Sign in with
Sign up | Sign in
Your question
Closed

FX Vs. Core i7: Exploring CPU Bottlenecks And AMD CrossFire

Tags:
Last response: in Reviews comments
Share
January 24, 2013 3:24:21 AM

Quote:
We were hoping that AMD's Piledriver update would break that trend, but even a handful of impressive advancements aren't enough to match the effectiveness of AMD's graphics team. Might Steamroller be the evolutionary step forward needed to unleash the GCN architecture's peak performance?


I disagree. What's needed is even stronger push on the developers to use more than four cores, effectively, not some 100% load on one core and 10% on the other five cores.
January 24, 2013 3:50:09 AM

Great article and very informative. The FX-8350 really held it's own until it came down to Skyrim.

A Bad DayI disagree. What's needed is even stronger push on the developers to use more than four cores, effectively, not some 100% load on one core and 10% on the other five cores.


I thought more cores were for multi-tasking, as in having multiple programs running simultaneously. It would suck to turn on BF3 and everything else running on my PC simply shut down because the CPU is under 100% utilization. How would i be able to play BF3 while streaming/playing some HD content on my TV that's hooked up to my same computer.


Related resources
January 24, 2013 3:58:09 AM

acktionhankGreat article and very informative. The FX-8350 really held it's own until it came down to Skyrim.I thought more cores were for multi-tasking, as in having multiple programs running simultaneously. It would suck to turn on BF3 and everything else running on my PC simply shut down because the CPU is under 100% utilization. How would i be able to play BF3 while streaming/playing some HD content on my TV that's hooked up to my same computer.


single core performance... look up some other benchmarks, where they use itunes to encode things, or when i believe winzip went from single core to multicore, it shows a GREAT difference more cores can do to performance.

the problem is that few games and few programs really scale, sure, pro applications almost always take advantage of whatever you put in them, but consumer, different story.

more cores can offer more multitasking, but they also allow the load to be shifted from one core to all 4 cores and get over all more performance when properly coded.
a c 124 à CPUs
a b À AMD
January 24, 2013 4:03:25 AM

Personally I'd like to see the i5-3570K included in here. It's closer in price to the 8350, but should perform more like the 3770K (as the games are unlikely to use more than 4 threads).
a b à CPUs
January 24, 2013 4:06:13 AM

A Bad DayI disagree. What's needed is even stronger push on the developers to use more than four cores, effectively, not some 100% load on one core and 10% on the other five cores.
I'm calling BS on this one because AMD's "eight cores" are actually four modules, on four front ends, with four FP units. Games have historically been limited by FP units specifically and front ends in general, no? What I'm seeing is that Intel's per-core IPC appears to be a little higher, when two different FOUR "full" CORE processors are compared.
a b à CPUs
a b À AMD
January 24, 2013 4:09:36 AM

There should be an i5 included just so you can have a middle ground.
a c 184 à CPUs
a b À AMD
January 24, 2013 4:09:50 AM

I'm really liking the new logo!
a b à CPUs
a b À AMD
January 24, 2013 4:44:16 AM

like the article.
woulda liked to see how a 3570k does against the fx8350 running the same cfx setup. impo, the price/perf woulda tipped further in favor of intel in configs like this.
lastly, woulda liked some newer games like sleeping dogs, far cry3, max payne 3 in the benches instead of the ol' bf3 single player. i hear bf3 sp doesn't stress cpus that much. may be bf3 skewed the benches in favor of amd as much as skyrim favored intel. :whistle: 
January 24, 2013 5:34:48 AM

all these benchmarks are manipulated. First, there is this site which claims the 7900 series does well even with mid level cpus in gaming scenarios. And now toms claim a high end cpu. There is some propoganda here.
January 24, 2013 5:47:42 AM

It would be nice to see prices for components similar the SMB. Not because I can't look them up, but because the article is very price/performance oriented
a b à CPUs
January 24, 2013 6:01:16 AM

acktionhankGreat article and very informative. The FX-8350 really held it's own until it came down to Skyrim.I thought more cores were for multi-tasking, as in having multiple programs running simultaneously. It would suck to turn on BF3 and everything else running on my PC simply shut down because the CPU is under 100% utilization. How would i be able to play BF3 while streaming/playing some HD content on my TV that's hooked up to my same computer.

Why not just use two computers?
January 24, 2013 6:19:33 AM

The final (biased/unbiased) comparisons are the oddest thing I've ever seen. You practically have to have a degree in statistics to understand them, since for us mere mortals the maximum value is 100%. How the fuck you got 115% of max performance I really do not know. What is your 100%??? In any case. On the value thing I also have to add that saying that the value of the whole system is the most balanced approach is nonsense. You compare individual parts with individual parts. If you care about value then you will hardly get the most expensive case,kbd,mouse,etc... You'll buy the best CPU and gfx card and memory you can afford and stick it in any damned old chassis that you can keep cool. Your "most balanced" approach is a crock. It's like saying that two people have close as the same test scores because the average of the test scores of their respective classes are just about the same. LOL So much of that comparison was meaningless to me. Sorry...
a b à CPUs
January 24, 2013 6:52:32 AM

The_TrutherizerThe final (biased/unbiased) comparisons are the oddest thing I've ever seen. You practically have to have a degree in statistics to understand them, since for us mere mortals the maximum value is 100%. How the *** you got 115% of max performance I really do not know. What is your 100%??? In any case. On the value thing I also have to add that saying that the value of the whole system is the most balanced approach is nonsense. You compare individual parts with individual parts. If you care about value then you will hardly get the most expensive case,kbd,mouse,etc... You'll buy the best CPU and gfx card and memory you can afford and stick it in any damned old chassis that you can keep cool. Your "most balanced" approach is a crock. It's like saying that two people have close as the same test scores because the average of the test scores of their respective classes are just about the same. LOL So much of that comparison was meaningless to me. Sorry...
OK, for mere mortals: When you make this type of calculation, the average is 100%. You have to subtract 100% in order to get a 0% average. In this case, the average was "average of all systems". Think of it like IQ (where the average also happens to be 100).

If you could buy $4 RAM instead of $40 RAM, but the $4 RAM made your system 50% slower, would you buy it? No, because it would make your $1000 PC perform like a $500 PC.

You can only do per-component value when you're only comparing one component. In this case, the graphics cards and CPUs were being tested as a pairing (just like the title says).
January 24, 2013 8:15:17 AM

The i5-3570 is about $100 cheaper and would play those benchmarked games pretty much just as fast as the i7-3770...
a b à CPUs
January 24, 2013 8:26:49 AM

Thanks for the article! I just wished all games took advantage of AMD CPU(s) threads. Hopefully AMD will get better with SteamRoller and improve it's power consumption and IPC.

Again, I enjoy reading the article. Get ready for b!tching by fanboies.... Tom. ;) 
January 24, 2013 8:28:59 AM

esreverThere should be an i5 included just so you can have a middle ground.


No need. My sister's FX 8350 kicks my 3570k's ass at 4.2 ghz consistently in most benchmarks. We both run GTX 480's
January 24, 2013 8:30:38 AM

belardoThe i5-3570 is about $100 cheaper and would play those benchmarked games pretty much just as fast as the i7-3770...


what world do you live in? I payed 200 euro for my i5 3570k while my sister's 8350 cost ~ 160 and gets better performance.
Anonymous
a b à CPUs
January 24, 2013 8:31:17 AM

If the AMD FX 8350 isn't fully loaded for games ... why do you use them to compare? Because it's the top CPU in AMD's portfolio?
Why don't use AMD FX 6x00? They are cheaper, almost 60€ in my country. You have compared AMD FX 4x00 already, but i don't see any review or article using a FX6x00 and i think it's the sweet spot for an all-in-one PC (game and work, with 8GB at least of RAM).

Sorry for my english.
Anonymous
a b à CPUs
January 24, 2013 8:32:18 AM

You must remember Guys that the Tech in AMD's now processors don't support the Old Optimisations 3DNOW! ETC so with certain coding techniques that still use the older stuff AMD suffers. Note that with Newer Applications, AMD Especially with OPENCL can simply whoop Intel out of the Park.

Your processor is only as good as the Programming that supports it, and Intel pays developers to use code that supports it and that is missing on AMD's architecture.
a b à CPUs
a b À AMD
January 24, 2013 9:27:19 AM

Compare a $200 AMD processor with a $330 Intel processor ?

Can you run the whole test again with a $200 intel quadcore

and ditch the old DX 9 game engines , too?
January 24, 2013 10:09:51 AM

Outlander_04 said:
Compare a $200 AMD processor with a $330 Intel processor ?

Can you run the whole test again with a $200 intel quadcore

and ditch the old DX 9 game engines , too?


Nah, it's totally fair.


Like comparing a top of the line Mercedes with a top of the line Ford. They're both top of the line so it's fair to compare even though they're in totally different price brackets :p 

(DISCLAIMER: I am in no way saying AMD processors are "Fords".)
January 24, 2013 10:18:56 AM

Quote:
Since an evenly-spread 20 FPS rate would consist of 200 50-ms frames, we’re using 50 ms as the cut-off for actual playability in today’s analysis.


Hello Mr. Soderstrom,
I always thought of FPS as evenly distributed throughout the second, so thanks for the initial example of 91FPS. Should the calculation in your second example be "20" 50-ms frames? 20 x 50ms = 1000ms = 1 sec...
a b à CPUs
January 24, 2013 10:41:19 AM

Outlander_04Compare a $200 AMD processor with a $330 Intel processor ?Can you run the whole test again with a $200 intel quadcore and ditch the old DX 9 game engines , too?
AMD sets the standard for the elite gaming market BY MAKING THE GRAPHICS CARDS in this test, and you want someone to gimp the CPU? What gamer would do that? At $800 for graphics, these cards deserve all the performance you can throw at them. If anything else was needed, it would have been a FASTER AMD processor that's worthy of this $800 graphics array.

nezzymightyHello Mr. Soderstrom,I always thought of FPS as evenly distributed throughout the second, so thanks for the initial example of 91FPS. Should the calculation in your second example be "20" 50-ms frames? 20 x 50ms = 1000ms = 1 sec...
Thanks for finding that typo! I was so certain that it originally had said 20 that I searched the original document, only to find the typo in the original document as well :( 
a b à CPUs
January 24, 2013 11:19:08 AM

CrashmanI'm calling BS on this one because AMD's "eight cores" are actually four modules, on four front ends, with four FP units. Games have historically been limited by FP units specifically and front ends in general, no? What I'm seeing is that Intel's per-core IPC appears to be a little higher, when two different FOUR "full" CORE processors are compared.


They are actual 8 cores
January 24, 2013 11:19:31 AM

CrashmanI'm calling BS on this one because AMD's "eight cores" are actually four modules, on four front ends, with four FP units. Games have historically been limited by FP units specifically and front ends in general, no? What I'm seeing is that Intel's per-core IPC appears to be a little higher, when two different FOUR "full" CORE processors are compared.

Depending on the AI pathfinding algorythm, the code can use only integer math. With a CPU that has so many integer cores, the pathfinding can be more complex, allowing for some new ways of AI behaviour.
Besides, I still wonder why no one ever thought of developing an engine based on integers, not on floating point, and abstracting it to developers and users.
January 24, 2013 11:30:40 AM

Thanks for the review. I was curious when the sub $200 cpu battle review was coming out (as mentioned in the best cpu/$.)
January 24, 2013 11:31:37 AM

It does not matter that cards can drive 200fps @2560x1600? I think someone forgot that 4k will be in your average conumers homes in 4 years.... (2017, some say 2015) and 8k is not far behind that, maybe 2 more years......

Yeah these are over kill now, but they are not even capable of driving the displays we will be using shortly.
January 24, 2013 11:44:45 AM

BigMack70lol CES has people stupidly optimistic about 4kNo way is that going to be in the "average consumer's" home in 4 years. Not a snowball's chance in hell of that happening. In 10 years? Maybe. In 4?

Agreed, how many people actually have a 1440 set up. There are a lot of people still playing on 1600x900 and 1680x1050, let alone 1080. We still have a while before even the average enthusiast will have 4k monitors.
January 24, 2013 11:46:03 AM

Nice article, though i'll echo the others about throwing in a i5 there too, since you ended up with the value analysis anyway.

And of course, the 4C/4T vs 4C/8T vs 4 module/8T comparison, which i've been dying to see, especially for newer games.
January 24, 2013 11:56:28 AM

The vast majority of people do not build ultra high end pc's with $800 worth of crossfire / sli graphics cards. The ones that do probably will not care about a measley $180 because they obviously want to build the highest end pc possible.

AMD's real value is for every one else. Those users that would rather take the $180 saved by using an AMD processor and contribute that money to a better graphics card. Most people build their pc's with a specific budget in mind. Therefore they look at it this way:

1. I can have an Intel Processor with a Radeon 7770 for x numbers of dollars or I can go with an AMD processor with a Radeon 7850 for the same amount of money. Which one is going to give me the best bang for the buck in the programs I want to run?

Additionally, I will be very curious how this article may change once the next generation of gaming consoles are introduced. From all early indications, it looks like both the PS4 and Xbox 720 will utilize 8 core cpu's at around 1.6ghz. To push games to the limits, developers will be multi-threading their games for all 8 cores. My guess is that the GPU will be the real bottleneck then no matter what processor is used. This might just change AMD's perceived value.
January 24, 2013 11:58:48 AM

BigMack70lol CES has people stupidly optimistic about 4kNo way is that going to be in the "average consumer's" home in 4 years. Not a snowball's chance in hell of that happening. In 10 years? Maybe. In 4?


4k will be here sooner than you think. It will not be for average consumer for sure. Just look at the 30" IPS they have been around for decade and maybe one in million PC user has it. 4k will be the new enthusiast's monitor. If it sells for $2k or less I 'd buy it tomorrow lol.

At 4k resolutions I believe there is no need for AAA or MSAA and tehrefore 8970/GTX880 or 9970/GTX980 will be able to handle it.
January 24, 2013 12:07:58 PM

For 99% of gamers, this article doesn't really mean anything.
Any AMD FX, Intel i5/i7 or even a high end Phenom II or Core 2 Quad are all fast enough for most gaming needs.
The graphics card is a lot more important, and very few people will be using a 7970.

January 24, 2013 12:41:55 PM

CrashmanAMD sets the standard for the elite gaming market BY MAKING THE GRAPHICS CARDS in this test, and you want someone to gimp the CPU? What gamer would do that? At $800 for graphics, these cards deserve all the performance you can throw at them. If anything else was needed, it would have been a FASTER AMD processor that's worthy of this $800 graphics array.

Half of the games tested weren't very CPU (much less utilized hyperthreading) bound and a i5-3570k would have destroyed the value comparison.
January 24, 2013 12:52:20 PM

salgado18Depending on the AI pathfinding algorythm, the code can use only integer math. With a CPU that has so many integer cores, the pathfinding can be more complex, allowing for some new ways of AI behaviour.Besides, I still wonder why no one ever thought of developing an engine based on integers, not on floating point, and abstracting it to developers and users.

That would never work. FP datatypes have around 10^300 times the range with constant precision, while an integer's significant digits end at the radix. Most trig functions have a domain and/or range the is less than one. All of the workarounds required would make the program much slower (not to mention impossible to understand) than using FP logic.
January 24, 2013 1:01:00 PM

AMD FX Processors are made different for the sake of cooling off their CPUs. Multiple cores on a 1 core is called cluster cores. AMD squeezed 2 cores on each processor instead of putting it directly on the bus (like Phenoms 1 + 2). So when you look at your Window Experience Index and it shows the number of cores remember I said. Example: FX-8350.......4 clusters cores times 2 per cluster equals 8 cores total.
January 24, 2013 1:05:38 PM

BigMack70lol CES has people stupidly optimistic about 4kNo way is that going to be in the "average consumer's" home in 4 years. Not a snowball's chance in hell of that happening. In 10 years? Maybe. In 4?


i think saying ten years is optimistic , look how long standard def hung around 60 almost 70 years now i know plenty people still using standard CRT TV's.

besides 4k resolution is only good for super enormous displays , (like 50+) and i just don't see 50 inch tv's getting pushed down to "every man's" price range , taht being 300-400 bucks
January 24, 2013 1:17:00 PM

FX-4300= 2 cluster cores x 2 native cores per cluster = 4 cores
FX-6300= 3 x2 = 6 cores
forget the assumptions.....AMD rid of 2 monikers cause they figure it is pointless to bring it up and all of the AMD brand will have it. "64-bit" and "Black Edition or BE"
a b à CPUs
January 24, 2013 1:28:58 PM

Great article, I enjoyed the read quite a bit. I especially like the fact your benches included ultra widescreen res. Hardware, including monitors, has hit a price point where if your a serious PC gamer you should be rocking 3, at least 1080p, screens.

Kudos to the 8350 as well. The "everything else is free" scenario exists for many people who already have an AM3+ system. I just went x6 to 8320 and could not be happier with my $180 upgrade.
a c 147 à CPUs
a b À AMD
January 24, 2013 1:43:41 PM

about what i expected.

we all know intel has better single core performance then AMD. About the only thing noteable in this whole thing was how well the fx8350 did. I didn't expect it to go toe to toe with an intel chip in gaming.

It seems like the high end cpus for both companies aren't the gaming bottleneck (unless you're playing with 3 monitors)... the GPUs are. If anything this article made clear it's that as long as you're playing on one monitor at 1080p resolution either CPU is more then enough CPU coupled with a high end GPU.
a b à CPUs
a b À AMD
January 24, 2013 1:44:33 PM

razzb3dNo need. My sister's FX 8350 kicks my 3570k's ass at 4.2 ghz consistently in most benchmarks. We both run GTX 480's

Assuming the same graphics setup:
Benchmarks, like 3dmark11 are far different, favor more cores, but games rarely take advantage of those cores.
Also, there are many other parts of your system, including install issues that can make the systems vary.
a b à CPUs
January 24, 2013 1:54:21 PM

A CPU bottleneck test and no starcraft 2 ? dosent seem like a serious test to me 8(.
January 24, 2013 1:59:33 PM

Great article crash, it shows in some games little difference, in some noticeable, and playing experience on either chip is good for the most part
TY
January 24, 2013 2:13:37 PM

So basically the AMD gets 60 fps or more for gaming and I need 140 fps because I can afford the expensive monitoring system and all. Bottom line AMD= Low cost basic performance Intel= High cost great performance. I see your graphs and all but at the end of the day I spend $140 less buying the AMD processor for a decent gaming performance. I get what your trying to say but end of the day its not always dollar to performance its what can I afford to a game at the needed level. Some of us have house payment and car payments and a family to raise. Its not all about the shiny jewel in my pc that lets me put a wall full of monitors up and Play BF3. I get it but people need to get you cant compare apples to prime rib. An apple will satisfy your appetite but a prime rib will do it with a nicer flavor in my book. But thanks for the input and all your hard work and benchmarks you all provide.
January 24, 2013 2:15:56 PM

HaserathWhy not just use two computers?


Give me a couple hundred dollars and i'll build myself one.
a b à CPUs
January 24, 2013 2:25:23 PM

Overall it was a good article, and I think you have shown once again that AMD is still not yet !/$ competitive with Intel, but it would seem to me that they are at least making some headway.

Anywho, my only complaint is that this is not how people build systems. Yes, if you want 'the best of the best' for a respective platform, then it is helpful to show that Intel wins (would be more helpful to know how nVidia cards react... though I imagine it would be much of the same story). But 90+% of buyers out there have a specific, or near specific, dollar point in mind when purchasing a total system, and so the question then becomes what is best at what price points.

In other words, I think the real question is not one of what platform is best, but as GPUs have vastly more influence on performance than the CPU, what level of CPU is needed for today's games?
So if I have a set budget of (lets just say) $1000, am I better off getting an i7 or 8350 with a single monster GPU? or would I actuially get better performance from a much slower/cheaper i3 or i5 (or AMD equivilant), but paired with perhaps 2 upper midrange cards?
January 24, 2013 2:39:57 PM

Please test a 3570k with the same GPU setup and the same games. At least there would be price parity with the cpus. I own a 8350 and I own 2 2500k rigs. My $$$ is on the 3570k winning. However, at least on a true cost basis, the comparison would nearly be equal.
January 24, 2013 2:41:36 PM

Amd beat Intel in multitasking ,cores , now games. I think Amd processors are value for money but Intel want more money for 6core processor .
January 24, 2013 3:13:14 PM

agent88 said:
The vast majority of people do not build ultra high end pc's with $800 worth of crossfire / sli graphics cards. The ones that do probably will not care about a measley $180 because they obviously want to build the highest end pc possible.

AMD's real value is for every one else. Those users that would rather take the $180 saved by using an AMD processor and contribute that money to a better graphics card. Most people build their pc's with a specific budget in mind. Therefore they look at it this way:

1. I can have an Intel Processor with a Radeon 7770 for x numbers of dollars or I can go with an AMD processor with a Radeon 7850 for the same amount of money. Which one is going to give me the best bang for the buck in the programs I want to run?


That seems like a rather convenient example. Don't forget to take power efficiency into account. To use an equally convenient example, if you save $30 on the initial purchase only to spend an extra $60 on your electricity bill over the next 3 years, you've gained nothing. (Obviously usage habits and power costs vary from person to person, locale to locale.)

In this case, the Intel solution costs significantly more than the AMD solution (about $100 more, off the top of my head), so you have a point. But as others have pointed out, the Intel Core i5 probably provides a more apt comparison. It's easy to say that AMD is for cost-conscious consumers and Intel is for high-end consumers, but the truth is that there isn't a hard-and-fast rule. Intel has a whole range of powerful and power-efficient CPUs ranging in price from about $120 to $1,000. (For the sake of argument, I'm ignoring everything below Core i3.)

AMD wins in the extreme-budget segment of the market, I think, on the strength of their superior integrated graphics' solutions. In the low-to-mid range, Intel and AMD trade blows; whether one or the other represents the better buy will depend on the consumer's particular needs.

agent88 said:
Additionally, I will be very curious how this article may change once the next generation of gaming consoles are introduced. From all early indications, it looks like both the PS4 and Xbox 720 will utilize 8 core cpu's at around 1.6ghz. To push games to the limits, developers will be multi-threading their games for all 8 cores. My guess is that the GPU will be the real bottleneck then no matter what processor is used. This might just change AMD's perceived value.

I'd like to think that all of that is true, but there are limits to how well certain tasks can be multi-threaded. Sometimes, step 1 must proceed to step 2, and so on. We'll see.
January 24, 2013 3:16:15 PM

could tomshardware or if anyone knows of any other websites who have done this please recommend benchmarks comparing the phenom II series with the bulldozer and piledriver fx series (ideally with plenty of gaming benchmarks)
!