Sign in with
Sign up | Sign in
Your question
Closed

Is the AMD FX 8350 good for gaming - Page 3

Last response: in CPUs
Share
March 30, 2013 7:32:06 AM

Yes, the i5 does better in games such as skyrim under Windows but then the FX does better in other games such as Showdown or Batman Arkham City under Windows (the FX perform at the same level that the i7-3770k in this latter game at 1080p). The FX is also at the i7 level on Battlefield 3 under Windows



Toms last advice (gaming cpu hierarchy chart update of 20 March 2013) is that, overall, you do "not notice a worthwhile difference in game performance" in Windows gaming. If you own a FX-8350, Toms recommend you to not upgrade to a more expensive chip. Moreover, neither Windows nor those games are using all the performance of the FX chips.

Moreover, at the time of writing this the FX is clearly faster (even beating the i7) under linux {*} and future games will be optimized for the AMD because just-ready consoles will be using AMD chips.

{*} Before anyone claim that linux is not for gaming, notice that developers such as Valve are abandoning Windows platform and porting games to linux.
March 30, 2013 7:43:41 AM

juanrga said:
Moreover, at the time of writing this the FX is clearly faster (even beating the i7) under linux {*} and future games will be optimized for the AMD because just-ready next consoles will be using AMD chips.


Yep, I think this is important. The PS4 is running on piledriver architecture and radeon graphics, AMD could easily pull ahead in gaming in the future because games well be optimized for their hardware.
March 30, 2013 9:22:25 AM

8350rocks said:
drinvis said:
Point here is gaming and i5 3570k till this date has more wins over fx8350 taking the CPU as same.
There is no point of saying which processor can do what in heavily multi-threaded renderings or encoding.
Just check out some Skyrim,Civilization 5,Dawn of war,etc i5 3570k has a good amount of lead.
Crysis 3 does well on fx8350 on the green scenes but rest does well on i5 3570k.In Toms test they were comparable.
For gaming i5 3570k this date is still better than fx8350.


We are discussing the reasons that the benchmarks reflect what they do...not what the current benchmarks say...

Also, take a look at openbenchmarking.org...they run non intel compiler programs and games on both systems and show you what you're REALLY getting into...


The thing is for many games the GPU is more important at higher resolutions with any capable CPU doing the job well enough.
Once you go to some games like Civilization5,SC2,Skyrim which demand more light threaded CPU power the fx8350 falls down which is never the case with
likes of i5 3570k.The point is for current games i5 3570k wins more and in some games the difference is more than noticeable.
Even in hitman absolution which is not that old game fx8350 gets considerably behind an i5 3570k http://www.techspot.com/review/608-hitman-absolution-pe...
Try to get the point,don't get blind in your support for a brand.

juanrga
Would you like to show the benches,then I can comment upon those.

I am not saying that fx8350 is a bad chip.It is a very good chip for the price considering it's all round abilities,full instruction set support unlike intel k series processors.But for gaming and things where per core performance is still very important i5 3570k still is a better choice.Well if one can get a better GPU with the saved money from the fx8350 set up then that is well and nice for gaming but otherwise i5 3570k is the way to go.

People please don't directly reach to the conclusion about gaming by directly comparing PS4 CPU specs with the PC CPU specs.The CPU in PS4 uses Jaguar cores not piledriver.In a dedicated platform like a PS4 the OS is much different,the layers of abstractions are much less.PS4 has unified memory for CPU and GPU,does PC have the same?I think it would be better if we talk about what we have seen or what we have now rather than talking about what will happen,just becoz a PS4 has 8 cores doesn't mean it is what you would take and directly compare with the PC CPUs with PC OSes and their ability to handle games.Gaming might become more threaded in future but per core performance of a processor is still quite relevant,which is why an i5 3570k in some games beats fx8350 by a good amount at this date.




Related resources
March 30, 2013 11:05:26 AM

I don't think AMD fans understand that the 8350 is already clocked to 4.0Ghz. The 3570K is clocked at 3.4Ghz. If the 3570K was overclocked to 4.0Ghz, it would beat the 8350 is most things.
a c 210 à CPUs
March 30, 2013 11:15:52 AM

robjon70 said:
I don't think AMD fans understand that the 8350 is already clocked to 4.0Ghz. The 3570K is clocked at 3.4Ghz. If the 3570K was overclocked to 4.0Ghz, it would beat the 8350 is most things.


You missed my post explaining what is going on with that...please refer to the bottom end of the page before for my example.

You are uninformed about how the CPUs work, I gave an example analogy in a long post about shopping lines at stores that explains the difference, it was a reply to ericjohn004.
March 30, 2013 1:08:17 PM

drinvis said:

juanrga
Would you like to show the benches,then I can comment upon those.


They were given before in the thread in several parts.

drinvis said:

I am not saying that fx8350 is a bad chip.It is a very good chip for the price considering it's all round abilities,full instruction set support unlike intel k series processors.But for gaming and things where per core performance is still very important i5 3570k still is a better choice.Well if one can get a better GPU with the saved money from the fx8350 set up then that is well and nice for gaming but otherwise i5 3570k is the way to go.


This was all addressed before.

drinvis said:

People please don't directly reach to the conclusion about gaming by directly comparing PS4 CPU specs with the PC CPU specs.The CPU in PS4 uses Jaguar cores not piledriver.In a dedicated platform like a PS4 the OS is much different,the layers of abstractions are much less.PS4 has unified memory for CPU and GPU,does PC have the same?I think it would be better if we talk about what we have seen or what we have now rather than talking about what will happen,just becoz a PS4 has 8 cores doesn't mean it is what you would take and directly compare with the PC CPUs with PC OSes and their ability to handle games.Gaming might become more threaded in future but per core performance of a processor is still quite relevant,which is why an i5 3570k in some games beats fx8350 by a good amount at this date.


What I have said is irrelevant to the core architecture. Moreover, I have not said you would wait the same amount of optimization than in the PS4 only you would wait optimization. It is not arbitrary that the PS4 will be an eight-core design. And if you learn how to parallelize your game for an eight-core design that game will run faster in an eight-core chip.
March 30, 2013 1:48:58 PM

juanrga
The PS4 thing that I posted was not for you,it was in general.

There are no links given.
Phoronix did a review about fx8350 under ubuntu12.10.
And most of the benches like NAS parallel benches,John the ripper etc are massively parallel,even then fx8350 with its 8 core gets beaten by i7 3770k in most cases,even if it wins in some the margin is not that high.
Games are different from things like NAS parallel benches,kernel compilation.
And it is not as if under linux fx8350 gets boosted by 40% than under windows7/8.
On overclocking an i5 3570k would pay off more in most of the things since it is more sensitive to frequency.
There is a reason why fx8350 is priced that low and that reason is not AMD are generous or are a no profit company,rather they don't have a good per core and light threaded performance,though it is still quite a good chip for the money.

The context here is gaming,and for gaming i5 3570k is a better chip than fx8350 as on this date,period.
a b à CPUs
March 30, 2013 1:49:04 PM

juanrga said:
Yes, the i5 does better in games such as skyrim under Windows but then the FX does better in other games such as Showdown or Batman Arkham City under Windows (the FX perform at the same level that the i7-3770k in this latter game at 1080p). The FX is also at the i7 level on Battlefield 3 under Windows



Toms last advice (gaming cpu hierarchy chart update of 20 March 2013) is that, overall, you do "not notice a worthwhile difference in game performance" in Windows gaming. If you own a FX-8350, Toms recommend you to not upgrade to a more expensive chip. Moreover, neither Windows nor those games are using all the performance of the FX chips.

Moreover, at the time of writing this the FX is clearly faster (even beating the i7) under linux {*} and future games will be optimized for the AMD because just-ready consoles will be using AMD chips.

{*} Before anyone claim that linux is not for gaming, notice that developers such as Valve are abandoning Windows platform and porting games to linux.


I can see someone comparing an 8350 to an i5 3570k in rendering and editing because if a 3570k and an 8350 are BOTH clocked at 4.5Ghz they both score a 7.5 in Cinebench 11.5. Granted I'm sure if an 8350 and 3570k were both at 4.5Ghz, I'm sure the 8350 would do better on MOST tests with anything that uses 5+ cores. Although the 3570k does win some tests that take advantage of 8 cores too. So I'll admit that. I compared some of my 3570k's benchmarks to a guy on Tom's that has an 8350. He beat me by 200 points in Passmark8, I score a 9700 and he scores a 9900. I beat him at Cinebench because I'm clocked .1Ghz higher than he is, other than that we would be the same, both scoring 7.50. And these scores are VERY similar to other websites I've viewed. And of course I would absolutely crush him in single threaded benchmarks such as SuperPi. Afteall an 8350 clocked at 7.8Ghz only scores an 11 second SuperPi score(I just looked this up). My 3570k at 4.6Ghz scores a 7.92 second SuperPi score. The 8350 and 3570k are just not comparable when it comes to single threaded benchmarks and programs. And fortunately for me, a lot of games and programs use less than 4 cores. Which is why I prefer the 3570k.

But to say that an 8350 compares to a 3770k OVERALL is just wrong. Compare the 3770k at 4.5Ghz vs. the 8350 at 4.5Ghz and the 3770k wins everything. Why compare them stock, when the 8350 is clocked at 4.0Ghz and the 3770k is clocked at 3.5? And even stock the 3770k wins alomst everything. Everyone knows they could make a 3770k 4.0Ghz stock too. So why not let it be equal? You may be able to find a couple oddball benchmarks that the 8350 win's in but if you use Sisoft Sandra, Passmark8, Aida 64, Cinebench 11.5, Handbrake, Fitz Chess Benchmark, 3DMark11 Physics, PCMark7, SuperPi, and ect. the 3770k wins in all of those benchmarks by a decent margin. It doesn't crush the 8350 in multithreading but it does beat it by a good 15-20%. And in single threaded benchmarks it's not even close. For example, the 3770k scores a 9 second SuperPi 1m score stock whereas an FX8350 overclocked to 7.8Ghz only scores an 11 second 1m score. So if the 8350 was stock, I'm guessing it would score somewhere around an 18-20 seconds 1m. Which is twice as slow. This benchmark goes to show you how much more powerful an i5 or i7 is while using 4 cores or less. And as I've said before, ALOT of programs and 99.999% of games don't use 5+ cores. Maybe in the future they will, but we're talking about right now.

Some FX fans completely forget about single threaded performance. Which is a BIG part of the overall picture. Not everything you do uses 5+ cores. Hell, EVERYTHING I do uses 4 cores of less during my everyday at my PC. This is why I love me 3570k, it's the best processor I could have for what I'm doing. If I would get an 8350 it would be a complete waste of 8 cores, plus I would get worse performance with everything I did, everyday. Sure I might not notice this one bit, but I like having high benchmark scores and impressive single threaded capability.

The thing is though, the 8350 only costs 199.99 on average. So it's not SUPPOSE to compete with the 3770k at 329.99. It's not even SUPPOSE to compete with the 3570k. As the 3570k is designed for people like me. People who do everyday computing and want that everyday computing and gaming to be the absolute best it could be. The 8350 is designed for multitaskers, and people who want high quality fast rendering on the cheap. Basically, an 8350 is for someone who wants a more cost effective version of a 3770k, but with a little less rendering muscle and very capable single threaded performance. So if you constantly doing this kind of stuff and want a great deal, I can see people buying the 8350. And the 8350 is a very good deal for the price and it overclocks like hell. I would buy one myself if it fit the profile of what I normally did with my PC.

What could possibly be wrong with what I've said? I'm not trying to be mean to FX owners, I just tell it like it is. But what I've said is true and I think everyone should agree. And if you only use multithreaded benchmarks to compare the 8350, 3770k, and 3570k then your missing a very large part of the overall picture because programs and games MOST people use, don't use 5+ cores.
a c 210 à CPUs
March 30, 2013 2:55:54 PM

You know, that's a fair assessment...however...I don't think an i3 at any price point makes sense over the FX6300...

I also think that people multitask more than they realize, and games coming soon are going to require more cores in the near future.

In less than 12 months anyone without a quad-core CPU (or more) is going to be handicapped.

That said, the architecture is going to start to be better utilized as software is more optimized for AMD protocols.

So, for a budget gaming build, I think an FX series CPU not only presents a more than valid way to go...it also leaves them far more choices down the road, and they get ALOT of performance now versus what they would pay for comparable intel performance.

Yes, the single threaded apps favor intel...of course...however, most budget gamers are not going to miss the 2-5 FPS difference.

If they can afford it now...and ONLY game, sure go i5/i7...especially if they can afford to upgrade in 12 months.
March 30, 2013 7:44:42 PM

drinvis said:
juanrga
There are no links given.


There was direct links and linked images of benchmark results.

drinvis said:

Phoronix did a review about fx8350 under ubuntu12.10.
And most of the benches like NAS parallel benches,John the ripper etc are massively parallel,


Precisely the reason for a multi-core design is exploiting parallel work. Nobody develops four or eight cores chips for running only one of them at once.

drinvis said:

even then fx8350 with its 8 core gets beaten by i7 3770k in most cases,even if it wins in some the margin is not that high.


The data says otherwise "FX 8350 competitive with Core i7 3770K". Moreover it seems you also missed the conclusion where explains that the current GGC is not using the advantages of the piledriver architecture over the Bulldozer one and how this explains the tests where the FX was not so good.

drinvis said:

On overclocking an i5 3570k would pay off more in most of the things since it is more sensitive to frequency.


Several benchmarks with the FX-8350 OC beating an i5 3570k OC were given.

drinvis said:

There is a reason why fx8350 is priced that low and that reason is not AMD are generous or are a no profit company,rather they don't have a good per core and light threaded performance,though it is still quite a good chip for the money.


It is not that the FX is priced low. It is Intel chips which are overpriced.

drinvis said:

The context here is gaming,and for gaming i5 3570k is a better chip than fx8350 as on this date,period.


Here you ignore all, including the Toms cpu gaming hierarchy chart.
March 30, 2013 8:16:01 PM

ericjohn004 said:

I can see someone comparing an 8350 to an i5 3570k in rendering and editing because if a 3570k and an 8350 are BOTH clocked at 4.5Ghz they both score a 7.5 in Cinebench 11.5. Granted I'm sure if an 8350 and 3570k were both at 4.5Ghz, I'm sure the 8350 would do better on MOST tests with anything that uses 5+ cores. Although the 3570k does win some tests that take advantage of 8 cores too. So I'll admit that. I compared some of my 3570k's benchmarks to a guy on Tom's that has an 8350. He beat me by 200 points in Passmark8, I score a 9700 and he scores a 9900. I beat him at Cinebench because I'm clocked .1Ghz higher than he is, other than that we would be the same, both scoring 7.50. And these scores are VERY similar to other websites I've viewed. And of course I would absolutely crush him in single threaded benchmarks such as SuperPi. Afteall an 8350 clocked at 7.8Ghz only scores an 11 second SuperPi score(I just looked this up). My 3570k at 4.6Ghz scores a 7.92 second SuperPi score. The 8350 and 3570k are just not comparable when it comes to single threaded benchmarks and programs. And fortunately for me, a lot of games and programs use less than 4 cores. Which is why I prefer the 3570k.

But to say that an 8350 compares to a 3770k OVERALL is just wrong. Compare the 3770k at 4.5Ghz vs. the 8350 at 4.5Ghz and the 3770k wins everything. Why compare them stock, when the 8350 is clocked at 4.0Ghz and the 3770k is clocked at 3.5? And even stock the 3770k wins alomst everything. Everyone knows they could make a 3770k 4.0Ghz stock too. So why not let it be equal? You may be able to find a couple oddball benchmarks that the 8350 win's in but if you use Sisoft Sandra, Passmark8, Aida 64, Cinebench 11.5, Handbrake, Fitz Chess Benchmark, 3DMark11 Physics, PCMark7, SuperPi, and ect. the 3770k wins in all of those benchmarks by a decent margin. It doesn't crush the 8350 in multithreading but it does beat it by a good 15-20%. And in single threaded benchmarks it's not even close. For example, the 3770k scores a 9 second SuperPi 1m score stock whereas an FX8350 overclocked to 7.8Ghz only scores an 11 second 1m score. So if the 8350 was stock, I'm guessing it would score somewhere around an 18-20 seconds 1m. Which is twice as slow. This benchmark goes to show you how much more powerful an i5 or i7 is while using 4 cores or less. And as I've said before, ALOT of programs and 99.999% of games don't use 5+ cores. Maybe in the future they will, but we're talking about right now.

Some FX fans completely forget about single threaded performance. Which is a BIG part of the overall picture. Not everything you do uses 5+ cores. Hell, EVERYTHING I do uses 4 cores of less during my everyday at my PC. This is why I love me 3570k, it's the best processor I could have for what I'm doing. If I would get an 8350 it would be a complete waste of 8 cores, plus I would get worse performance with everything I did, everyday. Sure I might not notice this one bit, but I like having high benchmark scores and impressive single threaded capability.

The thing is though, the 8350 only costs 199.99 on average. So it's not SUPPOSE to compete with the 3770k at 329.99. It's not even SUPPOSE to compete with the 3570k. As the 3570k is designed for people like me. People who do everyday computing and want that everyday computing and gaming to be the absolute best it could be. The 8350 is designed for multitaskers, and people who want high quality fast rendering on the cheap. Basically, an 8350 is for someone who wants a more cost effective version of a 3770k, but with a little less rendering muscle and very capable single threaded performance. So if you constantly doing this kind of stuff and want a great deal, I can see people buying the 8350. And the 8350 is a very good deal for the price and it overclocks like hell. I would buy one myself if it fit the profile of what I normally did with my PC.

What could possibly be wrong with what I've said? I'm not trying to be mean to FX owners, I just tell it like it is. But what I've said is true and I think everyone should agree. And if you only use multithreaded benchmarks to compare the 8350, 3770k, and 3570k then your missing a very large part of the overall picture because programs and games MOST people use, don't use 5+ cores.


I gave you the link to Passmark score with the FX-8350 on pair with an i7-3770K and beating your i5-3570K by a very wide margin.

I gave you best worldwide Cinebench value obtained by the FX-8350 and how it beats the best value obtained ever on an i5-3570K.

I gave you the link to the Wikipedia article explaining how the use of SuperPi for measuring chip performance is declining and why. You insist on evaluating the eight-cores FX-8350 with a test that only uses one of them, because this is how the i5-3570K can win thanks to a better performance using a single core.

You ignore the many benchmarks provided showing how "the AMD FX-8350 proved to be competitive with the Intel Core i7 3770K "Ivy Bridge" processor".

Your claim one cannot compare AMD and Intel processors unless the Intel chip is overclocked is so ridiculous that I don't need to answer this one.

You continue ignoring my emphasis on that current Windows games are not using all the performance of the AMD FX chip because were designed for an older architecture (as that in Intel i5-3570K), but that future games will be thanks to next consoles such as the PS4 which includes an 8-core chip from AMD.

You also ignore toms cpu gaming hierarchy, updated to this month, where they claim that if you were to upgrade from an FX-8350 to an expensive i7-3770K you would not notice overall gain in gaming performance.

You continue ignoring that lots of Windows programs are using the Intel compiler cheating, which forces the code to run slower when detects a non-Intel chip (did you read the links?). That is, your i5 is not so good as you believe it is, but the compiler is cheating about which is the real performance of the non-Intel chips.
March 31, 2013 9:29:22 AM

its very good for gaming but there are some great i5 that may perform better such as 3570 and 3570k!
a b à CPUs
March 31, 2013 10:21:50 AM

juanrga said:
ericjohn004 said:

I can see someone comparing an 8350 to an i5 3570k in rendering and editing because if a 3570k and an 8350 are BOTH clocked at 4.5Ghz they both score a 7.5 in Cinebench 11.5. Granted I'm sure if an 8350 and 3570k were both at 4.5Ghz, I'm sure the 8350 would do better on MOST tests with anything that uses 5+ cores. Although the 3570k does win some tests that take advantage of 8 cores too. So I'll admit that. I compared some of my 3570k's benchmarks to a guy on Tom's that has an 8350. He beat me by 200 points in Passmark8, I score a 9700 and he scores a 9900. I beat him at Cinebench because I'm clocked .1Ghz higher than he is, other than that we would be the same, both scoring 7.50. And these scores are VERY similar to other websites I've viewed. And of course I would absolutely crush him in single threaded benchmarks such as SuperPi. Afteall an 8350 clocked at 7.8Ghz only scores an 11 second SuperPi score(I just looked this up). My 3570k at 4.6Ghz scores a 7.92 second SuperPi score. The 8350 and 3570k are just not comparable when it comes to single threaded benchmarks and programs. And fortunately for me, a lot of games and programs use less than 4 cores. Which is why I prefer the 3570k.

But to say that an 8350 compares to a 3770k OVERALL is just wrong. Compare the 3770k at 4.5Ghz vs. the 8350 at 4.5Ghz and the 3770k wins everything. Why compare them stock, when the 8350 is clocked at 4.0Ghz and the 3770k is clocked at 3.5? And even stock the 3770k wins alomst everything. Everyone knows they could make a 3770k 4.0Ghz stock too. So why not let it be equal? You may be able to find a couple oddball benchmarks that the 8350 win's in but if you use Sisoft Sandra, Passmark8, Aida 64, Cinebench 11.5, Handbrake, Fitz Chess Benchmark, 3DMark11 Physics, PCMark7, SuperPi, and ect. the 3770k wins in all of those benchmarks by a decent margin. It doesn't crush the 8350 in multithreading but it does beat it by a good 15-20%. And in single threaded benchmarks it's not even close. For example, the 3770k scores a 9 second SuperPi 1m score stock whereas an FX8350 overclocked to 7.8Ghz only scores an 11 second 1m score. So if the 8350 was stock, I'm guessing it would score somewhere around an 18-20 seconds 1m. Which is twice as slow. This benchmark goes to show you how much more powerful an i5 or i7 is while using 4 cores or less. And as I've said before, ALOT of programs and 99.999% of games don't use 5+ cores. Maybe in the future they will, but we're talking about right now.

Some FX fans completely forget about single threaded performance. Which is a BIG part of the overall picture. Not everything you do uses 5+ cores. Hell, EVERYTHING I do uses 4 cores of less during my everyday at my PC. This is why I love me 3570k, it's the best processor I could have for what I'm doing. If I would get an 8350 it would be a complete waste of 8 cores, plus I would get worse performance with everything I did, everyday. Sure I might not notice this one bit, but I like having high benchmark scores and impressive single threaded capability.

The thing is though, the 8350 only costs 199.99 on average. So it's not SUPPOSE to compete with the 3770k at 329.99. It's not even SUPPOSE to compete with the 3570k. As the 3570k is designed for people like me. People who do everyday computing and want that everyday computing and gaming to be the absolute best it could be. The 8350 is designed for multitaskers, and people who want high quality fast rendering on the cheap. Basically, an 8350 is for someone who wants a more cost effective version of a 3770k, but with a little less rendering muscle and very capable single threaded performance. So if you constantly doing this kind of stuff and want a great deal, I can see people buying the 8350. And the 8350 is a very good deal for the price and it overclocks like hell. I would buy one myself if it fit the profile of what I normally did with my PC.

What could possibly be wrong with what I've said? I'm not trying to be mean to FX owners, I just tell it like it is. But what I've said is true and I think everyone should agree. And if you only use multithreaded benchmarks to compare the 8350, 3770k, and 3570k then your missing a very large part of the overall picture because programs and games MOST people use, don't use 5+ cores.


I gave you the link to Passmark score with the FX-8350 on pair with an i7-3770K and beating your i5-3570K by a very wide margin.

I gave you best worldwide Cinebench value obtained by the FX-8350 and how it beats the best value obtained ever on an i5-3570K.

I gave you the link to the Wikipedia article explaining how the use of SuperPi for measuring chip performance is declining and why. You insist on evaluating the eight-cores FX-8350 with a test that only uses one of them, because this is how the i5-3570K can win thanks to a better performance using a single core.

You ignore the many benchmarks provided showing how "the AMD FX-8350 proved to be competitive with the Intel Core i7 3770K "Ivy Bridge" processor".

Your claim one cannot compare AMD and Intel processors unless the Intel chip is overclocked is so ridiculous that I don't need to answer this one.

You continue ignoring my emphasis on that current Windows games are not using all the performance of the AMD FX chip because were designed for an older architecture (as that in Intel i5-3570K), but that future games will be thanks to next consoles such as the PS4 which includes an 8-core chip from AMD.

You also ignore toms cpu gaming hierarchy, updated to this month, where they claim that if you were to upgrade from an FX-8350 to an expensive i7-3770K you would not notice overall gain in gaming performance.

You continue ignoring that lots of Windows programs are using the Intel compiler cheating, which forces the code to run slower when detects a non-Intel chip (did you read the links?). That is, your i5 is not so good as you believe it is, but the compiler is cheating about which is the real performance of the non-Intel chips.


Yeah you STILL have no answer for single core performance, AND performance using 4 cores or less. And that's a HUGE, a HUGE part of the picture. Why do you neglect to go anywhere around single core and programs that use 4 cores or less? Because then you know your argument would go out the window. And you would have to start backtracking on some of your statements.

Yeah, everythings all great when you only compare threaded performance. Yeah then the FX8350 lives up to what it's suppose to be. But since 99.999% of all games AND at least half of all programs use 4 cores or less(probably more), I'd say single threading is far more important than what the processor can do while using 8 cores. And once you put the 3770k into the equation the 3770k wins in multithreading and single threaded performance so I don't see where your coming from.

I haven't even looked at one of your links. Because for some reason I'm sure your passmark8 links are going to have a different score than Tom's Hardware does. And Tom's Hardware clearly shows that a 3770k beats an 8350 in Passmark8. And if you don't trust them IDK what to say. And a 3570k an and 8350 are close. Yeah the 8350 beats the 3570k(I think), as I've already said, but that's just because passmark8, like Sisoft Sandra relies heavily on multithreading and hardly any in single threading and 4 cores or less.

And I think it's just hilarious how you have to go to websites that are specifically designed to show how the 8350 is awesome. You gotta go to some biased site where their motto is to prove the 8350 is better because of the "OS", "Intel's Compiler Programs", "SuperPi isn't a good benchmark", "Sysmark isn't a good benchmark". Why when Intel loses a benchmark they say "well we just lost that one". Why don't they make 1000 different excuses?

Because that's what your all about, EXCUSES.
a b à CPUs
March 31, 2013 10:22:02 AM

Intel's just so happened to be designed to where if you use 1-4 cores, you get 1-4 cores at 100% power. And if you use Hyperthreading, you get 8 cores at about 65% power. AMD's are designed in a way that is flawed. If you use 4 or less cores, you still only get cores at 65% since they ALL SHARE RESOURCES. A 3770k's cores only share if they are using Hyperthreading. THAT's why an 8350 is not nearly as good as Intel with single core and 4 cores or less performance. Because all of an 8350 cores are always running at 65% because they constantly have to share resources while a 3770k can run at 100% when it's not Hyperthreading.
a c 210 à CPUs
March 31, 2013 10:36:48 AM

ericjohn004 said:
Intel's just so happened to be designed to where if you use 1-4 cores, you get 1-4 cores at 100% power. And if you use Hyperthreading, you get 8 cores at about 65% power. AMD's are designed in a way that is flawed. If you use 4 or less cores, you still only get cores at 65% since they ALL SHARE RESOURCES. A 3770k's cores only share if they are using Hyperthreading. THAT's why an 8350 is not nearly as good as Intel with single core and 4 cores or less performance. Because all of an 8350 cores are always running at 65% because they constantly have to share resources while a 3770k can run at 100% when it's not Hyperthreading.


Well, you're barking up the right tree, but you went up the wrong branch...

Intel virtual cores (hyperthreading) is like each core in an intel being 1.2 cores in the chips that have it. So, in a 4 core processor with HT...you're kind of getting almost a 5 core CPU, in the grander scheme of things this is a decent approximation.

AMD cores on the other hand, share a floating point unit between 2 cores. Floating point units only come into play in certain specific scenarios...primarily physics calculations and rendering, which is why GPUs are so important in gaming...modern GPUs are physics crunching machines...however, back to my point. Only under unique circumstances will there be an occurence where you have resources shared between 2 AMD cores. The only situation I can think of would seriously be something like Crysis 3 MAYBE! Even then it must not have happened much because the benchmarks are basically mas o menos same/same between the 8350 and the intel i5s/i7s it was compared to.

Having said that, AMD cores are slightly stronger than your virtual cores in intel, because, unlike intel virtual cores, they don't have to tap resources from another core to run integer calculations or ANYTHING BUT FP calculations. So, an AMD Module, is 2 cores and 1 FPU, but it amounts to 1 module being approximately worth about ~1.6-1.8 Intel cores for raw horsepower. Intel chips run virtual cores that tax the resources of a "real" core, because you're sapping clock time that could be spent elsewhere.

So the reason AMD really shines at parallel coded applications (because of HSA)...is because it can run 8 separate integer calculation functions at 1 time without sharing resources, where as the i5 can only run 4 before it starts sharing resources.

Also, with more Cache on board the AMD chip, that makes it easier to keep parallel instructions fed to multiple cores, even when they do share resources under SUPER heavy load.

I think you're reading up more into this stuff...I am glad to see you're doing some self education on the subject.

a b à CPUs
March 31, 2013 1:30:17 PM

8350rocks said:
ericjohn004 said:
Intel's just so happened to be designed to where if you use 1-4 cores, you get 1-4 cores at 100% power. And if you use Hyperthreading, you get 8 cores at about 65% power. AMD's are designed in a way that is flawed. If you use 4 or less cores, you still only get cores at 65% since they ALL SHARE RESOURCES. A 3770k's cores only share if they are using Hyperthreading. THAT's why an 8350 is not nearly as good as Intel with single core and 4 cores or less performance. Because all of an 8350 cores are always running at 65% because they constantly have to share resources while a 3770k can run at 100% when it's not Hyperthreading.


Well, you're barking up the right tree, but you went up the wrong branch...

Intel virtual cores (hyperthreading) is like each core in an intel being 1.2 cores in the chips that have it. So, in a 4 core processor with HT...you're kind of getting almost a 5 core CPU, in the grander scheme of things this is a decent approximation.

AMD cores on the other hand, share a floating point unit between 2 cores. Floating point units only come into play in certain specific scenarios...primarily physics calculations and rendering, which is why GPUs are so important in gaming...modern GPUs are physics crunching machines...however, back to my point. Only under unique circumstances will there be an occurence where you have resources shared between 2 AMD cores. The only situation I can think of would seriously be something like Crysis 3 MAYBE! Even then it must not have happened much because the benchmarks are basically mas o menos same/same between the 8350 and the intel i5s/i7s it was compared to.

Having said that, AMD cores are slightly stronger than your virtual cores in intel, because, unlike intel virtual cores, they don't have to tap resources from another core to run integer calculations or ANYTHING BUT FP calculations. So, an AMD Module, is 2 cores and 1 FPU, but it amounts to 1 module being approximately worth about ~1.6-1.8 Intel cores for raw horsepower. Intel chips run virtual cores that tax the resources of a "real" core, because you're sapping clock time that could be spent elsewhere.

So the reason AMD really shines at parallel coded applications (because of HSA)...is because it can run 8 separate integer calculation functions at 1 time without sharing resources, where as the i5 can only run 4 before it starts sharing resources.

Also, with more Cache on board the AMD chip, that makes it easier to keep parallel instructions fed to multiple cores, even when they do share resources under SUPER heavy load.

I think you're reading up more into this stuff...I am glad to see you're doing some self education on the subject.



Thanks man I'm trying lol... Still not quite there yet. I can see how AMD cores are all equally powered no matter how many of them you use, whereas with Intel you only get, from what I read, about a 30% increase in performance which is about 65% of a full core. So the way AMD does it is superier to the way Intel does the Hyperthreading. The problem is, is that sometimes 1.3x a regular Intel core is more powerful than 2 AMD cores. And this is, I think, why sometimes the 3770k beats the 8350 in some multithreading. But the way AMD does it is a better way, in theory.

The way I see AMD really starting to be superior is if they really up the IPC on their chips. Or the per core power. Once they can compete, per core, then I can see an AMD 8 core chip beating a 4 core Intel chip plus Hyperthreading in ALL 5+ core tasks. But it almost seems like, from what I read, that AMD doesn't care about performance using 4 cores or less. It seems like they are really only trying in the multithreading game. Which is why they are so good at multithreaded applications. I mean if your really into things that use 5+ cores, an AMD is the better option from a price standpoint even though it, and the 3770k are similar in performance. The benchmarks I see have the 3770k winning by a handful of points the majority of the time but for the price you pay for the 8350 it makes the 8350 worth it. I can definately see why people buy the FX8350. Plus it is a little better overclocker than a 3770k I hear. Just by a couple Mhz though, but that does make a difference. But Intel kinda makes up fo that being that their chips start out at only 3.4 and 3.5Ghz.

If AMD can get their single threaded performance to about 75-85% of what an Intel core does, I think the 8350 would be a killer CPU. Then I would definately buy one of these for a future PC. But I think AMD's plans are for the future. When in 3-4 years, everything uses 8+ cores. But in the meantime, per core performance is a rather large part of the picture. And per core performance will always be important. When are AMD's next generation of CPU's coming out? The 4770k is coming out in June I think, although according to Tom's it's only anywhere from 6-13% faster than a 3770k. So for me, it's not worth the upgrade. Although by then I won't be able to claim I have the most powerful CPU, per core, on the market lol. But I think AMD's new CPU's normally come out long after the new Intel CPU's come out.

March 31, 2013 1:56:15 PM

juanrga said:
drinvis said:
juanrga
There are no links given.


There was direct links and linked images of benchmark results.

drinvis said:

Phoronix did a review about fx8350 under ubuntu12.10.
And most of the benches like NAS parallel benches,John the ripper etc are massively parallel,


Precisely the reason for a multi-core design is exploiting parallel work. Nobody develops four or eight cores chips for running only one of them at once.

drinvis said:

even then fx8350 with its 8 core gets beaten by i7 3770k in most cases,even if it wins in some the margin is not that high.


The data says otherwise "FX 8350 competitive with Core i7 3770K". Moreover it seems you also missed the conclusion where explains that the current GGC is not using the advantages of the piledriver architecture over the Bulldozer one and how this explains the tests where the FX was not so good.

drinvis said:

On overclocking an i5 3570k would pay off more in most of the things since it is more sensitive to frequency.


Several benchmarks with the FX-8350 OC beating an i5 3570k OC were given.

drinvis said:

There is a reason why fx8350 is priced that low and that reason is not AMD are generous or are a no profit company,rather they don't have a good per core and light threaded performance,though it is still quite a good chip for the money.


It is not that the FX is priced low. It is Intel chips which are overpriced.

drinvis said:

The context here is gaming,and for gaming i5 3570k is a better chip than fx8350 as on this date,period.


Here you ignore all, including the Toms cpu gaming hierarchy chart.


That review from phoronix clearly shows that i7 3770k is the better choice and it is not as portrayed by many as if the fx8350 is competitive with i7 3770k.Competitive in few benches doesn't mean it is overall competitive with i7 3770k.Check this :http://forums.anandtech.com/showpost.php?p=34472568&pos...
Even when HT is turned off i7 3770k still comes out ahead of fx8350 at same clocks,which says an i5 3570k would perform similar or better to fx8350 at same clocks in 3DS MAX 2013 rendering.An in 3DS MAX there are many things like previewing objects in viewport which depend upon single/light threaded performance,so i5 3570k is quite a good choice for 3DS MAX 2013 as it has good performance in both single threaded and heavily threaded tasks.
No wonder,one can say intel processors are overpriced and if fx8350 had all round performance with it then probably it would have been priced more than i5 3570k.Come on,AMD is not doing charity here,they need money as well.Even hd7970 was priced 550$,AMD afterwards got it priced to much lower levels as soon as they got competition.
Several benchmarks!!!...there are enough benchmarks where an OCed i5 3570k beats fx8350 quite well and vice-versa but for most of the things which utilise 4/5 or less cores i5 3570k wins hands down.Every program doesn't use 8 or more cores neither everything requires more cores.Never did I say that parallel things like NAS parallel benches should not use more cores,so what was your point in that context?
What I implied was you can't compare gaming performance by seeing those benches.

And about gaming I haven't said anything that is wrong.I have explained well enough that there are some situations where fx8350 is noticeably slower than i5 3570k.Check these




Fx8350 is a good processor for the price,quite fantastic for specific tasks like video encoding,compression etc.
Single core performance and light threaded performance of fx8350 is quite weak compared to i5 3570k.I have repeatedly said the context here is gaming and i5 3570k is the better one for it as of now.You clearly are missing the point here.So don't deviate more,rather put your logic on the context.
a c 210 à CPUs
March 31, 2013 4:56:15 PM

ericjohn004 said:

Thanks man I'm trying lol... Still not quite there yet. I can see how AMD cores are all equally powered no matter how many of them you use, whereas with Intel you only get, from what I read, about a 30% increase in performance which is about 65% of a full core. So the way AMD does it is superier to the way Intel does the Hyperthreading. The problem is, is that sometimes 1.3x a regular Intel core is more powerful than 2 AMD cores. And this is, I think, why sometimes the 3770k beats the 8350 in some multithreading. But the way AMD does it is a better way, in theory.


Well, I can get into why the i7-3770k still wins occasionally, I will address it further down though because it will be more pertinent.
ericjohn004 said:

The way I see AMD really starting to be superior is if they really up the IPC on their chips. Or the per core power. Once they can compete, per core, then I can see an AMD 8 core chip beating a 4 core Intel chip plus Hyperthreading in ALL 5+ core tasks. But it almost seems like, from what I read, that AMD doesn't care about performance using 4 cores or less. It seems like they are really only trying in the multithreading game. Which is why they are so good at multithreaded applications. I mean if your really into things that use 5+ cores, an AMD is the better option from a price standpoint even though it, and the 3770k are similar in performance. The benchmarks I see have the 3770k winning by a handful of points the majority of the time but for the price you pay for the 8350 it makes the 8350 worth it. I can definately see why people buy the FX8350. Plus it is a little better overclocker than a 3770k I hear. Just by a couple Mhz though, but that does make a difference. But Intel kinda makes up fo that being that their chips start out at only 3.4 and 3.5Ghz.


AMD's protocols (called HSA) are designed specifically to excel at multi threaded or parallel applications. You're right the IPC on the BD/PD architecture is still comparatively low to what intel is doing...this is why the i7-3770k occasionally still wins, what it lacks in comparatively raw horsepower, it makes up for in efficiency of clock time used. AMD is working on this...

ericjohn004 said:
If AMD can get their single threaded performance to about 75-85% of what an Intel core does, I think the 8350 would be a killer CPU. Then I would definately buy one of these for a future PC. But I think AMD's plans are for the future. When in 3-4 years, everything uses 8+ cores. But in the meantime, per core performance is a rather large part of the picture. And per core performance will always be important. When are AMD's next generation of CPU's coming out? The 4770k is coming out in June I think, although according to Tom's it's only anywhere from 6-13% faster than a 3770k. So for me, it's not worth the upgrade. Although by then I won't be able to claim I have the most powerful CPU, per core, on the market lol. But I think AMD's new CPU's normally come out long after the new Intel CPU's come out.



I agree, and so do many of the people with a similar level of knowledge...THIS, PRECISELY, is why the few that are following steamroller are excited about. Steamroller is going to be PD with a more refined protocol for handling instructions, which will make the architecture far more efficient using clock time. The IPC on Steamroller is supposed to be competitive to what ivy bridge is now...but with more raw horsepower under the hood. So, steamroller is going to be what everyone THOUGHT bulldozer was going to be, however, in order to keep something out there, AMD rushed their 1st generation chip out to a great deal of enthusiasm that was quickly glossed over by the fact that the chip didn't deliver on the claims...YET! But, the architecture is capable of competing, and with a new instruction protocol in place at the basic level that processes instructions more efficiently...it should effectively "steamroll" it's way through to being a very valid option for anyone that would save them quite a bit of money.

The expectations from AMD are a 30% gain in single threaded performance and 15-20% overall. A 30% gain puts the Steamroller architecture right in the thick of everything even with Haswell coming out, the advance in performance is only going to be a minor one...you don't hear about a 30% increase in CPU performance, it's unheard of...but the new instruction protocol is that drastic an upgrade that it's going to really put things in a new light.
a b à CPUs
March 31, 2013 4:57:29 PM

Juanrga's problem is he's only taking into account heavily threaded benchmarks. He says nothing about lightly threaded benchmarks, single threaded benchmarks, or gaming. The only thing he says about gaming is that the 8350 is almost as good as the 3570k. But in some games the 3570k dominates, like your benchmarks say(notice Tek Syndacate doesn't include Skyrim, they seem to be trying to prove a point though). In some games the 8350 actually wins by a small margin. The 8350 IS good for gaming. But not AS good as the 3570k for gaming OVERALL. Most games you really won't notice much of a difference, but then Skyrim comes along.

And I find, if you have to deviate from sites like Tom's Hardware and dig all through the internet to find certain websites that prove your point(and you can ALWAYS find some site that disagrees with reality), then your looking for something that's probably not there.

I'm not saying that there aren't some other really good benchmark sites, and I'm not saying those sites aren't correct in what they are reporting, I'm just saying that they aren't going to tell you the whole story. They may only come at it from one specific angle that's preferable to what they are trying to show you. And anything that might deviate from their specific agenda, well, they just won't include that in their analysis or they'll purposely avoid even running a certain benchmark just to further prove their point. That's why there are some sites that have an 8350 "winning" over a 3570k. But those sites are probably only using multithreaded benchmarks and won't include any single threaded benchmarks, just so one processor will "win" over the other.

So if you visit a site that uses a good general variety of popular benchmarks(like Anandtech or Tom's Hardware), and doesn't rely so heavily on multithreaded tests, you'll see that what I've said is indeed true. Let me point out though that it's not like I'm trying to prove the 8350 "loses" I'm just trying to get everyone to see the whole picture and not just a small piece of the whole puzzle.

This is why the 8350 is priced where it is, and the benchmarks show that. You are getting a great deal for the 8350 and you also usually get what you pay for.
a b à CPUs
March 31, 2013 5:07:37 PM

8350rocks said:
ericjohn004 said:

Thanks man I'm trying lol... Still not quite there yet. I can see how AMD cores are all equally powered no matter how many of them you use, whereas with Intel you only get, from what I read, about a 30% increase in performance which is about 65% of a full core. So the way AMD does it is superier to the way Intel does the Hyperthreading. The problem is, is that sometimes 1.3x a regular Intel core is more powerful than 2 AMD cores. And this is, I think, why sometimes the 3770k beats the 8350 in some multithreading. But the way AMD does it is a better way, in theory.


Well, I can get into why the i7-3770k still wins occasionally, I will address it further down though because it will be more pertinent.
ericjohn004 said:

The way I see AMD really starting to be superior is if they really up the IPC on their chips. Or the per core power. Once they can compete, per core, then I can see an AMD 8 core chip beating a 4 core Intel chip plus Hyperthreading in ALL 5+ core tasks. But it almost seems like, from what I read, that AMD doesn't care about performance using 4 cores or less. It seems like they are really only trying in the multithreading game. Which is why they are so good at multithreaded applications. I mean if your really into things that use 5+ cores, an AMD is the better option from a price standpoint even though it, and the 3770k are similar in performance. The benchmarks I see have the 3770k winning by a handful of points the majority of the time but for the price you pay for the 8350 it makes the 8350 worth it. I can definately see why people buy the FX8350. Plus it is a little better overclocker than a 3770k I hear. Just by a couple Mhz though, but that does make a difference. But Intel kinda makes up fo that being that their chips start out at only 3.4 and 3.5Ghz.


AMD's protocols (called HSA) are designed specifically to excel at multi threaded or parallel applications. You're right the IPC on the BD/PD architecture is still comparatively low to what intel is doing...this is why the i7-3770k occasionally still wins, what it lacks in comparatively raw horsepower, it makes up for in efficiency of clock time used. AMD is working on this...

ericjohn004 said:
If AMD can get their single threaded performance to about 75-85% of what an Intel core does, I think the 8350 would be a killer CPU. Then I would definately buy one of these for a future PC. But I think AMD's plans are for the future. When in 3-4 years, everything uses 8+ cores. But in the meantime, per core performance is a rather large part of the picture. And per core performance will always be important. When are AMD's next generation of CPU's coming out? The 4770k is coming out in June I think, although according to Tom's it's only anywhere from 6-13% faster than a 3770k. So for me, it's not worth the upgrade. Although by then I won't be able to claim I have the most powerful CPU, per core, on the market lol. But I think AMD's new CPU's normally come out long after the new Intel CPU's come out.



I agree, and so do many of the people with a similar level of knowledge...THIS, PRECISELY, is why the few that are following steamroller are excited about. Steamroller is going to be PD with a more refined protocol for handling instructions, which will make the architecture far more efficient using clock time. The IPC on Steamroller is supposed to be competitive to what ivy bridge is now...but with more raw horsepower under the hood. So, steamroller is going to be what everyone THOUGHT bulldozer was going to be, however, in order to keep something out there, AMD rushed their 1st generation chip out to a great deal of enthusiasm that was quickly glossed over by the fact that the chip didn't deliver on the claims...YET! But, the architecture is capable of competing, and with a new instruction protocol in place at the basic level that processes instructions more efficiently...it should effectively "steamroll" it's way through to being a very valid option for anyone that would save them quite a bit of money.


Yeah, now that you say this, I'm a lot more excited about the new AMD chips than the 4670k or the 4770k. Although that might be because I already know what they can do. But if AMD's doing what you say they are doing then that's definitely something to be excited about.

I would expect their pricing to stay below Intels for comparable performance. But I won't expect it to stay as low as say, 179.99, like you can get an 8350 for. If it's comparing with a 4770k then I would expect the price to be 279.99 or 249.99. And even if it's just as powerful they will have to still be cheaper just because of people's perception. People perceive Intel to be the highest quality and they perceive AMD to be cheaper but still quality 2nd best. And some people don't read enough benchmarks to overcome that perception.

I wonder when these new steamroller designed will be introduced? 2013? Beginning of 2014?

a c 210 à CPUs
March 31, 2013 5:29:10 PM

ericjohn004 said:
8350rocks said:
ericjohn004 said:

Thanks man I'm trying lol... Still not quite there yet. I can see how AMD cores are all equally powered no matter how many of them you use, whereas with Intel you only get, from what I read, about a 30% increase in performance which is about 65% of a full core. So the way AMD does it is superier to the way Intel does the Hyperthreading. The problem is, is that sometimes 1.3x a regular Intel core is more powerful than 2 AMD cores. And this is, I think, why sometimes the 3770k beats the 8350 in some multithreading. But the way AMD does it is a better way, in theory.


Well, I can get into why the i7-3770k still wins occasionally, I will address it further down though because it will be more pertinent.
ericjohn004 said:

The way I see AMD really starting to be superior is if they really up the IPC on their chips. Or the per core power. Once they can compete, per core, then I can see an AMD 8 core chip beating a 4 core Intel chip plus Hyperthreading in ALL 5+ core tasks. But it almost seems like, from what I read, that AMD doesn't care about performance using 4 cores or less. It seems like they are really only trying in the multithreading game. Which is why they are so good at multithreaded applications. I mean if your really into things that use 5+ cores, an AMD is the better option from a price standpoint even though it, and the 3770k are similar in performance. The benchmarks I see have the 3770k winning by a handful of points the majority of the time but for the price you pay for the 8350 it makes the 8350 worth it. I can definately see why people buy the FX8350. Plus it is a little better overclocker than a 3770k I hear. Just by a couple Mhz though, but that does make a difference. But Intel kinda makes up fo that being that their chips start out at only 3.4 and 3.5Ghz.


AMD's protocols (called HSA) are designed specifically to excel at multi threaded or parallel applications. You're right the IPC on the BD/PD architecture is still comparatively low to what intel is doing...this is why the i7-3770k occasionally still wins, what it lacks in comparatively raw horsepower, it makes up for in efficiency of clock time used. AMD is working on this...

ericjohn004 said:
If AMD can get their single threaded performance to about 75-85% of what an Intel core does, I think the 8350 would be a killer CPU. Then I would definately buy one of these for a future PC. But I think AMD's plans are for the future. When in 3-4 years, everything uses 8+ cores. But in the meantime, per core performance is a rather large part of the picture. And per core performance will always be important. When are AMD's next generation of CPU's coming out? The 4770k is coming out in June I think, although according to Tom's it's only anywhere from 6-13% faster than a 3770k. So for me, it's not worth the upgrade. Although by then I won't be able to claim I have the most powerful CPU, per core, on the market lol. But I think AMD's new CPU's normally come out long after the new Intel CPU's come out.



I agree, and so do many of the people with a similar level of knowledge...THIS, PRECISELY, is why the few that are following steamroller are excited about. Steamroller is going to be PD with a more refined protocol for handling instructions, which will make the architecture far more efficient using clock time. The IPC on Steamroller is supposed to be competitive to what ivy bridge is now...but with more raw horsepower under the hood. So, steamroller is going to be what everyone THOUGHT bulldozer was going to be, however, in order to keep something out there, AMD rushed their 1st generation chip out to a great deal of enthusiasm that was quickly glossed over by the fact that the chip didn't deliver on the claims...YET! But, the architecture is capable of competing, and with a new instruction protocol in place at the basic level that processes instructions more efficiently...it should effectively "steamroll" it's way through to being a very valid option for anyone that would save them quite a bit of money.


Yeah, now that you say this, I'm a lot more excited about the new AMD chips than the 4670k or the 4770k. Although that might be because I already know what they can do. But if AMD's doing what you say they are doing then that's definitely something to be excited about.

I would expect their pricing to stay below Intels for comparable performance. But I won't expect it to stay as low as say, 179.99, like you can get an 8350 for. If it's comparing with a 4770k then I would expect the price to be 279.99 or 249.99. And even if it's just as powerful they will have to still be cheaper just because of people's perception. People perceive Intel to be the highest quality and they perceive AMD to be cheaper but still quality 2nd best. And some people don't read enough benchmarks to overcome that perception.

I wonder when these new steamroller designed will be introduced? 2013? Beginning of 2014?



I have read...somewhere...that Steamroller may be Q4 2013, but will more likely come Q1 2014. Kaveri, the next iteration of the APU after Richland will be steamroller architecture, so if it hits for the holiday season, I would expect the steamroller CPUs to be right there as well. I am sure they will likely launch simultaneously.

If the original Bulldozer pricing estimate holds true, then I would expect the top of the line chip in the $229-239 price range, that's where bulldozer was going to fall retail when they launched it, then they cut it back to get market share and because it was less competitive than they had hoped.
a b à CPUs
March 31, 2013 6:20:44 PM

All I see is proof that the FX 8350 is adequate for gaming and CPU tasks that can use more than 4 cores. Please people, stop being fanboys and morons. We're all adults here... or most of us. Grow up. People buy the CPU they want or like. You can't go wrong with either processor.
March 31, 2013 8:46:57 PM

ericjohn004 said:

Yeah you STILL have no answer for single core performance, AND performance using 4 cores or less. And that's a HUGE, a HUGE part of the picture. Why do you neglect to go anywhere around single core and programs that use 4 cores or less? Because then you know your argument would go out the window. And you would have to start backtracking on some of your statements.

Yeah, everythings all great when you only compare threaded performance. Yeah then the FX8350 lives up to what it's suppose to be. But since 99.999% of all games AND at least half of all programs use 4 cores or less(probably more), I'd say single threading is far more important than what the processor can do while using 8 cores. And once you put the 3770k into the equation the 3770k wins in multithreading and single threaded performance so I don't see where your coming from.

I haven't even looked at one of your links. Because for some reason I'm sure your passmark8 links are going to have a different score than Tom's Hardware does. And Tom's Hardware clearly shows that a 3770k beats an 8350 in Passmark8. And if you don't trust them IDK what to say. And a 3570k an and 8350 are close. Yeah the 8350 beats the 3570k(I think), as I've already said, but that's just because passmark8, like Sisoft Sandra relies heavily on multithreading and hardly any in single threading and 4 cores or less.

And I think it's just hilarious how you have to go to websites that are specifically designed to show how the 8350 is awesome. You gotta go to some biased site where their motto is to prove the 8350 is better because of the "OS", "Intel's Compiler Programs", "SuperPi isn't a good benchmark", "Sysmark isn't a good benchmark". Why when Intel loses a benchmark they say "well we just lost that one". Why don't they make 1000 different excuses?

Because that's what your all about, EXCUSES.


1) Maybe you would go through the thread and count the number of times that I said that i5-3570k/i7-3770k usually has better performance per core.

2) Yes, there is older software which is single threaded (I also said this), but there multi-threaded versions of several of them available now, and the future goes toward multi-threading hardware not towards hyperclocked single cores.

3) My main point is that the FX-8350 is an excellent machine for current software and will be better for future software per point 2).

4) I have given you Toms benchmarks showing how the FX-8350 is on pair with a i7-3770k on several games, albeit the games are not still using the full potential of the FX chip. I have cited and quoted Toms gaming cpu hierarchy chart. And I have cited recent openbenchmarks that show the real potential of the FX beating a i7-3770k with modern software.

5) I have given you a link to the Wikipedia explaining why Super Pi is a bad tool for measuring chip performance and why "its use is declining". Of course you can continue reporting everyone else your Super Pi score. But pretending to evaluate a FX-8350 by insisting on an older software that only uses a 12.5% of the FX chip is kind of laughable.

6) I gave links to news reporting AMD, Nvidia, and VIA abandoning BAPCO because of the "SYSMARK bias" toward Intel products.

7) I gave links to a court case won by AMD because of the Intel compiler cheating.

8) "I haven't even looked at one of your links". This says it all and shows that there is not reason to further arguing with a blinded owner.
March 31, 2013 9:53:04 PM

drinvis said:

That review from phoronix clearly shows that i7 3770k is the better choice and it is not as portrayed by many as if the fx8350 is competitive with i7 3770k.Competitive in few benches doesn't mean it is overall competitive with i7 3770k.Check this :http://forums.anandtech.com/showpost.php?p=34472568&pos...
Even when HT is turned off i7 3770k still comes out ahead of fx8350 at same clocks,which says an i5 3570k would perform similar or better to fx8350 at same clocks in 3DS MAX 2013 rendering.An in 3DS MAX there are many things like previewing objects in viewport which depend upon single/light threaded performance,so i5 3570k is quite a good choice for 3DS MAX 2013 as it has good performance in both single threaded and heavily threaded tasks.
No wonder,one can say intel processors are overpriced and if fx8350 had all round performance with it then probably it would have been priced more than i5 3570k.Come on,AMD is not doing charity here,they need money as well.Even hd7970 was priced 550$,AMD afterwards got it priced to much lower levels as soon as they got competition.
Several benchmarks!!!...there are enough benchmarks where an OCed i5 3570k beats fx8350 quite well and vice-versa but for most of the things which utilise 4/5 or less cores i5 3570k wins hands down.Every program doesn't use 8 or more cores neither everything requires more cores.Never did I say that parallel things like NAS parallel benches should not use more cores,so what was your point in that context?
What I implied was you can't compare gaming performance by seeing those benches.

And about gaming I haven't said anything that is wrong.I have explained well enough that there are some situations where fx8350 is noticeably slower than i5 3570k.Check these




Fx8350 is a good processor for the price,quite fantastic for specific tasks like video encoding,compression etc.
Single core performance and light threaded performance of fx8350 is quite weak compared to i5 3570k.I have repeatedly said the context here is gaming and i5 3570k is the better one for it as of now.You clearly are missing the point here.So don't deviate more,rather put your logic on the context.


1) That review from phoronix clearly shows that the cheap FX-8350 is on pair with an i7-3770k (beating the Intel ivy Bridge in several tests). The review also emphasizes that the FX is competitive with the i7, albeit the current version of the compiler is not still using all the advantages of the Piledriver architecture over the Bulldozer one. The FX will be still better when the software was using all the hidden performance from the chip...

2) You write "Competitive in few benches doesn't mean it is overall competitive" but then spout a link to a forum presenting one closed benchmark avoiding hardware details (really any detail is avoided). And the benchmark is biased to every place posible: presenting the data in such a way that you visually compare the FX stock with the Intel overclocked, measuring power from the wall, using overclocked memory on the Intel but stock memory on the FX, extrapolating the Intel data far beyond its maximum speed... and all for finally obtaining a real 13% gain which reads to a few seconds.

3) I notice how every fanboy insists on the existence of little software that uses 8-cores. Not only they miss that an entire generation of next games will be optimized for 8-cores thanks to consoles as the PS4 using a 8-cores chip from AMD. But that the fanboys also omit that people often runs two-three applications at once. It is not usual to see a FX-8350 user gaming whereas doing some background work. Again this is a typical situation where the FX shows its real performance.

4) Yes you can pick a pair of selective games and configurations where the FX-8350 does not look so good. Toms have something important to say about your choice of Skyrim as example (bold font from mine):

Quote:
Skyrim appears to be the most CPU-dependent game in today’s suite. It also appears to be the most heavily slanted toward Intel's architecture. AMD's FX-8350 appears adequate across all of the tested settings, though we do have a little more data to discuss.


And still they say "AMD's FX-8350 appears adequate across all of the tested settings".

You cannot take one of the most unfavourable case possibles and pretend that it is something that happens more often.

It is also interesting that you give WoW as example of how you believe the FX "is noticeably slower". Are you serious? A difference of 10FPS above >80FPS? I suppose that you mean that those 10FPS are noticeable only when your see the bars in the graph, because you cannot detect the difference between >80FPS and 90FPS when playing the game in a 60Hz monitor, you cannot even using a 80Hz monitor...

And of course, one again, you ignore what Toms says in the cpu gaming hierachy chart (updated to this month), where the FX is placed near the top and emphasized that you do not see overall gain performance when going to a more expensive i5/i7. See also point (3).
April 1, 2013 1:50:34 AM

SO much people going back and fourth intel better amd better.... POINT is practically ALL games are gpu bottlenecked my best friend has a overclocked 680 with his phenom 2 x 4 clocked at 3.5ghz and he gets 60 frames in crysis 3 and far cry 3 and bio shock all at max in one monitor at 1080p..... point is the fx-8350 is more than enough all you need to worry about is GPU for gaming. i have an i7-3770k with a hd 7970 and i play all those games exactly the same at 60 frames.... get w/e cpu you want focus on the video card its where the performance is at an i5 or the fx-8350 you wont tell the differance
a b à CPUs
April 1, 2013 4:08:35 AM

griptwister said:
Lol, I remember when the i7 was hot shizz and intel fanboys were saying, "Multi-threading is the future" but now that AMD made decent multi-threaded CPU, the bandwagoners say, "Games only use 2-4 cores." If that was true, most modern games wouldn't be doing well on a vishera CPU. Infact. the FX processor still provides Stable FPS. Intel is king, but for the overall price, AMD is cheaper... And provides an upgrade path to Steamroller. Some of us don't have a Mere Hundred Dollars. The FX 8350 is right on the i5 and i7's a$$ in most benchmarks. Except skyrim lol.


a b à CPUs
April 1, 2013 4:16:15 AM

All this arguing about which processor is superior or if AMD can keep up with Intel is a moot point. Current Sandy/Ivy bridge chips destroy all FX chips when both processor are at same clocks. Why same clocks? Because both have identical overclocking capabilities and therefore they should be compared as such. Most people don't buy an unlocked CPU to leave at stock

An i5 at the SAME CLOCK as an FX 8350 will absolutely smoke it in anything using 4 cores or less and most likely come slightly behind or even mildly ahead in some multi threaded aps.

An i7 at the SAME CLOCK as an FX 8350 will DEMOLISH the FX is virtually everything outside of a very few select programs. And it will do all of this while using HALF the power and a POS $25 heatsink from NewEgg. Meanwhile smoke bellows from the H100 attempting to quench the power hungry beast of an FX chip.

Lets quit acting like these chips are even in the same league here.

a c 210 à CPUs
April 1, 2013 7:56:05 AM

bigj1985 said:
All this arguing about which processor is superior or if AMD can keep up with Intel is a moot point. Current Sandy/Ivy bridge chips destroy all FX chips when both processor are at same clocks. Why same clocks? Because both have identical overclocking capabilities and therefore they should be compared as such. Most people don't buy an unlocked CPU to leave at stock

An i5 at the SAME CLOCK as an FX 8350 will absolutely smoke it in anything using 4 cores or less and most likely come slightly behind or even mildly ahead in some multi threaded aps.

An i7 at the SAME CLOCK as an FX 8350 will DEMOLISH the FX is virtually everything outside of a very few select programs. And it will do all of this while using HALF the power and a POS $25 heatsink from NewEgg. Meanwhile smoke bellows from the H100 attempting to quench the power hungry beast of an FX chip.

Lets quit acting like these chips are even in the same league here.



Ok, OC either one of those to 5.2-5.6 GHz....then talk about OC'ing potential being equal.

April 1, 2013 9:26:33 AM

bigj1985 said:
All this arguing about which processor is superior or if AMD can keep up with Intel is a moot point. Current Sandy/Ivy bridge chips destroy all FX chips when both processor are at same clocks. Why same clocks? Because both have identical overclocking capabilities and therefore they should be compared as such. Most people don't buy an unlocked CPU to leave at stock

An i5 at the SAME CLOCK as an FX 8350 will absolutely smoke it in anything using 4 cores or less and most likely come slightly behind or even mildly ahead in some multi threaded aps.

An i7 at the SAME CLOCK as an FX 8350 will DEMOLISH the FX is virtually everything outside of a very few select programs. And it will do all of this while using HALF the power and a POS $25 heatsink from NewEgg. Meanwhile smoke bellows from the H100 attempting to quench the power hungry beast of an FX chip.

Lets quit acting like these chips are even in the same league here.



1) Comparing a stock FX to an overcloked Intel chip is biased.

2) The FX has superior overclocking capabilitites. It is easier to overclock, more stable, and has the world record on overclocking.

3) Rest of your points were addressed before as well.
April 1, 2013 1:30:38 PM

juanrga said:
drinvis said:

That review from phoronix clearly shows that i7 3770k is the better choice and it is not as portrayed by many as if the fx8350 is competitive with i7 3770k.Competitive in few benches doesn't mean it is overall competitive with i7 3770k.Check this :http://forums.anandtech.com/showpost.php?p=34472568&pos...
Even when HT is turned off i7 3770k still comes out ahead of fx8350 at same clocks,which says an i5 3570k would perform similar or better to fx8350 at same clocks in 3DS MAX 2013 rendering.An in 3DS MAX there are many things like previewing objects in viewport which depend upon single/light threaded performance,so i5 3570k is quite a good choice for 3DS MAX 2013 as it has good performance in both single threaded and heavily threaded tasks.
No wonder,one can say intel processors are overpriced and if fx8350 had all round performance with it then probably it would have been priced more than i5 3570k.Come on,AMD is not doing charity here,they need money as well.Even hd7970 was priced 550$,AMD afterwards got it priced to much lower levels as soon as they got competition.
Several benchmarks!!!...there are enough benchmarks where an OCed i5 3570k beats fx8350 quite well and vice-versa but for most of the things which utilise 4/5 or less cores i5 3570k wins hands down.Every program doesn't use 8 or more cores neither everything requires more cores.Never did I say that parallel things like NAS parallel benches should not use more cores,so what was your point in that context?
What I implied was you can't compare gaming performance by seeing those benches.

And about gaming I haven't said anything that is wrong.I have explained well enough that there are some situations where fx8350 is noticeably slower than i5 3570k.Check these




Fx8350 is a good processor for the price,quite fantastic for specific tasks like video encoding,compression etc.
Single core performance and light threaded performance of fx8350 is quite weak compared to i5 3570k.I have repeatedly said the context here is gaming and i5 3570k is the better one for it as of now.You clearly are missing the point here.So don't deviate more,rather put your logic on the context.


1) That review from phoronix clearly shows that the cheap FX-8350 is on pair with an i7-3770k (beating the Intel ivy Bridge in several tests). The review also emphasizes that the FX is competitive with the i7, albeit the current version of the compiler is not still using all the advantages of the Piledriver architecture over the Bulldozer one. The FX will be still better when the software was using all the hidden performance from the chip...

2) You write "Competitive in few benches doesn't mean it is overall competitive" but then spout a link to a forum presenting one closed benchmark avoiding hardware details (really any detail is avoided). And the benchmark is biased to every place posible: presenting the data in such a way that you visually compare the FX stock with the Intel overclocked, measuring power from the wall, using overclocked memory on the Intel but stock memory on the FX, extrapolating the Intel data far beyond its maximum speed... and all for finally obtaining a real 13% gain which reads to a few seconds.

3) I notice how every fanboy insists on the existence of little software that uses 8-cores. Not only they miss that an entire generation of next games will be optimized for 8-cores thanks to consoles as the PS4 using a 8-cores chip from AMD. But that the fanboys also omit that people often runs two-three applications at once. It is not usual to see a FX-8350 user gaming whereas doing some background work. Again this is a typical situation where the FX shows its real performance.

4) Yes you can pick a pair of selective games and configurations where the FX-8350 does not look so good. Toms have something important to say about your choice of Skyrim as example (bold font from mine):

Quote:
Skyrim appears to be the most CPU-dependent game in today’s suite. It also appears to be the most heavily slanted toward Intel's architecture. AMD's FX-8350 appears adequate across all of the tested settings, though we do have a little more data to discuss.


And still they say "AMD's FX-8350 appears adequate across all of the tested settings".

You cannot take one of the most unfavourable case possibles and pretend that it is something that happens more often.

It is also interesting that you give WoW as example of how you believe the FX "is noticeably slower". Are you serious? A difference of 10FPS above >80FPS? I suppose that you mean that those 10FPS are noticeable only when your see the bars in the graph, because you cannot detect the difference between >80FPS and 90FPS when playing the game in a 60Hz monitor, you cannot even using a 80Hz monitor...

And of course, one again, you ignore what Toms says in the cpu gaming hierachy chart (updated to this month), where the FX is placed near the top and emphasized that you do not see overall gain performance when going to a more expensive i5/i7. See also point (3).




I have said that on some CPU bound games like Civilization 5 and others the fx8350 can be noticeably slower.Even in hitman absoultion which is quite a recent game it was slower.Nobody is saying that fx8350 is a bad processor or something.Without a doubt the most important things for gaming in general is the GPU,but in some CPU heavy games the difference between an fx8350 and an i5 3570k is noticeable.Even in games like borderlands2 fx8350 lagged behind :http://www.xbitlabs.com/images/cpu/fx-8350-8320-6300-43... if GPU is taken common to both set ups then i5 3570k definitely looks the better option for gaming overall as of now.For most games there won't be any noticeable difference but for some like skyrim i5 3570k would be better.That's nothing to be selective about.The thing exists so it exists.Fx8350 doesn't perform well in some games that's it.There are enough guys with an 120Hz monitor as well.You are quoting toms.Let me quote one
"At the end of the day, AMD still has work to do in improving game performance. But Piledriver certainly does help rectify the slide backward we saw Bulldozer taking relative to some of AMD’s previous quad-core parts in processor-bound games."

About that thread check it out here:http://forums.anandtech.com/showthread.php?t=2294486
Idontcare is pretty reliable.Rendering on 3DS MAX which takes all cores to maximum,here i5 3570k was able to keep up with fx8350,that was the point to show how an OCed i5 3570k can keep up with an OCed fx8350 even on a thing which scales so well on cores.One just can't ignore the light threaded performance.It is not as if i5 3570k is a lame processor just becoz it has only 4 cores.It does the perform quite well with what it has and consumes lot lesser power when overclocked,though power consumption might not be a concern here.

World record!! Not practical at all.How many people use LN2 for daily stuffs?Both ivy i5/i7 k and fx8350 overclock fairly well.Problem with ivy processors is heat produced which limits OC potential which is not as good as SB processors as intel cheap out on TIM.Still ivy i5/i7 do 4.7-4.8GHz mostly on a good board.By the way though it is not something that would concern everyone but at higher clocks the fx8350 eats out much more power than an ivy i5/i7 at same clocks.http://www.tomshardware.com/reviews/build-a-pc-overcloc... does tell that i5 3570k when overclocked is very good and both fx8350 and i5 3570k traded blows in some.


Quotes from that page on toms "When it comes to overclocking though, Intel extends its lead with significantly lower power consumption and much better performance. If we were measuring efficiency, that'd be a home run"
"It'd be a great experiment, and we might even play around with it in the future, but it's clear that Intel's Core i5-3570K remains the better choice for overclockers in this price range."

For the things tested i5 3570k was overall better when both were overclocked.They are far more sensitive to frequency than fx8350 is.So simply talking about clocks won't do.IPC is a thing as well.So the so called "superior clocks" are just not the only thing,it's also important how much they are sensitive to those clocks.


ericjohn004 has explained things quite well.I think you are just not ready to look on other side.He and I were talking about both side of things but you just don't seem to acknowledge what is on other side.I do not support or am loyal to a company.I would pick whatever best for the money is available within what I can spend which is best suited to things I would do.We all want AMD to do well simply becoz it would result in more competition and better priced products from both sides.But that doesn't mean we should not look at the overall picture of the things now.I have made my points on the context,period.
April 1, 2013 4:21:39 PM

drinvis said:

I have said that on some CPU bound games like Civilization 5 and others the fx8350 can be noticeably slower.Even in hitman absoultion which is quite a recent game it was slower.Nobody is saying that fx8350 is a bad processor or something.Without a doubt the most important things for gaming in general is the GPU,but in some CPU heavy games the difference between an fx8350 and an i5 3570k is noticeable.Even in games like borderlands2 fx8350 lagged behind :http://www.xbitlabs.com/images/cpu/fx-8350-8320-6300-43... if GPU is taken common to both set ups then i5 3570k definitely looks the better option for gaming overall as of now.For most games there won't be any noticeable difference but for some like skyrim i5 3570k would be better.That's nothing to be selective about.The thing exists so it exists.Fx8350 doesn't perform well in some games that's it.There are enough guys with an 120Hz monitor as well.You are quoting toms.Let me quote one
"At the end of the day, AMD still has work to do in improving game performance. But Piledriver certainly does help rectify the slide backward we saw Bulldozer taking relative to some of AMD’s previous quad-core parts in processor-bound games."

About that thread check it out here:http://forums.anandtech.com/showthread.php?t=2294486
Idontcare is pretty reliable.Rendering on 3DS MAX which takes all cores to maximum,here i5 3570k was able to keep up with fx8350,that was the point to show how an OCed i5 3570k can keep up with an OCed fx8350 even on a thing which scales so well on cores.One just can't ignore the light threaded performance.It is not as if i5 3570k is a lame processor just becoz it has only 4 cores.It does the perform quite well with what it has and consumes lot lesser power when overclocked,though power consumption might not be a concern here.

World record!! Not practical at all.How many people use LN2 for daily stuffs?Both ivy i5/i7 k and fx8350 overclock fairly well.Problem with ivy processors is heat produced which limits OC potential which is not as good as SB processors as intel cheap out on TIM.Still ivy i5/i7 do 4.7-4.8GHz mostly on a good board.By the way though it is not something that would concern everyone but at higher clocks the fx8350 eats out much more power than an ivy i5/i7 at same clocks.http://www.tomshardware.com/reviews/build-a-pc-overcloc... does tell that i5 3570k when overclocked is very good and both fx8350 and i5 3570k traded blows in some.


Quotes from that page on toms "When it comes to overclocking though, Intel extends its lead with significantly lower power consumption and much better performance. If we were measuring efficiency, that'd be a home run"
"It'd be a great experiment, and we might even play around with it in the future, but it's clear that Intel's Core i5-3570K remains the better choice for overclockers in this price range."

For the things tested i5 3570k was overall better when both were overclocked.They are far more sensitive to frequency than fx8350 is.So simply talking about clocks won't do.IPC is a thing as well.So the so called "superior clocks" are just not the only thing,it's also important how much they are sensitive to those clocks.


ericjohn004 has explained things quite well.I think you are just not ready to look on other side.He and I were talking about both side of things but you just don't seem to acknowledge what is on other side.I do not support or am loyal to a company.I would pick whatever best for the money is available within what I can spend which is best suited to things I would do.We all want AMD to do well simply becoz it would result in more competition and better priced products from both sides.But that doesn't mean we should not look at the overall picture of the things now.I have made my points on the context,period.



1) What you really said is quoted above. Borderlands 2 is not using all the potential of the FX and still it was only slightly slower than the expensive i7-3770k



Hitman is not using all the potential of the FX, albeit you present it as new, "new" does not mean "parallelized for eight cores".

The point here is that FX is good enough for gaming and you are not going to change that by posting here.

2) It is good that you quote Toms now, but are you aware that the quote that you gave is a bit old? In their cpu gaming hierarchy cart (updated to March 2013). Toms puts the FX near the top and say, in the text, that upgrading from a FX-8350 to an expensive i5-3570K/i7-3770k does not offer overall gaming benefit. You and another user pretend otherwise but precisely both of you avoid Toms advice.

3) As explained before, Idontcare comparison was biased in any possible point. No, 3DS MAX is not taking all FX cores to maximum. Where did you got that exaggeration? In any case the same forum that you link says it clear

Quote:
So in these 2 workloads 8350 is actually on par or faster than 3770K






You pretend that "the i5 3570k was able to keep up with fx8350", still data shows otherwise. Moreover, the same 3DS MAX review says that the new FX chip allowed AMD to beat the entire range of LGA 1155 Intel chips:

Quote:
Ceci permet à AMD de prendre les devants sur toute la gamme LGA 1155.


4) Who said you that everyone will be obtaining a world record at her/his home? The world record shows that the design of the FX chip is superior to that of the Intel chips. A more solid design is the reason why it is so easy to overclock an AMD FX beyond Intel chips.

5) You quote toms overcloking review of the FX-8350, but once again you forget to quote relevant info. That review was made using older software, favourable to Intel chips as the i5. Toms write in the same page that you quote:

Quote:
This will likely change as we fold more heavily-threaded tests into the Marathon, starting this quarter.


Moreover, you also forgot to say that the default Intel builts used stock memory, but the default FX build used under-clocked memory, thus lowering its performance.

The point is: if you force yourself to using older software or one application at once the i5 looks a better option. If you run several applications at once and/or modern optimized software the FX looks better.
a b à CPUs
April 1, 2013 6:18:38 PM

shinte122305 said:
SO much people going back and fourth intel better amd better.... POINT is practically ALL games are gpu bottlenecked my best friend has a overclocked 680 with his phenom 2 x 4 clocked at 3.5ghz and he gets 60 frames in crysis 3 and far cry 3 and bio shock all at max in one monitor at 1080p..... point is the fx-8350 is more than enough all you need to worry about is GPU for gaming. i have an i7-3770k with a hd 7970 and i play all those games exactly the same at 60 frames.... get w/e cpu you want focus on the video card its where the performance is at an i5 or the fx-8350 you wont tell the differance


Until you start playing something like Skyrim and quite a few others. Then you'll notice a difference. But I get your point. But you can't assume a CPU doesn't matter at all. If that were the case then Tom's wouldn't recommend "Best gaming CPU's for the Money". And the 3570k is the best one available.
a b à CPUs
April 1, 2013 6:55:20 PM

juanrga said:
drinvis said:

I have said that on some CPU bound games like Civilization 5 and others the fx8350 can be noticeably slower.Even in hitman absoultion which is quite a recent game it was slower.Nobody is saying that fx8350 is a bad processor or something.Without a doubt the most important things for gaming in general is the GPU,but in some CPU heavy games the difference between an fx8350 and an i5 3570k is noticeable.Even in games like borderlands2 fx8350 lagged behind :http://www.xbitlabs.com/images/cpu/fx-8350-8320-6300-43... if GPU is taken common to both set ups then i5 3570k definitely looks the better option for gaming overall as of now.For most games there won't be any noticeable difference but for some like skyrim i5 3570k would be better.That's nothing to be selective about.The thing exists so it exists.Fx8350 doesn't perform well in some games that's it.There are enough guys with an 120Hz monitor as well.You are quoting toms.Let me quote one
"At the end of the day, AMD still has work to do in improving game performance. But Piledriver certainly does help rectify the slide backward we saw Bulldozer taking relative to some of AMD’s previous quad-core parts in processor-bound games."

About that thread check it out here:http://forums.anandtech.com/showthread.php?t=2294486
Idontcare is pretty reliable.Rendering on 3DS MAX which takes all cores to maximum,here i5 3570k was able to keep up with fx8350,that was the point to show how an OCed i5 3570k can keep up with an OCed fx8350 even on a thing which scales so well on cores.One just can't ignore the light threaded performance.It is not as if i5 3570k is a lame processor just becoz it has only 4 cores.It does the perform quite well with what it has and consumes lot lesser power when overclocked,though power consumption might not be a concern here.

World record!! Not practical at all.How many people use LN2 for daily stuffs?Both ivy i5/i7 k and fx8350 overclock fairly well.Problem with ivy processors is heat produced which limits OC potential which is not as good as SB processors as intel cheap out on TIM.Still ivy i5/i7 do 4.7-4.8GHz mostly on a good board.By the way though it is not something that would concern everyone but at higher clocks the fx8350 eats out much more power than an ivy i5/i7 at same clocks.http://www.tomshardware.com/reviews/build-a-pc-overcloc... does tell that i5 3570k when overclocked is very good and both fx8350 and i5 3570k traded blows in some.


Quotes from that page on toms "When it comes to overclocking though, Intel extends its lead with significantly lower power consumption and much better performance. If we were measuring efficiency, that'd be a home run"
"It'd be a great experiment, and we might even play around with it in the future, but it's clear that Intel's Core i5-3570K remains the better choice for overclockers in this price range."

For the things tested i5 3570k was overall better when both were overclocked.They are far more sensitive to frequency than fx8350 is.So simply talking about clocks won't do.IPC is a thing as well.So the so called "superior clocks" are just not the only thing,it's also important how much they are sensitive to those clocks.


ericjohn004 has explained things quite well.I think you are just not ready to look on other side.He and I were talking about both side of things but you just don't seem to acknowledge what is on other side.I do not support or am loyal to a company.I would pick whatever best for the money is available within what I can spend which is best suited to things I would do.We all want AMD to do well simply becoz it would result in more competition and better priced products from both sides.But that doesn't mean we should not look at the overall picture of the things now.I have made my points on the context,period.



1) What you really said is quoted above. Borderlands 2 is not using all the potential of the FX and still it was only slightly slower than the expensive i7-3770k



Hitman is not using all the potential of the FX, albeit you present it as new, "new" does not mean "parallelized for eight cores".

The point here is that FX is good enough for gaming and you are not going to change that by posting here.

2) It is good that you quote Toms now, but are you aware that the quote that you gave is a bit old? In their cpu gaming hierarchy cart (updated to March 2013). Toms puts the FX near the top and say, in the text, that upgrading from a FX-8350 to an expensive i5-3570K/i7-3770k does not offer overall gaming benefit. You and another user pretend otherwise but precisely both of you avoid Toms advice.

3) As explained before, Idontcare comparison was biased in any possible point. No, 3DS MAX is not taking all FX cores to maximum. Where did you got that exaggeration? In any case the same forum that you link says it clear

Quote:
So in these 2 workloads 8350 is actually on par or faster than 3770K






You pretend that "the i5 3570k was able to keep up with fx8350", still data shows otherwise. Moreover, the same 3DS MAX review says that the new FX chip allowed AMD to beat the entire range of LGA 1155 Intel chips:

Quote:
Ceci permet à AMD de prendre les devants sur toute la gamme LGA 1155.


4) Who said you that everyone will be obtaining a world record at her/his home? The world record shows that the design of the FX chip is superior to that of the Intel chips. A more solid design is the reason why it is so easy to overclock an AMD FX beyond Intel chips.

5) You quote toms overcloking review of the FX-8350, but once again you forget to quote relevant info. That review was made using older software, favourable to Intel chips as the i5. Toms write in the same page that you quote:

Quote:
This will likely change as we fold more heavily-threaded tests into the Marathon, starting this quarter.


Moreover, you also forgot to say that the default Intel builts used stock memory, but the default FX build used under-clocked memory, thus lowering its performance.

The point is: if you force yourself to using older software or one application at once the i5 looks a better option. If you run several applications at once and/or modern optimized software the FX looks better.


You like to quote Tom's Hardware right? Well I got quite a few things Tom's says about the 8350 RIGHT HERE.

Dude seriously just stop. I've never seen the word "biased" used so many times in my life. I GET YOUR POINT! MY point is that an FX is still kind of lacking as far as being an OVERALL great chip because of it's lack of single threaded performance. It's missing that.

To say that's Tom's Hardware or any other site is "bottlenecking" an AMD cpu just because they used 1600Mhz memory is absolutely ludicrous. Just read the Tom's review on the FX8350. It clearly states that they used 1600mhz memory because they tested with BOTH 1600mhz and 1866mhz and they got the same results with BOTH. And when they didn't get the same results with both, they stated so. And the only benchmark that happened in was, guess what? MEMORY BANDWIDTH! So they DID use both but only got different results with one test. And if you don't believe Tom's then IDK what to tell you. Tom's even has an entire article dedicated to proving this myth false. If you would believe facts, but sometimes I guess you choose not to. You like to quote Tom's hardware but it doesn't seem like you even read the 8350 review they did. If you quote Tom's trying to say they are 100% accurate then why don't you believe what they say about the 8350 in this review? Seems like your nit picking your talking points, heh?

I've seen reviews that completely dis the AMD FX8350. I've seen reviews that say the 8350 "beats" a 3570k. I've seen a whole lot more that say the 3570k "wins". The only review site that I trust is Tom's Hardware, and they give a VERY fair review of the 8350. Clearly, gaming/single threaded performance is not on par with the 3570k. And gaming/single threaded performance is at least 65% of the whole picture.

Multithreaded performance is BETTER on an 8350 than a 3570k but a 3570k's multithreaded performance is still very good nonetheless. Especially if you match them up clock for clock. So a 3570k doesn't lack at anything. So MY POINT is, is that a 3570k is a more well rounded chip. Because it's good with everything. And a 3770k is good with EVERYTHING, and is also better than the 8350 with multithreading as per Tom's Hardware. And since your such a fan of Tom's and because you like to quote Tom's, then Tom's must be right about this, RIGHT? Due to your logic, Tom's should be right, RIGHT?

Tom's has BEEN dispelled the rumor that the 8150's and 8350's perform that much better with 1866mhz memory. Why are you even arguing this point when it's clearly NOT true. The research has been done, the facts are out, your NOT right. Your WRONG. Stop pushing all this 8350 propaganda when it's clearly not true as far as Tom's Hardware is concerned. I just went from 1600 to 2133 and I saw a 3% performance gain. 3%! Going to 1866mhz would most likely be 1.5%. So ok, I'll give the 8350 a 1.5% advantage next time I take a look at a benchmark.

How can you possibly disagree with anything I've said? I know the facts, you probably know the facts but choose to only acknowledge the ones you like. Why hide from the truth. It'll set you free.
a b à CPUs
April 1, 2013 7:36:23 PM

You simply can't win an argument when your arguing with the facts themselves. Until you realize this no one will take you seriously
April 2, 2013 2:42:06 AM

ericjohn004 said:

You like to quote Tom's Hardware right? Well I got quite a few things Tom's says about the 8350 RIGHT HERE.

Dude seriously just stop. I've never seen the word "biased" used so many times in my life. I GET YOUR POINT! MY point is that an FX is still kind of lacking as far as being an OVERALL great chip because of it's lack of single threaded performance. It's missing that.

To say that's Tom's Hardware or any other site is "bottlenecking" an AMD cpu just because they used 1600Mhz memory is absolutely ludicrous. Just read the Tom's review on the FX8350. It clearly states that they used 1600mhz memory because they tested with BOTH 1600mhz and 1866mhz and they got the same results with BOTH. And when they didn't get the same results with both, they stated so. And the only benchmark that happened in was, guess what? MEMORY BANDWIDTH! So they DID use both but only got different results with one test. And if you don't believe Tom's then IDK what to tell you. Tom's even has an entire article dedicated to proving this myth false. If you would believe facts, but sometimes I guess you choose not to. You like to quote Tom's hardware but it doesn't seem like you even read the 8350 review they did. If you quote Tom's trying to say they are 100% accurate then why don't you believe what they say about the 8350 in this review? Seems like your nit picking your talking points, heh?

I've seen reviews that completely dis the AMD FX8350. I've seen reviews that say the 8350 "beats" a 3570k. I've seen a whole lot more that say the 3570k "wins". The only review site that I trust is Tom's Hardware, and they give a VERY fair review of the 8350. Clearly, gaming/single threaded performance is not on par with the 3570k. And gaming/single threaded performance is at least 65% of the whole picture.

Multithreaded performance is BETTER on an 8350 than a 3570k but a 3570k's multithreaded performance is still very good nonetheless. Especially if you match them up clock for clock. So a 3570k doesn't lack at anything. So MY POINT is, is that a 3570k is a more well rounded chip. Because it's good with everything. And a 3770k is good with EVERYTHING, and is also better than the 8350 with multithreading as per Tom's Hardware. And since your such a fan of Tom's and because you like to quote Tom's, then Tom's must be right about this, RIGHT? Due to your logic, Tom's should be right, RIGHT?

Tom's has BEEN dispelled the rumor that the 8150's and 8350's perform that much better with 1866mhz memory. Why are you even arguing this point when it's clearly NOT true. The research has been done, the facts are out, your NOT right. Your WRONG. Stop pushing all this 8350 propaganda when it's clearly not true as far as Tom's Hardware is concerned. I just went from 1600 to 2133 and I saw a 3% performance gain. 3%! Going to 1866mhz would most likely be 1.5%. So ok, I'll give the 8350 a 1.5% advantage next time I take a look at a benchmark.

How can you possibly disagree with anything I've said? I know the facts, you probably know the facts but choose to only acknowledge the ones you like. Why hide from the truth. It'll set you free.


In the past you already commented on stuff you did not even read, which says a lot of...

Now you chose to misinterpret me, to avoid relevant parts of my above message, and to try an appeal to authority fallacy. Toms' analysis of memory scaling of the AMD FX was not correct (they even selected memory modules designed for a specific Intel chipset). Moreover their test was for the Bulldozer architecture not Piledriver. The facts are that people have noted increase in gaming performance by using faster memory in an FX-8350 chip.
a b à CPUs
April 2, 2013 4:43:27 AM

shinte122305 said:
SO much people going back and fourth intel better amd better.... POINT is practically ALL games are gpu bottlenecked my best friend has a overclocked 680 with his phenom 2 x 4 clocked at 3.5ghz and he gets 60 frames in crysis 3 and far cry 3 and bio shock all at max in one monitor at 1080p..... point is the fx-8350 is more than enough all you need to worry about is GPU for gaming. i have an i7-3770k with a hd 7970 and i play all those games exactly the same at 60 frames.... get w/e cpu you want focus on the video card its where the performance is at an i5 or the fx-8350 you wont tell the differance



This.
a b à CPUs
April 2, 2013 8:54:36 AM

juanrga said:
ericjohn004 said:

You like to quote Tom's Hardware right? Well I got quite a few things Tom's says about the 8350 RIGHT HERE.

Dude seriously just stop. I've never seen the word "biased" used so many times in my life. I GET YOUR POINT! MY point is that an FX is still kind of lacking as far as being an OVERALL great chip because of it's lack of single threaded performance. It's missing that.

To say that's Tom's Hardware or any other site is "bottlenecking" an AMD cpu just because they used 1600Mhz memory is absolutely ludicrous. Just read the Tom's review on the FX8350. It clearly states that they used 1600mhz memory because they tested with BOTH 1600mhz and 1866mhz and they got the same results with BOTH. And when they didn't get the same results with both, they stated so. And the only benchmark that happened in was, guess what? MEMORY BANDWIDTH! So they DID use both but only got different results with one test. And if you don't believe Tom's then IDK what to tell you. Tom's even has an entire article dedicated to proving this myth false. If you would believe facts, but sometimes I guess you choose not to. You like to quote Tom's hardware but it doesn't seem like you even read the 8350 review they did. If you quote Tom's trying to say they are 100% accurate then why don't you believe what they say about the 8350 in this review? Seems like your nit picking your talking points, heh?

I've seen reviews that completely dis the AMD FX8350. I've seen reviews that say the 8350 "beats" a 3570k. I've seen a whole lot more that say the 3570k "wins". The only review site that I trust is Tom's Hardware, and they give a VERY fair review of the 8350. Clearly, gaming/single threaded performance is not on par with the 3570k. And gaming/single threaded performance is at least 65% of the whole picture.

Multithreaded performance is BETTER on an 8350 than a 3570k but a 3570k's multithreaded performance is still very good nonetheless. Especially if you match them up clock for clock. So a 3570k doesn't lack at anything. So MY POINT is, is that a 3570k is a more well rounded chip. Because it's good with everything. And a 3770k is good with EVERYTHING, and is also better than the 8350 with multithreading as per Tom's Hardware. And since your such a fan of Tom's and because you like to quote Tom's, then Tom's must be right about this, RIGHT? Due to your logic, Tom's should be right, RIGHT?

Tom's has BEEN dispelled the rumor that the 8150's and 8350's perform that much better with 1866mhz memory. Why are you even arguing this point when it's clearly NOT true. The research has been done, the facts are out, your NOT right. Your WRONG. Stop pushing all this 8350 propaganda when it's clearly not true as far as Tom's Hardware is concerned. I just went from 1600 to 2133 and I saw a 3% performance gain. 3%! Going to 1866mhz would most likely be 1.5%. So ok, I'll give the 8350 a 1.5% advantage next time I take a look at a benchmark.

How can you possibly disagree with anything I've said? I know the facts, you probably know the facts but choose to only acknowledge the ones you like. Why hide from the truth. It'll set you free.


In the past you already commented on stuff you did not even read, which says a lot of...

Now you chose to misinterpret me, to avoid relevant parts of my above message, and to try an appeal to authority fallacy. Toms' analysis of memory scaling of the AMD FX was not correct (they even selected memory modules designed for a specific Intel chipset). Moreover their test was for the Bulldozer architecture not Piledriver. The facts are that people have noted increase in gaming performance by using faster memory in an FX-8350 chip.


You should be banned for spreading misinformation and your biased thoughts and opinions. You should honestly be banned. How do you disagree with benchmarks?! How?!! It's evidence right in front of you and you still argue with it! WHY?! Why do you people not get it?

3570k>8350 for gaming

/thread
a c 210 à CPUs
April 2, 2013 9:22:00 AM

trogdor796 said:

You should be banned for spreading misinformation and your biased thoughts and opinions. You should honestly be banned. How do you disagree with benchmarks?! How?!! It's evidence right in front of you and you still argue with it! WHY?! Why do you people not get it?

3570k>8350 for gaming

/thread


That's a gross over generalization...

In single threaded games 3570k > 8350

In multi-threaded games 8350 > 3570k in many instances...

What it boils down to, is that the 8350 is about on par with a 3570k across a broad spectrum of games, and it is right on the heels of a 3770k in a lot of things outside of gaming.
a b à CPUs
April 2, 2013 10:12:10 AM

8350rocks said:
trogdor796 said:

You should be banned for spreading misinformation and your biased thoughts and opinions. You should honestly be banned. How do you disagree with benchmarks?! How?!! It's evidence right in front of you and you still argue with it! WHY?! Why do you people not get it?

3570k>8350 for gaming

/thread


That's a gross over generalization...

In single threaded games 3570k > 8350

In multi-threaded games 8350 > 3570k in many instances...

What it boils down to, is that the 8350 is about on par with a 3570k across a broad spectrum of games, and it is right on the heels of a 3770k in a lot of things outside of gaming.


Seriously? No, it's not. No legit benchmark exists that backs up your claims.

I've seen your posts, and all of them are filled with BS just like this one. You will never be able to prove your claim, because it's false. The 8350 is a good processor, but it doesn't touch the 3570k in gaming, and is destroyed by a 3770k. To say the 8350 is right on the heels of a 3770k proves how little you know and how biased you are.

Seriously, stop spreading misinformation and spewing out your biased thoughts.
a c 210 à CPUs
April 2, 2013 10:34:52 AM

trogdor796 said:
8350rocks said:
trogdor796 said:

You should be banned for spreading misinformation and your biased thoughts and opinions. You should honestly be banned. How do you disagree with benchmarks?! How?!! It's evidence right in front of you and you still argue with it! WHY?! Why do you people not get it?

3570k>8350 for gaming

/thread


That's a gross over generalization...

In single threaded games 3570k > 8350

In multi-threaded games 8350 > 3570k in many instances...

What it boils down to, is that the 8350 is about on par with a 3570k across a broad spectrum of games, and it is right on the heels of a 3770k in a lot of things outside of gaming.


Seriously? No, it's not. No legit benchmark exists that backs up your claims.

I've seen your posts, and all of them are filled with BS just like this one. You will never be able to prove your claim, because it's false. The 8350 is a good processor, but it doesn't touch the 3570k in gaming, and is destroyed by a 3770k. To say the 8350 is right on the heels of a 3770k proves how little you know and how biased you are.

Seriously, stop spreading misinformation and spewing out your biased thoughts.


Show me one Crysis 3 Benchmark or Far Cry 3 Benchmark or BF3 or Metro 2033 where the 3570k destroys the 8350. You can't produce one...a few FPS one way or another is less than the margin for error on the benchmark so that is an equivalent draw. You would need a greater than 10% difference in FPS to say that one is significantly better than the other as the margin for error on ALL your benchmarks is 5-10%. So go ahead and spend the next few days hunting for a benchmark on any of those 4 new games that proves your "unbiased" point.

Otherwise, go talk about the greatness of intel elsewhere...where some fool who hasn't done the research might believe you.
April 2, 2013 12:48:30 PM

To extend a bit on the memory issue.

As said above Toms tested the sensibility of bulldozer chips to memory speed but they did not use AMD optimized modules but four Intel optimized modules. Moreover at least two of the modules used in that test were explicitly designed for an Intel chipset (the X79) and failed at higher speeds.

Toms only tested two games: dirt 3 and metro 2033. Both games are well-suited for FX chips (the piledriver 8350 run them so fast as an i7-3770k) and thus they did not benefit from faster ram. Going from 1600 to 1866 only generated about a 1% more FPS.

However there are games more sensible to faster ram like the next



where going from 1600 to 1866 gives about a 4% more performance on Bulldozer FX-8150. This 4% increase is 2.7 times more than the fair estimation made by the well-informed poster ericjohn004.

Yes, 4% is little, but I have not said that it is the maximum increase possible. I am going to show how we can go from "no benefit" to 4% by selecting a different game. People reports up to a 25% increase in frame rate from going to 1333 to 2133 on Civilization V on the full render mode. Taking a linear scaling this gives about a 10% increase when going from 1600 to 1866 on Civ V. Faster RAM is also noticeable when running multiple apps at once or in specific tasks such as transcoding.

No here where overall I find good info, but I find lots of reviews on the internet that insist on feeding the FX processors with underclocked memory, selecting applications not specially favourable on the FX side, some non-optimized memory modules, and avoiding OS patches and a pair of tricks more... and you are going to change significantly the conclusion of a 'review'.
a b à CPUs
April 2, 2013 12:49:58 PM

8350rocks, I know you've said that the FX8350 doesn't run as fast when it's not using 1866mhz memory. I and I trust what you say to be true. But you did say that while both of us were arguing and sometimes we may have just said certain things to prove our point. And I'm not sure whether you said that to prove a point of whether you said that because it was really true.

From what Tom's says, an 8150 doesn't benefit with 1866mhz memory. They say one benchmark gave a 6% better results but most did nothing. They did an entire article on it. And if you read their 8350 article, they say they used both 1866 and 1600 for the review and only saw a difference in memory bandwidth. So they decided to stick with 1600mhz to be equal. They did a benchmark with 1066, 1333, 1600, and 1866. And performance got better with each change of Mhz but it stopped after 1600mhz. And 1600mhz and 1866mhz showed the same results.

Now, me personally, I do know that higher Mhz memory does give additional performance. I just went up from 1600-2133mhz and not only did my RAM scores shoot up I also got a 3% increase in most things I did with my CPU too. And that's like being clocked .1Ghz higher. Some benchmarks don't show this increase and some did. So I'm happy with it. So therefore, I know that an AMD 8350 moving up from 1600mhz to 1866mhz, will have a benefit in some benchmarks. But to me, that benefit should only be 1-2%, if that, and it should only benefit in specific benchmarks that are memory intensive.

So are the benchmark results really drastically different like some people would like you to believe? Will it make the 8350 all of a sudden surpass the 3770k in a benchmark it would normally fall behind? Or can this just sometimes be an 8350 fans talking point? From what I've researched, it's basically just a talking point because you should see a 1-2% gain. And in gaming I see no difference with the different RAM. Not that I can physically see.

BTW, you won't see any BF3 or Crysis 3 or Metro 2033 benchmark that a 3570k destroys an 8350. The only games, as I'm sure you know, that do show a difference, are games that use 4 cores or less and require a lot of CPU power. Like Skyrim.

Most games, especially at 1080p and higher resolutions, rely so much on the graphics card that the CPU doesn't matter as much. So basically, the higher settings and the higher resolution you use, the more the CPU doesn't even matter. If you gaming at like 1200x900 at low settings on any game that's not brand new, you'll want a 3570k. Otherwise, it won't matter too much unless your playing something like Skyrim. When games can use 8 cores the 8350 is actually really good. Like in 3DMark11 Physics test.
a b à CPUs
April 2, 2013 12:56:28 PM

FANBOYS!!! I'm sick and tired of your bickering... The FX 8350 is enough to maintain 60FPS (even in skyrim), and If you have a 60Hz monitor, the difference is negligible between the i7 and the FX 8350 because you have to turn on vsync anyways or you're going to get screen tearing. And the 8 cores make for a better multitasking experience. Even a stock FX 8350 with crossfire 7970s on a 2560x1600 monitor is going to be pretty close to intel in gaming performance. Also, clock for clock, INTEL IS FASTER. But performance isn't measured that way... If you were to measure performance 500Mhz past stock clocks on each system, that would be a fair benchmark. You guys have no idea what you're talking about if you're saying a i5 at 4GHz pwns a FX 8350 at 4 GHz. DUH! No shiznit sherlock, that's the FX 8350's stock clock. And, if you didn't know, most benchmarks are done at stock clocks with the AMD Processors, so you don't see that extra 200Mhz. I've seen multiple benchmarks done, and I've caught this.

The only time I would recommend a i5 over the FX 8350 is if you're going to be using a 120Hz monitor. If you check out (not the Tom's or Anand) benchmarks,
Multiple sources show that the FX 8350 beats out the i7 3770k in Metro 2033.

In case you guys didn't know... The horse is dead (and so was the thread), stop trying to kill it.
a b à CPUs
April 2, 2013 12:57:32 PM

juanrga said:
To extend a bit on the memory issue.

As said above Toms tested the sensibility of bulldozer chips to memory speed but they did not use AMD optimized modules but four Intel optimized modules. Moreover at least two of the modules used in that test were explicitly designed for an Intel chipset (the X79) and failed at higher speeds.

Toms only tested two games: dirt 3 and metro 2033. Both games are well-suited for FX chips (the piledriver 8350 run them so fast as an i7-3770k) and thus they did not benefit from faster ram. Going from 1600 to 1866 only generated about a 1% more FPS.

However there are games more sensible to faster ram like the next



where going from 1600 to 1866 gives about a 4% more performance on Bulldozer FX-8150. This 4% increase is 2.7 times more than the fair estimation made by the well-informed poster ericjohn004.

Yes, 4% is little, but I have not said that it is the maximum increase possible. I am going to show how we can go from "no benefit" to 4% by selecting a different game. People reports up to a 25% increase in frame rate from going to 1333 to 2133 on Civilization V on the full render mode. Taking a linear scaling this gives about a 10% increase when going from 1600 to 1866 on Civ V. Faster RAM is also noticeable when running multiple apps at once or in specific tasks such as transcoding.

No here where overall I find good info, but I find lots of reviews on the internet that insist on feeding the FX processors with underclocked memory, selecting applications not specially favourable on the FX side, some non-optimized memory modules, and avoiding OS patches and a pair of tricks more... and you are going to change significantly the conclusion of a 'review'.


Yeah, I agree with you there. Some people are totally not fair to the FX. I read this review yesterday. And everything the FX did great, they were like "well it seems to be ok at this, but look at this single threaded benchmark!". So I see why if you own an FX you'd be completely pissed at some of the reviews out there.

When I actually read Tom's review on the 8350, it surprised the hell out of me. Not only was it literally nipping at the heels of the 3770k like you say, in some of the benchmarks, but it wasn't as bad at single threaded tasks like I had once thought. I reread the review yesterday since we were chatting about this and I was surprised the say the least. And Tom's does do a very fair review. They don't even use SuperPi and some of the other stuff that's not optimized for AMD.

And like I said in my last post. I'm sure 1866mhz memory does give the 8350 an improvement. I just don't think it's quite as much as some people would like to believe. Maybe it does, IDK, I haven't read as much into it as you have.

But yeah, the 8350 is really super impressive as far as multithreading goes. And in theoretical benchmarks like Sisoft Sandra(which I hate, it give my 3570k low scores), the 8350 does really well. And for only 179.99, I can definitely see why people get this processor. I'm going to be doing a build for my little nephew. He wants a cheaper gaming computer. And I think I'm going to stick an FX6300 in there. You absolutely can't beat the FX6300 for only 129.99. Can't beat it.

a c 210 à CPUs
April 2, 2013 1:01:17 PM

ericjohn004 said:
8350rocks, I know you've said that the FX8350 doesn't run as fast when it's not using 1866mhz memory. I and I trust what you say to be true. But you did say that while both of us were arguing and sometimes we may have just said certain things to prove our point. And I'm not sure whether you said that to prove a point of whether you said that because it was really true.

From what Tom's says, an 8150 doesn't benefit with 1866mhz memory. They say one benchmark gave a 6% better results but most did nothing. They did an entire article on it. And if you read their 8350 article, they say they used both 1866 and 1600 for the review and only saw a difference in memory bandwidth. So they decided to stick with 1600mhz to be equal. They did a benchmark with 1066, 1333, 1600, and 1866. And performance got better with each change of Mhz but it stopped after 1600mhz. And 1600mhz and 1866mhz showed the same results.


That is a discrepancy, that, as a hardware guy, I cannot explain...there should be an increase in performance there...unless...the motherboard did not support higher memory bandwidth than 1600 MHz.

Quote:
Now, me personally, I do know that higher Mhz memory does give additional performance. I just went up from 1600-2133mhz and not only did my RAM scores shoot up I also got a 3% increase in most things I did with my CPU too. And that's like being clocked .1Ghz higher. Some benchmarks don't show this increase and some did. So I'm happy with it. So therefore, I know that an AMD 8350 moving up from 1600mhz to 1866mhz, will have a benefit in some benchmarks. But to me, that benefit should only be 1-2%, if that, and it should only benefit in specific benchmarks that are memory intensive.

So are the benchmark results really drastically different like some people would like you to believe? Will it make the 8350 all of a sudden surpass the 3770k in a benchmark it would normally fall behind? Or can this just sometimes be an 8350 fans talking point? From what I've researched, it's basically just a talking point because you should see a 1-2% gain. And in gaming I see no difference with the different RAM. Not that I can physically see.


In serial applications it makes less difference, in highly threaded applications it makes more difference...though your number is close enough to accurate...you should typically see not more than 5% performance gain on highly threaded applications.

Quote:
BTW, you won't see any BF3 or Crysis 3 or Metro 2033 benchmark that a 3570k destroys an 8350. The only games, as I'm sure you know, that do show a difference, are games that use 4 cores or less and require a lot of CPU power. Like Skyrim.


I know, that's why I said what I did to trogdar or whatever his name was...his assertions were foolish and uninformed...so by inviting him to prove me wrong when I know he cannot, he will have to do his own research and LEARN the facts about what he is talking about rather than spewing trash about "i5>8350 /thread". Those comments were not addressed to you...they were addressed to him.

Quote:
Most games, especially at 1080p and higher resolutions, rely so much on the graphics card that the CPU doesn't matter as much. So basically, the higher settings and the higher resolution you use, the more the CPU doesn't even matter. If you gaming at like 1200x900 at low settings on any game that's not brand new, you'll want a 3570k. Otherwise, it won't matter too much unless your playing something like Skyrim. When games can use 8 cores the 8350 is actually really good. Like in 3DMark11 Physics test.


+1 THIS has been my point EXACTLY all along, to get people to do some research enough to see reality. That the difference between the 2 slightly favors one over the other in certain areas, but it's not enough to sweat anything on the weak points of either side. In a perfect world, I would love to see more support for AMD to continue competition between the 2, but it seems some intel fanboys need a dose of reality every now and again when they pop into a thread and spew BS about dominant supremacy from intel...the facts are...the performance gains are mostly trivial with some exceptions. I will grant you Skyrim as an exception...and I am sure there are a few others...but in general...buy what you can afford and try to buy the best VALUE for the money. You always get the most value for your $ that way.
a b à CPUs
April 2, 2013 1:15:16 PM

8350rocks said:
ericjohn004 said:
8350rocks, I know you've said that the FX8350 doesn't run as fast when it's not using 1866mhz memory. I and I trust what you say to be true. But you did say that while both of us were arguing and sometimes we may have just said certain things to prove our point. And I'm not sure whether you said that to prove a point of whether you said that because it was really true.

From what Tom's says, an 8150 doesn't benefit with 1866mhz memory. They say one benchmark gave a 6% better results but most did nothing. They did an entire article on it. And if you read their 8350 article, they say they used both 1866 and 1600 for the review and only saw a difference in memory bandwidth. So they decided to stick with 1600mhz to be equal. They did a benchmark with 1066, 1333, 1600, and 1866. And performance got better with each change of Mhz but it stopped after 1600mhz. And 1600mhz and 1866mhz showed the same results.


That is a discrepancy, that, as a hardware guy, I cannot explain...there should be an increase in performance there...unless...the motherboard did not support higher memory bandwidth than 1600 MHz.

Quote:
Now, me personally, I do know that higher Mhz memory does give additional performance. I just went up from 1600-2133mhz and not only did my RAM scores shoot up I also got a 3% increase in most things I did with my CPU too. And that's like being clocked .1Ghz higher. Some benchmarks don't show this increase and some did. So I'm happy with it. So therefore, I know that an AMD 8350 moving up from 1600mhz to 1866mhz, will have a benefit in some benchmarks. But to me, that benefit should only be 1-2%, if that, and it should only benefit in specific benchmarks that are memory intensive.

So are the benchmark results really drastically different like some people would like you to believe? Will it make the 8350 all of a sudden surpass the 3770k in a benchmark it would normally fall behind? Or can this just sometimes be an 8350 fans talking point? From what I've researched, it's basically just a talking point because you should see a 1-2% gain. And in gaming I see no difference with the different RAM. Not that I can physically see.


In serial applications it makes less difference, in highly threaded applications it makes more difference...though your number is close enough to accurate...you should typically see not more than 5% performance gain on highly threaded applications.

Quote:
BTW, you won't see any BF3 or Crysis 3 or Metro 2033 benchmark that a 3570k destroys an 8350. The only games, as I'm sure you know, that do show a difference, are games that use 4 cores or less and require a lot of CPU power. Like Skyrim.


I know, that's why I said what I did to trogdar or whatever his name was...his assertions were foolish and uninformed...so by inviting him to prove me wrong when I know he cannot, he will have to do his own research and LEARN the facts about what he is talking about rather than spewing trash about "i5>8350 /thread". Those comments were not addressed to you...they were addressed to him.

Quote:
Most games, especially at 1080p and higher resolutions, rely so much on the graphics card that the CPU doesn't matter as much. So basically, the higher settings and the higher resolution you use, the more the CPU doesn't even matter. If you gaming at like 1200x900 at low settings on any game that's not brand new, you'll want a 3570k. Otherwise, it won't matter too much unless your playing something like Skyrim. When games can use 8 cores the 8350 is actually really good. Like in 3DMark11 Physics test.


+1 THIS has been my point EXACTLY all along, to get people to do some research enough to see reality. That the difference between the 2 slightly favors one over the other in certain areas, but it's not enough to sweat anything on the weak points of either side. In a perfect world, I would love to see more support for AMD to continue competition between the 2, but it seems some intel fanboys need a dose of reality every now and again when they pop into a thread and spew BS about dominant supremacy from intel...the facts are...the performance gains are mostly trivial with some exceptions. I will grant you Skyrim as an exception...and I am sure there are a few others...but in general...buy what you can afford and try to buy the best VALUE for the money. You always get the most value for your $ that way.


Most benchmarks sites purposely try to isolate the CPU just to show you what CPU is better for gaming. But what they don't realize, is that no gamer with a decent card is going to game at 1200x900 and on low settings. And these benchmarks are the exact reason why most people, myself included, always thought the 3570k was dominant in gaming. But if you are a high end gamer on a budget and don't care too much about games like Skyrim, that are just not optimized for AMD, then I really don't think you'll ever notice that you have an 8350 and not a 3570k. The 3570k is technically "better" but it's not like your going to notice while gaming at 1920x1080 at Ultra settings.

I used to think totally different about this because of all the benchmarks I had read and how far they all had the 3570k. But the more research I do the more I find information like what I've said above.

a c 210 à CPUs
April 2, 2013 1:25:14 PM

ericjohn004 said:


Most benchmarks sites purposely try to isolate the CPU just to show you what CPU is better for gaming. But what they don't realize, is that no gamer with a decent card is going to game at 1200x900 and on low settings. And these benchmarks are the exact reason why most people, myself included, always thought the 3570k was dominant in gaming. But if you are a high end gamer on a budget and don't care too much about games like Skyrim, that are just not optimized for AMD, then I really don't think you'll ever notice that you have an 8350 and not a 3570k. The 3570k is technically "better" but it's not like your going to notice while gaming at 1920x1080 at Ultra settings.

I used to think totally different about this because of all the benchmarks I had read and how far they all had the 3570k. But the more research I do the more I find information like what I've said above.



You have realized that which the masses have not...

Benchmarks are a useful yet skewed interpretation of possibilities...

As I said to you once before, you don't spend a chunk of money on a gaming rig to run every game on super low settings. You want to run them on Max settings if you can...in that realm, the difference is nitpicking in 90% of the cases out there...

The FX6300 is a great buy...however, there is a FX6350 coming in June that will increase clock speed and tweak a few things on the FX6300...if you can wait a few months on the CPU, then it will be a similar price point, as the FX series CPUs except for the 8350 will see a decrease in price across the board.
a b à CPUs
April 2, 2013 1:27:24 PM

Lol, my earlier post was directed at the earlier post. I was a bit heated... Any who, The memory controller on AMD CPUs have been terrible for quite some time... How ever, they're getting closer to having a good one like Intel's (hence you see the difference between the FX 8350 and the FX 8150). And like 8350Rocks was saying, if you can afford a full upgrade in 12 months, the i5 or i7 would be worth it. But the difference isn't enough at the price point IMO.
April 2, 2013 4:04:55 PM

juanrga said:
drinvis said:

I have said that on some CPU bound games like Civilization 5 and others the fx8350 can be noticeably slower.Even in hitman absoultion which is quite a recent game it was slower.Nobody is saying that fx8350 is a bad processor or something.Without a doubt the most important things for gaming in general is the GPU,but in some CPU heavy games the difference between an fx8350 and an i5 3570k is noticeable.Even in games like borderlands2 fx8350 lagged behind :http://www.xbitlabs.com/images/cpu/fx-8350-8320-6300-43... if GPU is taken common to both set ups then i5 3570k definitely looks the better option for gaming overall as of now.For most games there won't be any noticeable difference but for some like skyrim i5 3570k would be better.That's nothing to be selective about.The thing exists so it exists.Fx8350 doesn't perform well in some games that's it.There are enough guys with an 120Hz monitor as well.You are quoting toms.Let me quote one
"At the end of the day, AMD still has work to do in improving game performance. But Piledriver certainly does help rectify the slide backward we saw Bulldozer taking relative to some of AMD’s previous quad-core parts in processor-bound games."

About that thread check it out here:http://forums.anandtech.com/showthread.php?t=2294486
Idontcare is pretty reliable.Rendering on 3DS MAX which takes all cores to maximum,here i5 3570k was able to keep up with fx8350,that was the point to show how an OCed i5 3570k can keep up with an OCed fx8350 even on a thing which scales so well on cores.One just can't ignore the light threaded performance.It is not as if i5 3570k is a lame processor just becoz it has only 4 cores.It does the perform quite well with what it has and consumes lot lesser power when overclocked,though power consumption might not be a concern here.

World record!! Not practical at all.How many people use LN2 for daily stuffs?Both ivy i5/i7 k and fx8350 overclock fairly well.Problem with ivy processors is heat produced which limits OC potential which is not as good as SB processors as intel cheap out on TIM.Still ivy i5/i7 do 4.7-4.8GHz mostly on a good board.By the way though it is not something that would concern everyone but at higher clocks the fx8350 eats out much more power than an ivy i5/i7 at same clocks.http://www.tomshardware.com/reviews/build-a-pc-overcloc... does tell that i5 3570k when overclocked is very good and both fx8350 and i5 3570k traded blows in some.


Quotes from that page on toms "When it comes to overclocking though, Intel extends its lead with significantly lower power consumption and much better performance. If we were measuring efficiency, that'd be a home run"
"It'd be a great experiment, and we might even play around with it in the future, but it's clear that Intel's Core i5-3570K remains the better choice for overclockers in this price range."

For the things tested i5 3570k was overall better when both were overclocked.They are far more sensitive to frequency than fx8350 is.So simply talking about clocks won't do.IPC is a thing as well.So the so called "superior clocks" are just not the only thing,it's also important how much they are sensitive to those clocks.


ericjohn004 has explained things quite well.I think you are just not ready to look on other side.He and I were talking about both side of things but you just don't seem to acknowledge what is on other side.I do not support or am loyal to a company.I would pick whatever best for the money is available within what I can spend which is best suited to things I would do.We all want AMD to do well simply becoz it would result in more competition and better priced products from both sides.But that doesn't mean we should not look at the overall picture of the things now.I have made my points on the context,period.



1) What you really said is quoted above. Borderlands 2 is not using all the potential of the FX and still it was only slightly slower than the expensive i7-3770k



Hitman is not using all the potential of the FX, albeit you present it as new, "new" does not mean "parallelized for eight cores".

The point here is that FX is good enough for gaming and you are not going to change that by posting here.

2) It is good that you quote Toms now, but are you aware that the quote that you gave is a bit old? In their cpu gaming hierarchy cart (updated to March 2013). Toms puts the FX near the top and say, in the text, that upgrading from a FX-8350 to an expensive i5-3570K/i7-3770k does not offer overall gaming benefit. You and another user pretend otherwise but precisely both of you avoid Toms advice.

3) As explained before, Idontcare comparison was biased in any possible point. No, 3DS MAX is not taking all FX cores to maximum. Where did you got that exaggeration? In any case the same forum that you link says it clear

Quote:
So in these 2 workloads 8350 is actually on par or faster than 3770K






You pretend that "the i5 3570k was able to keep up with fx8350", still data shows otherwise. Moreover, the same 3DS MAX review says that the new FX chip allowed AMD to beat the entire range of LGA 1155 Intel chips:

Quote:
Ceci permet à AMD de prendre les devants sur toute la gamme LGA 1155.


4) Who said you that everyone will be obtaining a world record at her/his home? The world record shows that the design of the FX chip is superior to that of the Intel chips. A more solid design is the reason why it is so easy to overclock an AMD FX beyond Intel chips.

5) You quote toms overcloking review of the FX-8350, but once again you forget to quote relevant info. That review was made using older software, favourable to Intel chips as the i5. Toms write in the same page that you quote:

Quote:
This will likely change as we fold more heavily-threaded tests into the Marathon, starting this quarter.


Moreover, you also forgot to say that the default Intel builts used stock memory, but the default FX build used under-clocked memory, thus lowering its performance.

The point is: if you force yourself to using older software or one application at once the i5 looks a better option. If you run several applications at once and/or modern optimized software the FX looks better.


Check borderlands 2 here: http://www.xbitlabs.com/images/cpu/fx-8350-8320-6300-43...
GPU used was a GTX680.So it quite clear that these are not any bogus links.

That's a pretty lame comment you made by saying Hitman absolution is not using Fx8350's full potential.
You expect every game out there to scale to eight cores on this date,really?Not every game engine is same,not every development is done on the same way.How many games from 2008-2012 are threaded enough to scale well across 8 cores??

And with quoting those 3DS MAX benchmarks you tried to twist things again in your favor.
Let me correct you:Idontcare tested with 3DS MAX 2013 and in that i7 3770k was the best of the three while
i5 3570k OCed was as good or better than fx8350 OCed,and for your information 3D rendering softwares can easily task all the cores to full.Just checkout what Idoncare had to say about rendering in 3DS MAX2013.There is no exaggeration here.
The 2 pics you showed out of that forum is of 3DS MAX 2011.
Priviewing things in viewport and many other things in 3DS MAX are light threaded which benchmarks never show so again i5 3570k with all round performance was capability is a pretty good choice.It would have been better if you had read that whole thread rather than just picking one thing which was a benchmark on an older version of the software.
And that thing only was quoted since you implied that an OCed fx8350 would be lot better than an OCed i5 3570k.
How many people care about which achieves the highest clock??i5 3570k is able to do at the level of 4.7-4.8 and is more responsive to frequency by design.Most people don't buy an unlocked processor to run it at stock.
In the softwares used in system builder marathon by 2013 OCed performance of i5 3570k was good 20-25% better than fx8350.The gap would reduce if more multi-threaded softwares are used even though toms did a pretty good job by running many kinds of apps which are used by many of us on a daily basis.

Nobody ever said that fx8350 was a bad chip for gaming.It is just in some cases/games it is not able to keep with i5 3570k(already quoted cases like Civilization 5,skyrim etc).In my country the fx8350 setup costs much lesser than an i5 3570k set up,so if I was money-limited and my sole purpose was gaming I would take the fx8350 and invest the rest on a better GPU.But that is not the context here.If I was not money bound and gaming is not important then i would choose whatever best suits for the tasks I do out of both fx8350 and i5 3570k.
fx8350 is a pretty solid value for the price especially for video encoding and some other heavily threaded tasks.It has full instruction set support,not cut any corners like intel with their k series processor like by not having VT-d.
Per core performance is not on par with i5 3570k.What i5 3570k has is good all round performance i.e.good single threaded as well as multi-threaded performance.I don't think any of these chips can be termed overall bad or something.
ericjohn004's point on gaming and how it is tested on some sites is right.Like many sites test on 1280*800 to show CPU side of things but that is not relevant to most.



a b à CPUs
April 2, 2013 6:21:57 PM

Wow I just realized this thread has 4 freaking pages of comments and it has been read almost 10,000 times. I hope some people out there can use all of our bickering, and rage, but also some really quite useful information to learn about the 3570k and FX8350.

This theads going to be like the top 10 searches of "3570k vs. 8350".
!