AMD CPU speculation... and expert conjecture

[mayankleoboy1]

rofl, you and your love for techspot's alternate wording for minimum fps measured in 1 frame intervals.

let me expand that : "minimum Frames Per Second measured in 1 frame intervals" .
What does this even mean? . Have you ever patiently read the methodology done by techreport ?

like I said. all those "cpu bound" games are supported by Intel, posted on Intel's website for being optimized for Intel. That makes it ok to bench AMD hardware on and expect the results to be accurate simply because it runs slower on AMD?

And the other games are AMD supported. So they cant be compared on Intel hardware. Also, TWIMTBP games shouldnt be tested on AMD GPU's, right ?

So basically everyone should own 4 computers : intel+nv, intel+ATI, AMD+nv, AMD+ATI. And there is no performance comparison possible between the four systems. If you dont own any of the four systems, you automatically are a Fanboi.

 

[de5_Roy]

.....
So basically everyone should own 4 computers : intel+nv, intel+ATI, AMD+nv, AMD+ATI. And there is no performance comparison possible between the four systems. If you dont own any of the four systems, you automatically are a Fanboi.

then, i am not a fanboy if i get a chromebook with samsung exynos and play angry birds..

anyway...
Richland successor in 2014 is Kaveri
http://www.fudzilla.com/home/item/29986-richland-successor-in-2014-is-kaveri
Kabini can be coupled with discrete HD 8000 cards
http://www.fudzilla.com/home/item/29985-kabini-can-be-coupled-with-discrete-hd-8000-cards

if noob's theory(!) is right, then upcoming consoles are going to suck so bad at games... may be that is why wii u has so weak hardware, nintendo gave up getting more powerful hw knowing they will not win against intel the chipzilla. the upcoming xbox720 and ps4 will suck for this very reason (at least the one that has amd hw in it).

XBox Next/720 silicon production day arrives
http://semiaccurate.com/2013/01/02/xbox-next720-silicon-production-day-arrives/

 

[noob2222]

let me expand that : "minimum Frames Per Second measured in 1 frame intervals" .
What does this even mean? . Have you ever patiently read the methodology done by techreport ?

like I said. all those "cpu bound" games are supported by Intel, posted on Intel's website for being optimized for Intel. That makes it ok to bench AMD hardware on and expect the results to be accurate simply because it runs slower on AMD?

And the other games are AMD supported. So they cant be compared on Intel hardware. Also, TWIMTBP games shouldnt be tested on AMD GPU's, right ?

So basically everyone should own 4 computers : intel+nv, intel+ATI, AMD+nv, AMD+ATI. And there is no performance comparison possible between the four systems. If you dont own any of the four systems, you automatically are a Fanboi.

rofl. TWIMBP can be turned off ... what a concept, turn off physx ... lets turn off the Intel compiler in the other games ... oh wait, you can't, you don't get to choose what compiler is used when you run or install the game.

funny thing is some twimbp games run faster on ATI hardware when you turn off physx.

Its a know fact that Intel disabled all functions on AMD processors with their compilers, SSE, MMX, ect. Functions that AMD paid roalty fees to Intel to be allowed to put them on the AMD cpus. Brilliant business indeed, sell a side-product to competitor and disable that side-product with your software.

http://downloadsquad.switched.com/2010/01/04/intel-forced-to-provide-a-compiler-that-isnt-crippled-for-amd-processors/

what was Intel's solution? you have to specify the compiler to enable extenstions on AMD hardware. If you just run it, its still disabled by default. Intel also added this clause to their software.

Intel® compilers, associated libraries and associated development tools may or may not optimize to the same degree for non-Intel microprocessors for optimizations that are not unique to Intel microprocessors. These optimizations include Intel® Streaming SIMD Extensions 2 (Intel® SSE2), Intel® Streaming SIMD Extensions 3 (Intel® SSE3), and Supplemental Streaming SIMD Extensions 3 (Intel® SSSE3) instruction sets and other optimizations.

some fix huh? Intel did as little as possible to make the courts happy and called it a day.

yet its a conspiracy to believe that Intel didn't do all that was possible to make AMD hardware just as optomized.

 

[gamerk316]

so if you tell the program to run a thread on X cpu and only X cpu ... it will randomly run or not run ... makes perfect sense ... especially when you program that particular program to use the same core over and over? What programmer does this? then again, maybe you should quit trying to run 2-3 high cpu usage programs at the same time.

Ill run skyrim, bf3, and world of warcraft all at the same time ... because thats the only way to test my program to see if it slows down.

If this is how programs are tested, no wonder they will never scale past 2 cores. you need the other 6 cores for the other 3 programs that are running.

Someone doesn't get the concept of "benchmarking".

rofl, you and your love for techspot's alternate wording for fps measured in 1 frame intervals.

Apparently, you haven't read the methodology behind it at all. Or more likely, you choose to ignore it because it disagrees with your misconceptions.

rofl. TWIMBP can be turned off ... what a concept, turn off physx ... lets turn off the Intel compiler in the other games ... oh wait, you can't, you don't get to choose what compiler is used when you run or install the game.

Name me one game that uses Intels compiler. Just one.

And no, naming a random game without any proof doesn't count.

yet its a conspiracy to believe that Intel didn't do all that was possible to make AMD hardware just as optomized.

Find me a compiler that results in a faster executable for AMD hardware then Intels compiler.

Heres a hint: It doesn't exist.


So yeah, I'm calling you a fanboy (as if we didn't know that already).

 

[mayankleoboy1]

@ Noob222 :
AMD's own compiler, AMD64 sucks big time. And its Linux only.

I really hope you know what GCC is. Its an open source compiler, that 99.999% of linux devs use. What that means is that it cant be biased against any one company. Go to Phoronix.com. They do compiler testing. Go read some benchmarks.

I hear that majorly only HPC people use ICC.

 

[mayankleoboy1]

@ Gamerk316 : What do major game devs use for compiling ? or is the concept of 'building' not so simple there?

 

[mayankleoboy1]

Really? can't at all ... ya ..

http://www.ncsa.illinois.edu/UserInfo/Resources/Software/Intel/Compilers/10.0/main_cls/mergedProjects/optaps_cls/common/optaps_openmp_thread_affinity.htm

http://msdn.microsoft.com/en-us/library/ms684251(v=vs.85).aspx

Kinda looks like to me you can pretty much tell a specific thread from letting the os decide to narrowing it down to 4, 2, or 1 specific core in an 8 core system.

Its even available on the Intel compiler. http://software.intel.com/en-us/articles/thread-affinity-compiler-options-and-environment-variables/

Did you even read what you have linked ? :lol:
Google search for "windows thread scheduling" gives me the same link the contents are quite different.

 

[gamerk316]

@ Gamerk316 : What do major game devs use for compiling ? or is the concept of 'building' not so simple there?

Depends. Anything cross platform, and you see more usage of GCC. Otherwise, I'd expect MSVC or some middleware engine, similar to the rest of the industry. Only way to know for sure though is to look at the compiled program for "signatures" that compilers leave behind.

 

[JAYDEEJOHN]

http://www.xtremesystems.org/forums/showthread.php?284523-If-the-frequency-is-everything....WR-of-Andre-Yang!&p=5162203&viewfull=1#post5162203

 

[JAYDEEJOHN]

Parallel command streams.....

 

[noob2222]

Name me one game that uses Intels compiler. Just one.

And no, naming a random game without any proof doesn't count.

I guess Intel Optimized means compiled with microsoft c++ because who would ever use the Intel compiler if Intel offered it for free and supplied several software engineers on hand if they had any questions. Microsoft would love to have intel rework their compiler to optimize code for Intel cpus, after all isn't that how you support your competition by rewriting their software for them?

Then again, shouldn't it be blatently obvious when a program is plastered with the words Intel at every stage of development that its most likely writtein on ICC, including when the beta testing is promoted by Intel?

then again, Intel optimized must mean that the microsoft compiler was used.

Find me a compiler that results in a faster executable for AMD hardware then Intels compiler.

Heres a hint: It doesn't exist.

http://agner.org/optimize/blog/read.php?i=179#179

Choosing the most efficient function library can be a nightmare to a programmer. I have tried to calculate the cosine function with different libraries and compare the calculation time. The best version is 19 times faster than the worst!

its near the bottom, 2012-4-18 is the date. here is the kicker ... the worst ... is Intel's default code for AMD's BD cpu ...

In 64 bit mode, MS compiler was the faster until you get to the SVML.

Interpetation of "it doesn't exist" depends on wether or not you know how to enable SSE for AMD cpus. All other compilers are faster if you use the default Intel.

 

[noob2222]

Did you even read what you have linked ? :lol:
Google search for "windows thread scheduling" gives me the same link the contents are quite different.

Thats not the point. Apparently you can't schedule a high priority thread to its own core becuse some other program might interfere with it somehow. Instead you have to program like its the 80s and let the os handle all of the scheduling for you. You want the high priority threads bouncing from core to core so that you cannot predict when the os is going to scheule it. You most definately can't hand schedule affinity just in case some moron wants to see what happens if he runs bf3, skyrim and WoW at the same time.

If his machine crashes, its the game developers fault for locking high-priority threads and trying to make their program more efficient in a multi-core system. This is why you can't program for multi cores. It doesn't make sense to spend all that time trying to make it run faster when some moron will just crash his computer some day by doing something stupid.

 

[gamerk316]

I guess Intel Optimized means compiled with microsoft c++ because who would ever use the Intel compiler if Intel offered it for free and supplied several software engineers on hand if they had any questions. Microsoft would love to have intel rework their compiler to optimize code for Intel cpus, after all isn't that how you support your competition by rewriting their software for them?

MSVC has been around a LOT longer then Intels compiler. Plus you get better drop in support for new Windows features (remember the primary development platform here). Nevermind MSVC has, by far, the best debug toolchain. Nothing else comes close.

Only time I would ever use Intel's compiler for development would be if I were writing a native X86 application. Otherwise, I'll stick with MSVC. The ease of development outweighs the possible performance gain of using Intels compiler.

Then again, shouldn't it be blatently obvious when a program is plastered with the words Intel at every stage of development that its most likely writtein on ICC, including when the beta testing is promoted by Intel?

Not really. Apparently, you REALLY don't get how marketing works.

Find me a compiler that results in a faster executable for AMD hardware then Intels compiler.

Heres a hint: It doesn't exist.

http://agner.org/optimize/blog/read.php?i=179#179

Choosing the most efficient function library can be a nightmare to a programmer. I have tried to calculate the cosine function with different libraries and compare the calculation time. The best version is 19 times faster than the worst!

its near the bottom, 2012-4-18 is the date. here is the kicker ... the worst ... is Intel's default code for AMD's BD cpu ...

In 64 bit mode, MS compiler was the faster until you get to the SVML.

Interpetation of "it doesn't exist" depends on wether or not you know how to enable SSE for AMD cpus. All other compilers are faster if you use the default Intel.

Out of curiosity, do you even know what SVML is? The Simple Vector Math Library is a very high performance math library for 2d/3d/4d vector types. Its worth noting, SVML does not work under MSVC. So that explains the speedup against MSVC rather nicely. I also note SVML is fully supported by AMD CPU's, blowing your bias argument out of the water as well.

Secondly, the quote you quoted was referring to choice of library, NOT the difference in speed due to compilation between two compilers. You are simply continuing to prove you have NO IDEA WHATSOEVER about software development.

Thirdly, no one compiles with the default compiler switches. Or are we going to degenerate into an argument that compiler X is faster then compiler Y because compiler Y has /O3 set by default? When comparing compiler performance, you ALWAYS do so with the same compile options set.

 

[noob2222]

Only time I would ever use Intel's compiler for development would be if I were writing a native X86 application. Otherwise, I'll stick with MSVC. The ease of development outweighs the possible performance gain of using Intels compiler.

Not really. Apparently, you REALLY don't get how marketing works.

you apparently don't either. Prove your competition is inferior by whatever means necessary, promote the heck out of that product and watch the reviews fly. If Intel came up to you and offered their support for your program, paid you to use their software, would you turn them down?

OMG AMD STRAIGHT SUCKS, OMG LOOK HOW BAD AMD IS, OMG INTEL RULZ, AMD SUKORS. DUED NOT EVEN A 5.0 GHZ AMD CAN KEEP UP WITH A 2.5 GHZ INTEL.

ya, thats not marketing at all, AMD just sucks.

http://www.intel.com/content/www/us/en/gaming/overclocking/starcraft-wallpaper-3.html
http://software.intel.com/sites/billboard/article/blizzard-entertainment-re-imagines-starcraft-intels-help
http://software.intel.com/en-us/blogs/2010/03/09/say-the-password-to-get-on-the-beta-list-for-starcraft-ii/
http://software.intel.com/en-us/blogs/2010/03/22/visualize-this-blizzard-on-starcraft-2-and-intel-core-i7/

ya, Intel had no intention of making AMD look bad with starcraft II, its just AMD sucks.


____________________

 

[esrever]

I guess we are back to this argument again.

 

[gamerk316]

you apparently don't either. Prove your competition is inferior by whatever means necessary, promote the heck out of that product and watch the reviews fly. If Intel came up to you and offered their support for your program, paid you to use their software, would you turn them down?

http://images.anandtech.com/graphs/graph6396/51141.png

OMG AMD STRAIGHT SUCKS, OMG LOOK HOW BAD AMD IS, OMG INTEL RULZ, AMD SUKORS. DUED NOT EVEN A 5.0 GHZ AMD CAN KEEP UP WITH A 2.5 GHZ INTEL.

ya, thats not marketing at all, AMD just sucks.

http://www.intel.com/content/www/us/en/gaming/overclocking/starcraft-wallpaper-3.html
http://software.intel.com/sites/billboard/article/blizzard-entertainment-re-imagines-starcraft-intels-help
http://software.intel.com/en-us/blogs/2010/03/09/say-the-password-to-get-on-the-beta-list-for-starcraft-ii/
http://software.intel.com/en-us/blogs/2010/03/22/visualize-this-blizzard-on-starcraft-2-and-intel-core-i7/

ya, Intel had no intention of making AMD look bad with starcraft II, its just AMD sucks.


____________________

Starcraft doesn't scale beyond two cores, so performance is dominated by clockspeed and IPC. That favors Intel. Adjusting for clocks, you see a constant 1.5-1.6 performance factor, which is about the IPC difference between Intel and AMD. You see these same numbers repeat in any game that does not scale well beyond 2 cores, which points to Intels superior IPC, and not compilation enhancements.

noob, what you are trying to do is take Intels superior performance as "proof" that everyone uses a compiler that is disadvantageous to AMD, despite the fact that the vast majority of the industry doesn't use said compiler, and the fact said compiler offers better performance for AMD then MSVC (as far as I've tested anyway). You are simply incapable of accepting that AMD's performance in non-threaded performance is inferior, despite all evidence to the contrary.

 

[keithlm]

Name me one game that uses Intels compiler. Just one.

The Lord of the Rings Online.

EDIT: Actually just re-scanned it. Apparently they stopped using ICC in the last 4 or 5 months since I last did a scan for ICC.

But they did use ICC from the release in 1997 until just recently. (It is not surprising they recently changed since they are supposedly working on a MAC version of the executable... so I'm sure they moved to a cross platform compiler.)

BTW: Adobe still uses it. (Lightroom, Flash, etc.) And so does Futuremark, Valve (Steam), and Cinebench.

Thirdly, no one compiles with the default compiler switches.

Actually MOST developers use the compiler defaults. Perhaps not a game developer... but probably 90% of the rest of all developers.

(I have had to fight with lead programmers to get them to adopt more optimal compiler switches in the past.)

 

[m32]

@gamerk316 Thanks for calling him out!

 

[noob2222]

Starcraft doesn't scale beyond two cores, so performance is dominated by clockspeed and IPC. That favors Intel. Adjusting for clocks, you see a constant 1.5-1.6 performance factor, which is about the IPC difference between Intel and AMD. You see these same numbers repeat in any game that does not scale well beyond 2 cores, which points to Intels superior IPC, and not compilation enhancements.

noob, what you are trying to do is take Intels superior performance as "proof" that everyone uses a compiler that is disadvantageous to AMD, despite the fact that the vast majority of the industry doesn't use said compiler, and the fact said compiler offers better performance for AMD then MSVC (as far as I've tested anyway). You are simply incapable of accepting that AMD's performance in non-threaded performance is inferior, despite all evidence to the contrary.

rofl, thats marketing BS and you bought 100% of it.

Note the 8150 is faster than the 1100T

See how much slower the 1100T is than the 8150? Oh wait, its not ... ya, try again.

 

[jdwii]

rofl, thats marketing BS and you bought 100% of it.

http://images.anandtech.com/graphs/graph4955/41703.png

Note the 8150 is faster than the 1100T

http://media.bestofmicro.com/M/A/310546/original/lame per core.png

See how much slower the 1100T is than the 8150? Oh wait, its not ... ya, try again.

One uses the FPU and one uses the integer.

 

[Chad Boga]

Poor little Noob, it is always big bad Intel to blame for all of AMD's woes.

AMD is never responsible for their own failures.

 

[mayankleoboy1]

Noob, you persist on wildly speculating about compiler shenanigns in closed source games, but are completely blind to open source application benchmarks.

 

[dailybargainshop]

I am newbie here, and searching some entertainment gadgets..

 

[dailybargainshop]

Hey dear, Nice information keep sharing !!

 

[noob2222]

Poor little Noob, it is always big bad Intel to blame for all of AMD's woes.

AMD is never responsible for their own failures.

Same can be said for Intel fanboys that claim Intel can't be blamed for anything.

Nothing is Intel's fault, even with intel's compiler, its 100% because its AMD, and has nothing to with the fact that Intel pushed things farther from the truth than they actually are.

Id say more 50/50 Intel's fault.