Sign in with
Sign up | Sign in
Your question
Solved

Will an AMD FX-6350 bottleneck an HD Radeon 7870?

Last response: in CPUs
Share
July 18, 2013 3:18:38 PM

I currently have the QuadCore AMD Phenom II X4 Black Edition 965 in my system along with the MSI HD Radeon 7870 OC. Within the next few weeks, I plan on getting the FX 6350 to put in my system because I feel that my Phenom is bottlenecking my GPU already. Will the 6350 be an improvement from the Phenom? And will I get a higher performance from my PC? Thanks in advance!

Best solution

July 18, 2013 3:24:29 PM

I dont see how a x4 965 is bottlenecking your GPU you might see from frame advantage going to the 6300 which is a recommended CPU in toms top CPU's for the money. Or you could spend like 10 bucks more and get an 8320
Share
July 18, 2013 3:29:55 PM

mourice12 said:
I dont see how a x4 965 is bottlenecking your GPU you might see from frame advantage going to the 6300 which is a recommended CPU in toms top CPU's for the money. Or you could spend like 10 bucks more and get an 8320


Well, to be honest, I don't even know if a higher clock speed or more cores is better. The 6350 has 6 cores at 3.9 GHz while the 8320 has 8 cores at 3.5 GHz. But I don't do much multitasking, so I don't think I need 8 cores. I just want to have higher FPS and better gaming performance so from my knowledge, a higher clock speed can give me that. I may be wrong, but still, I think the FX-6350 will be better for me.
m
0
l
Related resources
July 18, 2013 3:40:18 PM

its up to you man, but idk id get the best for my buck
m
0
l
a b à CPUs
July 18, 2013 4:33:53 PM

you can easily OC a 8320 to 4GHz with a decent cooler
m
0
l
July 18, 2013 4:37:32 PM

yea very easily. most ppl get anywhere from 4.5-4.8
m
0
l
July 21, 2013 6:10:16 PM

But what is the difference between the FX-6300 and the FX-6350? I hear that the 6300 uses less watage, but the 6350 performs slightly better. Is this true?
m
0
l
July 22, 2013 6:09:20 AM

the only difference between the two is that the 6350 comes clocked at a higher speed and cost 20 dollars more. Thats the only difference honestly. 3.9Ghz on the 6350 and 3.5ghz on the 6300.
m
0
l
a b à CPUs
July 22, 2013 6:10:01 AM

I think the 6300 and 6350 are the same chip with just a different default clock speed. You really should get the 8320 because what a lot of people don't realize is that these new AMD chips don't really have as many cores as they claim. The 8320 is more like a quad core with hyper threading and the 6300 is like a triple core with hyper threading, only a little different because they have 8 and 6 integer clusters. Many programs will report that the 8320 has 4 processors and 8 logical processors.
m
0
l
a b à CPUs
August 16, 2013 9:03:24 AM

Ok so the best bang/buck would be the 6300, the 6350 is just higher binned and higher clocked with higher default VCore. I also do not think that your CPU is what the bottleneck in your system is, it is usually RAM or HDD. 8320 vs 8350 is the same deal just with binning, clocks, and VCore but to be warned 8320 is last and 8350 is 3rd out of the 4 processors that use this silicon, the FX 9XXX series get the highest quality silicon.
m
0
l
August 16, 2013 1:18:05 PM

Bryan McCumber said:
I currently have the QuadCore AMD Phenom II X4 Black Edition 965 in my system along with the MSI HD Radeon 7870 OC. Within the next few weeks, I plan on getting the FX 6350 to put in my system because I feel that my Phenom is bottlenecking my GPU already. Will the 6350 be an improvement from the Phenom? And will I get a higher performance from my PC? Thanks in advance!


Get the Fx 8320 more cores IS better i can source things to prove it.

Also the new ps4 will be using apu amd 8 core there for future gaming will be moving in that defection I just build my friend a system with the his Iceq 7870 ghz edition and the fx 8230 and the machine is awesome all stock speeds and going to be amazing when i overclock it once he gets some coolers it.


It's just the better way to go if you're already trying for great gaming graphic.
m
0
l
a b à CPUs
August 26, 2013 12:23:45 AM

SilverfoxHound said:

Get the Fx 8320 more cores IS better i can source things to prove it.

Also the new ps4 will be using apu amd 8 core there for future gaming will be moving in that defection I just build my friend a system with the his Iceq 7870 ghz edition and the fx 8230 and the machine is awesome all stock speeds and going to be amazing when i overclock it once he gets some coolers it.


It's just the better way to go if you're already trying for great gaming graphic.


The PS4 APU uses 8 Cores to be able to run background programs and other stuff concurrently with games. Now with 8 cores their is an obvious advantage but is the extra cost worth the performance bump for him. If he is going for the HD 7870 over the HD 7950 or GTX 660Ti there is an obvious value oriented enthusiast build. I have a FX 6300 overclocked to 4.3 GHz with a Radeon HD 7870 running at 1050 MHz and have no problems once I got my RAM to 1866 and switched to an SSD. Even without the RAM overclock and SSD games were running at 60 FPS and smoothly, I just overclocked and got the SSD to get faster loading times and a safer drive for my OS. 6 cores/ 3 modules is the sweet spot for gaming now and with OS advancements in handling more cores the 6 cores will be fine for a good while and will last you just as long if not longer than the Phenom.
m
0
l
a c 210 à CPUs
a b À AMD
August 26, 2013 11:10:13 AM

Bryan McCumber said:
I currently have the QuadCore AMD Phenom II X4 Black Edition 965 in my system along with the MSI HD Radeon 7870 OC. Within the next few weeks, I plan on getting the FX 6350 to put in my system because I feel that my Phenom is bottlenecking my GPU already. Will the 6350 be an improvement from the Phenom? And will I get a higher performance from my PC? Thanks in advance!


The 6350 would be an improvement, Tom's Hardware has an article about the K10 architecture versus several of the FX line CPUs, and the 6350 was the best of the bunch they tested.

You would definitely not bottleneck your GPU, and you would likely see some good improvement in several games.
m
0
l
a c 210 à CPUs
a b À AMD
August 26, 2013 11:11:54 AM

swilczak said:
I think the 6300 and 6350 are the same chip with just a different default clock speed. You really should get the 8320 because what a lot of people don't realize is that these new AMD chips don't really have as many cores as they claim. The 8320 is more like a quad core with hyper threading and the 6300 is like a triple core with hyper threading, only a little different because they have 8 and 6 integer clusters. Many programs will report that the 8320 has 4 processors and 8 logical processors.


This is inaccurate...the 8320/8350 are 8 core CPUs, with 4 flex FPUs. AMD does not use HTT in any manner for any of their architectures.
m
0
l
a b à CPUs
August 26, 2013 12:13:53 PM

8350rocks said:
swilczak said:
I think the 6300 and 6350 are the same chip with just a different default clock speed. You really should get the 8320 because what a lot of people don't realize is that these new AMD chips don't really have as many cores as they claim. The 8320 is more like a quad core with hyper threading and the 6300 is like a triple core with hyper threading, only a little different because they have 8 and 6 integer clusters. Many programs will report that the 8320 has 4 processors and 8 logical processors.


This is inaccurate...the 8320/8350 are 8 core CPUs, with 4 flex FPUs. AMD does not use HTT in any manner for any of their architectures.


I have more than one program that reports my 8320 as being a quad core with hyperthreading. I never said it was hyperthreading, I said it's like it but different. AMD claims that they have eight cores but they are really a cheap half core design, 8 integer clusters, 4 modules, the 8 "cores" share cache in the 4 modules. A true 8 core would have it's own cache for each core.
m
0
l
a c 210 à CPUs
a b À AMD
August 26, 2013 12:17:15 PM

swilczak said:
8350rocks said:
swilczak said:
I think the 6300 and 6350 are the same chip with just a different default clock speed. You really should get the 8320 because what a lot of people don't realize is that these new AMD chips don't really have as many cores as they claim. The 8320 is more like a quad core with hyper threading and the 6300 is like a triple core with hyper threading, only a little different because they have 8 and 6 integer clusters. Many programs will report that the 8320 has 4 processors and 8 logical processors.


This is inaccurate...the 8320/8350 are 8 core CPUs, with 4 flex FPUs. AMD does not use HTT in any manner for any of their architectures.


I have more than one program that reports my 8320 as being a quad core with hyperthreading. I never said it was hyperthreading, I said it's like it but different. AMD claims that they have eight cores but they are really a cheap half core design, 8 integer clusters, 4 modules, the 8 "cores" share cache in the 4 modules. A true 8 core would have it's own cache for each core.


The cache is double sized so the cores don't lose performance...
m
0
l
a b à CPUs
August 26, 2013 12:50:28 PM

8350rocks said:
swilczak said:
8350rocks said:
swilczak said:
I think the 6300 and 6350 are the same chip with just a different default clock speed. You really should get the 8320 because what a lot of people don't realize is that these new AMD chips don't really have as many cores as they claim. The 8320 is more like a quad core with hyper threading and the 6300 is like a triple core with hyper threading, only a little different because they have 8 and 6 integer clusters. Many programs will report that the 8320 has 4 processors and 8 logical processors.


This is inaccurate...the 8320/8350 are 8 core CPUs, with 4 flex FPUs. AMD does not use HTT in any manner for any of their architectures.


I have more than one program that reports my 8320 as being a quad core with hyperthreading. I never said it was hyperthreading, I said it's like it but different. AMD claims that they have eight cores but they are really a cheap half core design, 8 integer clusters, 4 modules, the 8 "cores" share cache in the 4 modules. A true 8 core would have it's own cache for each core.


The cache is double sized so the cores don't lose performance...


I'm sure that's how AMD wanted it to work
m
0
l
a b à CPUs
August 26, 2013 1:59:29 PM

swilczak said:

I'm sure that's how AMD wanted it to work


AMD made a new architecture based on more ALUs and less FPUs to deliver more real-world performance where it counts while still having floating point calculations available to those who need it. When launching the FX processors they clearly said that they were re-inventing what a core should be, so not it is not Hyperthreading at all. Some programs will report each module as a core instead of each core. 1 module contains two cores so it may seem that way but they are all physical cores with no special thread handling.
m
0
l
a b à CPUs
August 26, 2013 2:42:36 PM

rishiswaz said:
swilczak said:

I'm sure that's how AMD wanted it to work


AMD made a new architecture based on more ALUs and less FPUs to deliver more real-world performance where it counts while still having floating point calculations available to those who need it. When launching the FX processors they clearly said that they were re-inventing what a core should be, so not it is not Hyperthreading at all. Some programs will report each module as a core instead of each core. 1 module contains two cores so it may seem that way but they are all physical cores with no special thread handling.


The Tom's Hardware website commented that the lower-than-expected performance in multi-threaded workloads may be because of the way Windows 7 currently schedules threads to the cores. They point out that "if Windows were able to utilize an FX-8150's four modules first, and then backfill each module's second core, it'd maximize performance with up to four threads running concurrently." This is similar to what happens on Intel CPUs with HyperThreading – Windows 7 "schedules to physical cores before utilizing logical (HyperThreaded) cores

What it is and how it actually works in windows are two different things. It works like hyperthreading just like I said.
http://en.wikipedia.org/wiki/Bulldozer_(microarchitecture)
m
0
l
a c 210 à CPUs
a b À AMD
August 26, 2013 2:46:30 PM

swilczak said:
rishiswaz said:
swilczak said:

I'm sure that's how AMD wanted it to work


AMD made a new architecture based on more ALUs and less FPUs to deliver more real-world performance where it counts while still having floating point calculations available to those who need it. When launching the FX processors they clearly said that they were re-inventing what a core should be, so not it is not Hyperthreading at all. Some programs will report each module as a core instead of each core. 1 module contains two cores so it may seem that way but they are all physical cores with no special thread handling.


The Tom's Hardware website commented that the lower-than-expected performance in multi-threaded workloads may be because of the way Windows 7 currently schedules threads to the cores. They point out that "if Windows were able to utilize an FX-8150's four modules first, and then backfill each module's second core, it'd maximize performance with up to four threads running concurrently." This is similar to what happens on Intel CPUs with HyperThreading – Windows 7 "schedules to physical cores before utilizing logical (HyperThreaded) cores

What it is and how it actually works in windows are two different things. It works like hyperthreading just like I said.
http://en.wikipedia.org/wiki/Bulldozer_(microarchitecture)


You do realize the only shared cache is L3 right? Each core otherwise has separate L1 and L2 cache, which are far more crucial...

Additionally...for scheduling purposes it would work better that way...however, windows scheduler is typically poor compared to other things. If you look at the PD performance in Linux you can see the difference between a "efficient" and "less efficient" scheduler.
m
0
l
a b à CPUs
August 26, 2013 3:25:15 PM

8350rocks said:
swilczak said:
rishiswaz said:
swilczak said:

I'm sure that's how AMD wanted it to work


AMD made a new architecture based on more ALUs and less FPUs to deliver more real-world performance where it counts while still having floating point calculations available to those who need it. When launching the FX processors they clearly said that they were re-inventing what a core should be, so not it is not Hyperthreading at all. Some programs will report each module as a core instead of each core. 1 module contains two cores so it may seem that way but they are all physical cores with no special thread handling.


The Tom's Hardware website commented that the lower-than-expected performance in multi-threaded workloads may be because of the way Windows 7 currently schedules threads to the cores. They point out that "if Windows were able to utilize an FX-8150's four modules first, and then backfill each module's second core, it'd maximize performance with up to four threads running concurrently." This is similar to what happens on Intel CPUs with HyperThreading – Windows 7 "schedules to physical cores before utilizing logical (HyperThreaded) cores

What it is and how it actually works in windows are two different things. It works like hyperthreading just like I said.
http://en.wikipedia.org/wiki/Bulldozer_(microarchitecture)


You do realize the only shared cache is L3 right? Each core otherwise has separate L1 and L2 cache, which are far more crucial...

Additionally...for scheduling purposes it would work better that way...however, windows scheduler is typically poor compared to other things. If you look at the PD performance in Linux you can see the difference between a "efficient" and "less efficient" scheduler.


Yes I realize that, but since only 5% of computer users use Linux.....need I say more? I'm tired of arguing about this, I showed proof from tomshardware quoted on wiki, so I'm done. You can sit in your moms basement all day and try to prove me wrong.
m
0
l
a c 210 à CPUs
a b À AMD
August 26, 2013 3:50:16 PM

swilczak said:
8350rocks said:
swilczak said:
rishiswaz said:
swilczak said:

I'm sure that's how AMD wanted it to work


AMD made a new architecture based on more ALUs and less FPUs to deliver more real-world performance where it counts while still having floating point calculations available to those who need it. When launching the FX processors they clearly said that they were re-inventing what a core should be, so not it is not Hyperthreading at all. Some programs will report each module as a core instead of each core. 1 module contains two cores so it may seem that way but they are all physical cores with no special thread handling.


The Tom's Hardware website commented that the lower-than-expected performance in multi-threaded workloads may be because of the way Windows 7 currently schedules threads to the cores. They point out that "if Windows were able to utilize an FX-8150's four modules first, and then backfill each module's second core, it'd maximize performance with up to four threads running concurrently." This is similar to what happens on Intel CPUs with HyperThreading – Windows 7 "schedules to physical cores before utilizing logical (HyperThreaded) cores

What it is and how it actually works in windows are two different things. It works like hyperthreading just like I said.
http://en.wikipedia.org/wiki/Bulldozer_(microarchitecture)


You do realize the only shared cache is L3 right? Each core otherwise has separate L1 and L2 cache, which are far more crucial...

Additionally...for scheduling purposes it would work better that way...however, windows scheduler is typically poor compared to other things. If you look at the PD performance in Linux you can see the difference between a "efficient" and "less efficient" scheduler.


Yes I realize that, but since only 5% of computer users use Linux.....need I say more? I'm tired of arguing about this, I showed proof from tomshardware quoted on wiki, so I'm done. You can sit in your moms basement all day and try to prove me wrong.


So, let me get this straight...

You're basing your argument on windows scheduler being a poor scheduler. Because it's a poor scheduler, if it handled the scheduling more efficiently...like it does for a HTT enabled Intel CPU...it may improve performance. Since that may improve performance, you are insinuating that the modular design of AMD's CPU architecture is similar to HTT...? All this, even though HTT has no physical resources outside of an extra register stack, and AMD architecture has a physical core with 2 ALUs and dedicated L1 and L2 cache??

Hmm...then you justify this by quoting a misinformed writer, trying to make an off the cuff comparison, that was quoted by Wikipedia?

That argument holds as much water as a minnow bucket...(if you're not a fisherman, they have holes in them from top to bottom)
m
0
l
a b à CPUs
August 26, 2013 4:08:02 PM

8350rocks said:
swilczak said:
8350rocks said:
swilczak said:
rishiswaz said:
swilczak said:

I'm sure that's how AMD wanted it to work


AMD made a new architecture based on more ALUs and less FPUs to deliver more real-world performance where it counts while still having floating point calculations available to those who need it. When launching the FX processors they clearly said that they were re-inventing what a core should be, so not it is not Hyperthreading at all. Some programs will report each module as a core instead of each core. 1 module contains two cores so it may seem that way but they are all physical cores with no special thread handling.


The Tom's Hardware website commented that the lower-than-expected performance in multi-threaded workloads may be because of the way Windows 7 currently schedules threads to the cores. They point out that "if Windows were able to utilize an FX-8150's four modules first, and then backfill each module's second core, it'd maximize performance with up to four threads running concurrently." This is similar to what happens on Intel CPUs with HyperThreading – Windows 7 "schedules to physical cores before utilizing logical (HyperThreaded) cores

What it is and how it actually works in windows are two different things. It works like hyperthreading just like I said.
http://en.wikipedia.org/wiki/Bulldozer_(microarchitecture)


You do realize the only shared cache is L3 right? Each core otherwise has separate L1 and L2 cache, which are far more crucial...

Additionally...for scheduling purposes it would work better that way...however, windows scheduler is typically poor compared to other things. If you look at the PD performance in Linux you can see the difference between a "efficient" and "less efficient" scheduler.


Yes I realize that, but since only 5% of computer users use Linux.....need I say more? I'm tired of arguing about this, I showed proof from tomshardware quoted on wiki, so I'm done. You can sit in your moms basement all day and try to prove me wrong.


So, let me get this straight...

You're basing your argument on windows scheduler being a poor scheduler. Because it's a poor scheduler, if it handled the scheduling more efficiently...like it does for a HTT enabled Intel CPU...it may improve performance. Since that may improve performance, you are insinuating that the modular design of AMD's CPU architecture is similar to HTT...? All this, even though HTT has no physical resources outside of an extra register stack, and AMD architecture has a physical core with 2 ALUs and dedicated L1 and L2 cache??

Hmm...then you justify this by quoting a misinformed writer, trying to make an off the cuff comparison, that was quoted by Wikipedia?

That argument holds as much water as a minnow bucket...(if you're not a fisherman, they have holes in them from top to bottom)

Yes, AMD was trying to follow in Intels footsteps like they always do and create something that was hyperthreading but better and they failed. The only area that AMD CPU's work better than intels is integer operations because of the 8 integer clusters. I've done my own tests so I know. If you look at the bottom of the Wikipedia page it shows a link to the tomshardware article that stated this. It's not like somebody just edited the page with some nonsense. You can write whatever you want from hear on out on this thread but I assure you that I won't read it because you are a troll and I see you on every CPU thread trying to argue with people because you have no life and you think you know everything.
m
0
l
a c 210 à CPUs
a b À AMD
August 26, 2013 4:24:34 PM

swilczak said:
8350rocks said:
swilczak said:
8350rocks said:
swilczak said:
rishiswaz said:
swilczak said:

I'm sure that's how AMD wanted it to work


AMD made a new architecture based on more ALUs and less FPUs to deliver more real-world performance where it counts while still having floating point calculations available to those who need it. When launching the FX processors they clearly said that they were re-inventing what a core should be, so not it is not Hyperthreading at all. Some programs will report each module as a core instead of each core. 1 module contains two cores so it may seem that way but they are all physical cores with no special thread handling.


The Tom's Hardware website commented that the lower-than-expected performance in multi-threaded workloads may be because of the way Windows 7 currently schedules threads to the cores. They point out that "if Windows were able to utilize an FX-8150's four modules first, and then backfill each module's second core, it'd maximize performance with up to four threads running concurrently." This is similar to what happens on Intel CPUs with HyperThreading – Windows 7 "schedules to physical cores before utilizing logical (HyperThreaded) cores

What it is and how it actually works in windows are two different things. It works like hyperthreading just like I said.
http://en.wikipedia.org/wiki/Bulldozer_(microarchitecture)


You do realize the only shared cache is L3 right? Each core otherwise has separate L1 and L2 cache, which are far more crucial...

Additionally...for scheduling purposes it would work better that way...however, windows scheduler is typically poor compared to other things. If you look at the PD performance in Linux you can see the difference between a "efficient" and "less efficient" scheduler.


Yes I realize that, but since only 5% of computer users use Linux.....need I say more? I'm tired of arguing about this, I showed proof from tomshardware quoted on wiki, so I'm done. You can sit in your moms basement all day and try to prove me wrong.


So, let me get this straight...

You're basing your argument on windows scheduler being a poor scheduler. Because it's a poor scheduler, if it handled the scheduling more efficiently...like it does for a HTT enabled Intel CPU...it may improve performance. Since that may improve performance, you are insinuating that the modular design of AMD's CPU architecture is similar to HTT...? All this, even though HTT has no physical resources outside of an extra register stack, and AMD architecture has a physical core with 2 ALUs and dedicated L1 and L2 cache??

Hmm...then you justify this by quoting a misinformed writer, trying to make an off the cuff comparison, that was quoted by Wikipedia?

That argument holds as much water as a minnow bucket...(if you're not a fisherman, they have holes in them from top to bottom)

Yes, AMD was trying to follow in Intels footsteps like they always do and create something that was hyperthreading but better and they failed. The only area that AMD CPU's work better than intels is integer operations because of the 8 integer clusters. I've done my own tests so I know. If you look at the bottom of the Wikipedia page it shows a link to the tomshardware article that stated this. It's not like somebody just edited the page with some nonsense. You can write whatever you want from hear on out on this thread but I assure you that I won't read it because you are a troll and I see you on every CPU thread trying to argue with people because you have no life and you think you know everything.


Your Wikipedia article doesn't equate it to hyper threading by the way.

Quote:
AMD has re-introduced the "Clustered Integer Core" micro-architecture, an architecture developed by DEC in 1996 with the RISC microprocessor Alpha 21264. This technology is informally called CMT (Clustered Multi-Thread) and formally called "module" by the AMD. In terms of hardware complexity and functionality, this "module" is equal to a dual-core processor in its integer power (each thread having a fully independent integer core). While it is similar to a single core processor that has the SMT ability, which can create a dual threads processor but with the power of one (each thread shares the resources of the module with the other thread) in terms of floating point performance.


It states exactly what it is...

So where you got your info from is a bit confusing...
m
0
l
a b à CPUs
August 26, 2013 4:28:13 PM

Wow..... sometimes you just need to step away from the keyboard.
m
0
l
a b à CPUs
August 26, 2013 4:30:49 PM

8350rocks said:
swilczak said:
rishiswaz said:
swilczak said:

I'm sure that's how AMD wanted it to work


AMD made a new architecture based on more ALUs and less FPUs to deliver more real-world performance where it counts while still having floating point calculations available to those who need it. When launching the FX processors they clearly said that they were re-inventing what a core should be, so not it is not Hyperthreading at all. Some programs will report each module as a core instead of each core. 1 module contains two cores so it may seem that way but they are all physical cores with no special thread handling.


The Tom's Hardware website commented that the lower-than-expected performance in multi-threaded workloads may be because of the way Windows 7 currently schedules threads to the cores. They point out that "if Windows were able to utilize an FX-8150's four modules first, and then backfill each module's second core, it'd maximize performance with up to four threads running concurrently." This is similar to what happens on Intel CPUs with HyperThreading – Windows 7 "schedules to physical cores before utilizing logical (HyperThreaded) cores

What it is and how it actually works in windows are two different things. It works like hyperthreading just like I said.
http://en.wikipedia.org/wiki/Bulldozer_(microarchitecture)


You do realize the only shared cache is L3 right? Each core otherwise has separate L1 and L2 cache, which are far more crucial...

Additionally...for scheduling purposes it would work better that way...however, windows scheduler is typically poor compared to other things. If you look at the PD performance in Linux you can see the difference between a "efficient" and "less efficient" scheduler.


Oh and one more thing, fx processors have shared level 2 cache and shared level 3 cache, so I don't know where you get your info from but it's wrong. Someone should take away your CPU master badge because you obviously don't know that much.

http://www.cpu-world.com/CPUs/Bulldozer/AMD-FX-Series%2...
m
0
l
a c 210 à CPUs
a b À AMD
August 26, 2013 4:32:20 PM

CooLWoLF said:
Wow..... sometimes you just need to step away from the keyboard.


Entirely off topic...how do you like your MB? I thought about going with the M5A99X EVO R2.0 a few times along the way to eventually buy the M5A97 I got...sometimes I wish I went with the 990 board...though at the time there was about $40 difference between the 2 and I couldn't justify it. ($119 vs. $79)
m
0
l
a b à CPUs
August 26, 2013 4:43:27 PM

8350rocks said:
CooLWoLF said:
Wow..... sometimes you just need to step away from the keyboard.


Entirely off topic...how do you like your MB? I thought about going with the M5A99X EVO R2.0 a few times along the way to eventually buy the M5A97 I got...sometimes I wish I went with the 990 board...though at the time there was about $40 difference between the 2 and I couldn't justify it. ($119 vs. $79)

Absolutely love it! Easily overclocked my 8350 up to 4.5ghz with little effort at all. My buddy got the board as well and cranked his 8350 up to 4.8ghz, although he is using an H100i so he has better cooling than me. Again, he pushed it to that speed with no sweat. I am going to try hitting 4.8ghz after getting some better fans (higher pressure) for my current water cooler. I am so glad I made the switch to this board vs. my original MSI 970a board.
m
0
l
a c 210 à CPUs
a b À AMD
August 26, 2013 4:47:31 PM

swilczak said:
8350rocks said:
swilczak said:
rishiswaz said:
swilczak said:

I'm sure that's how AMD wanted it to work


AMD made a new architecture based on more ALUs and less FPUs to deliver more real-world performance where it counts while still having floating point calculations available to those who need it. When launching the FX processors they clearly said that they were re-inventing what a core should be, so not it is not Hyperthreading at all. Some programs will report each module as a core instead of each core. 1 module contains two cores so it may seem that way but they are all physical cores with no special thread handling.


The Tom's Hardware website commented that the lower-than-expected performance in multi-threaded workloads may be because of the way Windows 7 currently schedules threads to the cores. They point out that "if Windows were able to utilize an FX-8150's four modules first, and then backfill each module's second core, it'd maximize performance with up to four threads running concurrently." This is similar to what happens on Intel CPUs with HyperThreading – Windows 7 "schedules to physical cores before utilizing logical (HyperThreaded) cores

What it is and how it actually works in windows are two different things. It works like hyperthreading just like I said.
http://en.wikipedia.org/wiki/Bulldozer_(microarchitecture)


You do realize the only shared cache is L3 right? Each core otherwise has separate L1 and L2 cache, which are far more crucial...

Additionally...for scheduling purposes it would work better that way...however, windows scheduler is typically poor compared to other things. If you look at the PD performance in Linux you can see the difference between a "efficient" and "less efficient" scheduler.


Oh and one more thing, fx processors have shared level 2 cache and shared level 3 cache, so I don't know where you get your info from but it's wrong. Someone should take away your CPU master badge because you obviously don't know that much.

http://www.cpu-world.com/CPUs/Bulldozer/AMD-FX-Series%2...


L3 cache is the only thing shared across the CPU, the L2 cache is shared per "module" however, the L2 cache on a bulldozer module is 2.5x what it is per core on the IB architecture.

So essentially, if you're faulting the architecture for that, you need to also fault Intel for not giving you enough cache in the first place per core (or you could accept that it isn't an issue, because it's not). As the 8350 per core has 1.25x the L2 cache of a 3570k per core.
m
0
l
a c 210 à CPUs
a b À AMD
August 26, 2013 4:48:36 PM

CooLWoLF said:
8350rocks said:
CooLWoLF said:
Wow..... sometimes you just need to step away from the keyboard.


Entirely off topic...how do you like your MB? I thought about going with the M5A99X EVO R2.0 a few times along the way to eventually buy the M5A97 I got...sometimes I wish I went with the 990 board...though at the time there was about $40 difference between the 2 and I couldn't justify it. ($119 vs. $79)

Absolutely love it! Easily overclocked my 8350 up to 4.5ghz with little effort at all. My buddy got the board as well and cranked his 8350 up to 4.8ghz, although he is using an H100i so he has better cooling than me. Again, he pushed it to that speed with no sweat. I am going to try hitting 4.8ghz after getting some better fans (higher pressure) for my current water cooler. I am so glad I made the switch to this board vs. my original MSI 970a board.


I really liked some of the features on that board. May bite the bullet and get something similar later on...
m
0
l
a b à CPUs
August 26, 2013 5:47:01 PM

swilczak said:

Oh and one more thing, fx processors have shared level 2 cache and shared level 3 cache, so I don't know where you get your info from but it's wrong. Someone should take away your CPU master badge because you obviously don't know that much.

http://www.cpu-world.com/CPUs/Bulldozer/AMD-FX-Series%2...


When "facts" fail resort to personal attacks. Just because Windows' scheduler reports it as 2/3/4 cores with 4/6/8 threads does not change that it has 4/6/8 cores with them being in groups of 2 per module having 2/3/4 modules. Because it is reported like Hyperthreading does not make it HTT. So far he has shown more CPU knowledge than you have and has been more courteous in his replies so I think the CPU Master badge seems to be well earned.
m
0
l
a b à CPUs
August 26, 2013 5:56:13 PM

rishiswaz said:
swilczak said:

Oh and one more thing, fx processors have shared level 2 cache and shared level 3 cache, so I don't know where you get your info from but it's wrong. Someone should take away your CPU master badge because you obviously don't know that much.

http://www.cpu-world.com/CPUs/Bulldozer/AMD-FX-Series%2...


When "facts" fail resort to personal attacks. Just because Windows' scheduler reports it as 2/3/4 cores with 4/6/8 threads does not change that it has 4/6/8 cores with them being in groups of 2 per module having 2/3/4 modules. Because it is reported like Hyperthreading does not make it HTT. So far he has shown more CPU knowledge than you have and has been more courteous in his replies so I think the CPU Master badge seems to be well earned.

I know you haven't been reading what we've been talking about because I was talking about how windows utilizes the cores, not just what it reports. If you think I'm being a jerk then please report me. I've just had enough of seeing the same two people on every cpu thread, we've got 8350rocks (AMD fanboy) and Hajifur (Intel Fanboy) on almost every CPU thread. Tomshardware is suppose to be about facts, not people defending whatever CPU they own. I don't even own any Intel gaming cpu and never have, I just know they are better than AMD. It doesn't take a genius to know that.
m
0
l
August 26, 2013 9:28:41 PM

OK Why don't we just stop this nonsense and leave. I think the general consensus is that no, the 6350 will not bottleneck the 7870. Done. No need to spew crap about the details concerning how these architectures work.
m
3
l
!