AMD CPU speculation... and expert conjecture

[JAYDEEJOHN]

PPC = 4800+ = Intel @ 4.8 Ghz

@ gamer
Realize, the runt frames arent unrealistic numbers, its most simply bad frames, and by the time the apus are out, the drivers will have corrected much of this, as the 7990 is just around the corner, and every bencher this side of Mars is waiting with baited breath as to whether AMD has/will corrected this.

Wavy Davy said waaay back they were cfx compatible
Im not so sure RR is happy with perf or timeliness as concerning the driver team, but I think this will change

 

[JAYDEEJOHN]

When it comes to xbit and Intel and AMD, its more like exchewedup

 

[truegenius]

if core i3 3225 can do that, who knows how hard haswell will hit... i hear there's a new driver out boosting 10% moar...

time to test intel igpu for bad/runt frames

Ironic that you brought it up, the last news out is yes it supports dual graphics with GCN(or GCN 2) and VLIW parts, so yes the Kaveri APU's can operate with existing Dual Graphics support DDR3 and DDR5 compatibile parts. Not only socket stability but GPU support stability which is great for users who may not be able to pay $60-80 for a HD8600 discrete part.

The HD6770 is not a supported GPU for DG, you need Turks based GPU's and the not yet revealed Sea Islands and Solar System parts.

:/
for current apus , gpu cores and performance are near hdx6xx level
but next gen is supposed to be equal to hd7700 series so i thought that they may include hdx7xx series too and thus hd6770 too (because upto hd7600 and upto hd6600 series is same arch as that of hd6770 but with different core counts)

 

[de5_Roy]

time to test intel igpu for bad/runt frames

good call!

 

[Ranth]

then how about this one:
http://www.xbitlabs.com/images/cpu/trinity-vs-ivy-bridge/f12012-2.png
if core i3 3225 can do that, who knows how hard haswell will hit... i hear there's a new driver out boosting 10% moar... ;D
don't blame me, i see techspot's bf3 bench being posted here all the time.

Well that was one out of 8 I'm not sure if you're just teasing but I think you should have mentioned the Hitman absolution and the sleeping dogs runs as well
http://www.xbitlabs.com/articles/cpu/display/trinity-vs-ivy-bridge_10.html

 

[de5_Roy]

Well that was one out of 8 I'm not sure if you're just teasing but I think you should have mentioned the Hitman absolution and the sleeping dogs runs as well
http://www.xbitlabs.com/articles/cpu/display/trinity-vs-ivy-bridge_10.html

i know that. but after hearing repeated ramblings about how toms dared to compare intel and amd igpus... i find that little anomaly, it was like a miracle...

TSMC 20nm Fab 14 coming on line soon
http://www.fudzilla.com/home/item/30939-tsmc-20nm-fab-14-coming-on-line-soon
if apple doesn't monopolize 20nm, amd has a chance to jump way ahead in gpu and low power apu manufacturing (hint: rival arm vendor).

Report: No easing in sight for PC memory prices
http://techreport.com/news/24596/report-no-easing-in-sight-for-pc-memory-prices

 

[sarinaide]

then how about this one:
http://www.xbitlabs.com/images/cpu/trinity-vs-ivy-bridge/f12012-2.png
if core i3 3225 can do that, who knows how hard haswell will hit... i hear there's a new driver out boosting 10% moar... ;D
don't blame me, i see techspot's bf3 bench being posted here all the time.

http://www.kitguru.net/components/graphic-cards/zardon/asus-f2a85-v-pro-amd-a10-5800k-w-hd7660d/21/

True APU result, Intel HD made worse than it is - AMD bias

http://www.hardwareheaven.com/reviews/1583/pg9/amd-a10-5800k-apu-performance-review-gaming-performance.html

Better results but not accurate. Yes that is the result of a i7 3770K, so we are starting to paint a picture here of reviewers inflating intel iGPU numbers.

Here is the absolute kicker, this has basically made me view xBit as a micky mouse site, these results are about right and we have tested the 2012 results with i7, i5 and 13 HD4000 and all APU's and the results at Ultra settings at 1080P with 2xMSAA gets you around 39-43 Average FPS on a HD7660D and around 21-24 on HD4000 (regardless of the CPU) basically xBit fucked up there own review.

http://www.xbitlabs.com/articles/graphics/display/amd-trinity-graphics_7.html (EDIT Forgot the link)

Complete and utter balls up. If they got F1 2012 with a in game Benchmark tool wrong I shudder to even contemplate how wrong everything else is. Since F1 2012 is a game I play very often and part of Apex racing league I can record my game for you, record the frames and frame times, record max, min and average, show you the game is on ultra settings on the apu where intel including haswell is barely playable on medium at 1080.

BF3 and Metro 2033 I often use because they stress GPU performance, in games that flex the iGPU performance to its max the AMD HD7660D is about 2-3x faster than HD4000 and about 1.5-2x faster than HD4600, in percentages its around 55%+ faster than HD4000 and around 35% faster than HD4600. 60+ FPS at 1280x768 lowest BF3 settings, take the 4770K if you like and go try beat it

10% boost if lucky, in a game like F1 2010 thats about 1-2 FPS which is barely exceptional and to make matters worse intels top dog iGPU solution at $350 is really only the match of dual core llano and trinity parts or lowest end iGPU which is sub $100 territory.

 

[sarinaide]

if core i3 3225 can do that, who knows how hard haswell will hit... i hear there's a new driver out boosting 10% moar...

time to test intel igpu for bad/runt frames

Ironic that you brought it up, the last news out is yes it supports dual graphics with GCN(or GCN 2) and VLIW parts, so yes the Kaveri APU's can operate with existing Dual Graphics support DDR3 and DDR5 compatibile parts. Not only socket stability but GPU support stability which is great for users who may not be able to pay $60-80 for a HD8600 discrete part.

The HD6770 is not a supported GPU for DG, you need Turks based GPU's and the not yet revealed Sea Islands and Solar System parts.

:/
for current apus , gpu cores and performance are near hdx6xx level
but next gen is supposed to be equal to hd7700 series so i thought that they may include hdx7xx series too and thus hd6770 too (because upto hd7600 and upto hd6600 series is same arch as that of hd6770 but with different core counts)

Intel HD even if its playable FPS has some serious latency, F1 2012 even at lowest settings feels like a slideshow, Dirt 3 is very slideshow, BF3 is unplayable at all, Metro kills a titan so not much hope here, Crysis 2 is a stutterfest at lowest settings. Intel HD is hardware output not much more irrespective of what frame rates say.

 

[8350rocks]

Well, if you have 4 cores that can run 32 IPC versus 8 cores that can run 44 IPC, I would think, even with less efficiency of code processing, you'll flat out get more operations out of the AMD chip, versus the intel.

He's right on that everyone here has been using "IPC" incorrectly and out of context. IPC is not how fast a CPU is, its an engineering term used to measure the efficiency of the integer processor and used almost exclusively in RISC uArchs. In x86 it has no meaning for one very simple reason. x86 Instructions have variable execution time with most requiring more then one clock cycle to actually execute. RISC design's on the other hand almost always have 1 cycle execution times for all instructions, they is why they use explicit load / store instructions rather then combined instructions. This limit is actually what had many of the original RISC design's beating out the x86 ones for performance (SPARC / PPC / MIPS) until Intel made the Pentium. That CPU did something unique in that it wasn't an x86 CPU, it only emulated one. Internally it was a RISC processor that had a front end x86 decoder / scheduler bolted on to trick the system into thinking it was an x86 CPU. Using this technique Intel was able to put the functional equivalent of multiple processors inside a single logical CPU that could then rapidly execute multiple instructions. The last true x86 CPU was the i486 era CPUs.

That is all important to keep in mind because when your trying to count instructions you need to specify which instructions your wanting to count as their not all the same. AMD and Intel both have different internal microprocessor languages, the real instructions being executed are masked from the OS and so we have no idea what exactly is going on inside (not without access to confidential engineering information). One particular instruction make take longer on one CPU then on another. AMD's BD in particular has a very sensitive cache access method that could easily destroy performance if not handled properly.

Anyhow the take away from all this is to stop using the word "IPC". You can simply say processing efficiency, or single core capacity. Capacity would be the proper term to use as it describes the maximum amount of work able to be done, either by a single unit or by the whole entity. Also do not use clock as a measurement of anything involving two different CPU designs. For the above reasons clock rates have dramatically different impacts depending on the design. The two best metrics are per cost and per energy used as they describe what the user / integrator is paying for that capacity.

I know what IPC is, and it doesn't have a real bearing on speed, only efficiency of the internal architecture. If AMD architecture is more optimized...(Currently the FX-8350 has a maximum IPC of 28 where intel is at 32, while Steamroller will have a 8 core CPU with a maximum IPC of 44). So you are literally gaining ~50% internal efficiency. That will lead to a greater instruction execution rate at the same clock speed as a stock 8350 has right now. So a 4.0 GHz Steamroller 8 core will be operating at ~150% of the efficiency internally, even though the clock speed did not technically increase, the instructions handled per cycle increased massively making for more efficient use of clocktime. So the CPU would feel ~50% faster in many applications even though the clock speed never changed...it was just capable of handling more instructions at once than it previously was...

On the coding end of things and the runtime for instructions, things begin to get more convoluted...I am a hardware guy...so I will gracefully decline to get into that debate.

 

[JAYDEEJOHN]

Obviously Intel seeks SFF/screens and good enough strategy.
My concern is, as we scrutinize AMD and runt frames, as was mentioned, since this market is a growing dynamic market, shouldnt Intel be held to the same scrutiny?
I dont want to see after, its good enough, as sales mount, and a lessor game experience is set for average Joe, I want to see Intel in gfx with their big boy pants on, and go thru what everyone has had to do, be it runt frames, blurienss, etc etc

 

[sarinaide]

Obviously Intel seeks SFF/screens and good enough strategy.
My concern is, as we scrutinize AMD and runt frames, as was mentioned, since this market is a growing dynamic market, shouldnt Intel be held to the same scrutiny?
I dont want to see after, its good enough, as sales mount, and a lessor game experience is set for average Joe, I want to see Intel in gfx with their big boy pants on, and go thru what everyone has had to do, be it runt frames, blurienss, etc etc

You honestly think they will do something that potentially will tell the truth about intels iGPU that it is not a gaming part, that it is video out at best, you have to be joking. Its easier to inflate numbers, pander on about 2x the performance while trying to mask over that its simply not good, if you have any ambitions on playing games on integrated solutions the A10 5800K is cheaper than every Intel part and murders them all past, present and to come, looking at the road maps it will be sometime in 2016 where Intel may just about match Trinity, who knows what AMD has by then, obviously if they are still in business.

 

[griptwister]

Crap x 2 = 2ce the crap

Intel HD graphics are terrible... I could care less what anyone else says... You know no one is going to be honest on graphics performance... I'd rather have a single A10 APU (heck, even an A8) for gaming than a single i7 4770k for gaming.

That power issue that intel is have, I'm pretty sure its true... have you seen how many power phases those friggin z87 boards have?!?

 

[de5_Roy]

http://www.kitguru.net/components/graphic-cards/zardon/asus-f2a85-v-pro-amd-a10-5800k-w-hd7660d/21/

True APU result, Intel HD made worse than it is - AMD bias

am i supposed to believe an amd biased site now? neither kitguru nor hardwareheaven provide intel/amd gfx driver versions (or any driver versions), windows 8, updated softwares. xbitlabs' test systems actually have higher bw ram not underspecced like the other two's. in your own posted links, hd4k can keep up with 7660d rather well in borderlands 2, f1 2012 and dota benches either in min. fps or avg. fps or both. so the xbitlabs' f1 2012 @1080p lowQ is not that far fetched. shows that intel can make improvements as well, even though kg and hh both used more intense gfx settings.

Spoiler

http://www.hardwareheaven.com/reviews/1583/pg9/amd-a10-5800k-apu-performance-review-gaming-performance.html

Better results but not accurate.

Here is the absolute kicker, this has basically made me view xBit as a micky mouse site, these results are about right and we have tested the 2012 results with i7, i5 and 13 HD4000 and all APU's and the results at Ultra settings at 1080P with 2xMSAA gets you around 39-43 Average FPS on a HD7660D and around 21-24 on HD4000 (regardless of the CPU) basically xBit fucked up there own review.

Complete and utter balls up. If they got F1 2012 with a in game Benchmark tool wrong I shudder to even contemplate how wrong everything else is. Since F1 2012 is a game I play very often and part of Apex racing league I can record my game for you, record the frames and frame times, record max, min and average, show you the game is on ultra settings on the apu where intel including haswell is barely playable on medium at 1080.

rofl. reminds me that time when techreport fell off amd fanboy favor when they used cinebench to test intel and amd cpus....
if xbitlabs !@#$ed up their own review, the other benches where 7660d leads might be invalid as well.

 

[sarinaide]

When it comes to xbit and Intel and AMD, its more like exchewedup

Apparently they cannot get their own benches right, so they inflate i3 performances higher than the i7 and get busted. Now De5_Roy is harbouring thoughts that Haswell's HD is exceptional, hell this was the part that was heralded the AMD slayer a year ago, it was playing Skyrim at 1080 resolution with Ultra settings, fast forward a year, haswell has been delayed twice, the leaks have shown nowhere near what it was puffing and ES samples back it up, I think ironically Haswell has become the intel iGPU slayer, twice the resources for moderate gains.

Broadwell, motherwell, cant you tell, oh well, sky lake, bake a cake Intel and iGPU is just not fun. I can't wait until they hit cell phones territory, I will need to walk around with a portable charger kit to keep that puppy charged.

 

[sarinaide]

http://www.kitguru.net/components/graphic-cards/zardon/asus-f2a85-v-pro-amd-a10-5800k-w-hd7660d/21/

True APU result, Intel HD made worse than it is - AMD bias

am i supposed to believe an amd biased site now? neither kitguru nor hardwareheaven provide intel/amd gfx driver versions (or any driver versions), windows 8, updated softwares. xbitlabs' test systems actually have higher bw ram not underspecced like the other two's. in your own posted links, hd4k can keep up with 7660d rather well in borderlands 2, f1 2012 and dota benches either in min. fps or avg. fps or both. so the xbitlabs' f1 2012 @1080p lowQ is not that far fetched. shows that intel can make improvements as well, even though kg and hh both used more intense gfx settings.

Spoiler

http://www.hardwareheaven.com/reviews/1583/pg9/amd-a10-5800k-apu-performance-review-gaming-performance.html

Better results but not accurate.

Here is the absolute kicker, this has basically made me view xBit as a micky mouse site, these results are about right and we have tested the 2012 results with i7, i5 and 13 HD4000 and all APU's and the results at Ultra settings at 1080P with 2xMSAA gets you around 39-43 Average FPS on a HD7660D and around 21-24 on HD4000 (regardless of the CPU) basically xBit fucked up there own review.

Complete and utter balls up. If they got F1 2012 with a in game Benchmark tool wrong I shudder to even contemplate how wrong everything else is. Since F1 2012 is a game I play very often and part of Apex racing league I can record my game for you, record the frames and frame times, record max, min and average, show you the game is on ultra settings on the apu where intel including haswell is barely playable on medium at 1080.

rofl. reminds me that time when techreport fell off amd fanboy favor when they used cinebench to test intel and amd cpus....
if xbitlabs !@#$ed up their own review, the other benches where 7660d leads might be invalid as well.

1 - Did say KitGuru are biased
2 - Techheaven whomever it was dont know them but there results are better
3 - Xbitlabs contradicted there earlier results making a i3 outscore a i7 - Thats comical at best.

And here is why the APU scores are not wrong within margin of error, They are consistent with all bench sites, and APU's use catalyst drivers, i have 13.1 running and I honestly believe that catalyst drivers are better than the 10% crap that comes out of Intelmaginationland.

F1 2012 plays at ultra settings at 1080 with 2x MSAA around 40FPS, this is consistent in almost every review, I can do this myself to, yet its LQ result was what its maxed settings result was, and intels i3 3225 outscored its i7 part by double the rates, how did this happen? In a year AMD scored less than in the past which was benched full settings with improved drivers and intel improved double its previous benches, sometimes its okay to just accept it as crap. I don't go around telling people that AMD x86 performance is the best in the world because its incorrect, I have seen to many people talk up intel iGPU's as awesome when quite honestly its not even half as strong and bigger than AMD's.

 

[de5_Roy]

Spoiler

Apparently they cannot get their own benches right, so they inflate i3 performances higher than the i7 and get busted.

Now De5_Roy is harbouring thoughts that Haswell's HD is exceptional,

Spoiler

hell this was the part that was heralded the AMD slayer a year ago, it was playing Skyrim at 1080 resolution with Ultra settings, fast forward a year, haswell has been delayed twice, the leaks have shown nowhere near what it was puffing and ES samples back it up, I think ironically Haswell has become the intel iGPU slayer, twice the resources for moderate gains.

Broadwell, motherwell, cant you tell, oh well, sky lake, bake a cake Intel and iGPU is just not fun. I can't wait until they hit cell phones territory, I will need to walk around with a portable charger kit to keep that puppy charged.

don't kill the messenger.
i assumed haswell's igpu would be capable of being 'exceptional'(depends on how you use the word) based on the previews and leaks. i think both intel and amd future igpus have potential but in both cases i'd wait for believable reviews.
once again, you're overexaggerating and the only one using the terms 'amd slayer', lol. no one else called it that.
haswell's actually been delayed less times than steamroller and kaveri due to both amd and glofo.

Spoiler

1 - Did say KitGuru are biased
2 - Techheaven whomever it was dont know them but there results are better
3 - Xbitlabs contradicted there earlier results making a i3 outscore a i7 - Thats comical at best.
And here is why the APU scores are not wrong within margin of error, They are consistent with all bench sites, and APU's use catalyst drivers, i have 13.1 running and I honestly believe that catalyst drivers are better than the 10% crap that comes out of Intelmaginationland.

huh, must have missed with the edits.
did you think that there might be reasons why core i3's hd4k performs better than core i7's other than improved drivers? i'll leave that to you to figure it out.
how are the apu gaming performance consistent if sites use different settings? or is it because only the cases where 7660d has lead are relevant?
haswell's gt2, at least for desktop, will be available for all dt skus if toms' preview was right. so the 'intel's top dt igpu costs $350' excuse can be laid to rest. i just posted a bench showing a core i3, similarly priced as an a10, marginally beating it in one bench, didn't i? who knows if more of those come out in the future.... ;D
and in mobile, gt3 and gt3e will be available on more skus - that's where the amd will compete against intel.
p.s. - the questions in this post are rhetorical.

 

[sarinaide]

Spoiler

Apparently they cannot get their own benches right, so they inflate i3 performances higher than the i7 and get busted.

Now De5_Roy is harbouring thoughts that Haswell's HD is exceptional,

Spoiler

hell this was the part that was heralded the AMD slayer a year ago, it was playing Skyrim at 1080 resolution with Ultra settings, fast forward a year, haswell has been delayed twice, the leaks have shown nowhere near what it was puffing and ES samples back it up, I think ironically Haswell has become the intel iGPU slayer, twice the resources for moderate gains.

Broadwell, motherwell, cant you tell, oh well, sky lake, bake a cake Intel and iGPU is just not fun. I can't wait until they hit cell phones territory, I will need to walk around with a portable charger kit to keep that puppy charged.

don't kill the messenger.
i assumed haswell's igpu would be capable of being 'exceptional'(depends on how you use the word) based on the previews and leaks. i think both intel and amd future igpus have potential but in both cases i'd wait for believable reviews.
once again, you're overexaggerating and the only one using the terms 'amd slayer', lol. no one else called it that.
haswell's actually been delayed less times than steamroller and kaveri due to both amd and glofo.

You didn't call it a AMD slayer but there were people that honestly believed the first leaks, and basically from a reviewer whom I trust for insight basically told me a year ago Haswell is impressive from a intel stand point but you would not trade up a Llano HD6550 let alone a Trinity HD7660D for any Haswell iGPU because Trinity was still going to be substantially faster, at best it would match Llano which in some instances it does but Llano is around 33% slower than Trinity, with Richland adding another 15-20% just widens the gap.

Haswell has so far not delivered any of the performance touted x86 or iGPU wise, and soon power numbers will be out and I don't think they will be that impressive by intel standards. Yet there is abject belief in Intel leaks but when there is AMD fact ie the Trinity has been out for nearly a year and there are cold hard numbers yet its met with abject disbelief, it is often called inferior or the numbers and experience are fake. I would love to send you my APU parts or just borrow you a APU test bed so you can experience what something you don't seem to want to believe, it will leave you impressed, in fact you would never bother to even regard a i3 with HD4000 anywhere in the same zip code as a cheaper A10 because the truth is no matter how high the intel numbers are fabrication or truth what is cold hard truth and TR did it on latency test is that HD experience is just not in the league of AMD nor should it be hinted at being.

 

[JAYDEEJOHN]

"did you think that there might be reasons why core i3's hd4k performs better than core i7's other than improved drivers? i'll leave that to you to figure it out."
But, as we go larger, despite process, we lose the smaller, aimed more at gpu divy of power. Or more efficient, as in GCN or latter iterations, coming soon.
As well as catch up from better power resourcing, and also better cpu push/IPC/PPC whatever you want to call it.

If removing power from cpu to sacrifice to gpu, when most of the low hanging fruit has been taken up for power management, against someone who already has the gpu edge, is incurring power management benefits, and has better gfx solutions, I dont see it, unless what weve seen so far is garbage, and HSW is much better graphically than what weve seen in reality, and all without the true testing , in depth testing.
When it comes to this real true testing, all Im saying is, this is the future, or a huge part of it, SFF and mobile gfx capability, as seen by Intels very approach, where higher resolutions show them failing miserably, the tech sites better wise up and start treating this market segment accordingly

 

[de5_Roy]

Spoiler

You didn't call it a AMD slayer but there were people that honestly believed the first leaks, and basically from a reviewer whom I trust for insight basically told me a year ago Haswell is impressive from a intel stand point but you would not trade up a Llano HD6550 let alone a Trinity HD7660D for any Haswell iGPU because Trinity was still going to be substantially faster, at best it would match Llano which in some instances it does but Llano is around 33% slower than Trinity, with Richland adding another 15-20% just widens the gap.

Haswell has so far not delivered any of the performance touted x86 or iGPU wise, and soon power numbers will be out and I don't think they will be that impressive by intel standards. Yet there is abject belief in Intel leaks but when there is AMD fact ie the Trinity has been out for nearly a year and there are cold hard numbers yet its met with abject disbelief, it is often called inferior or the numbers and experience are fake. I would love to send you my APU parts or just borrow you a APU test bed so you can experience what something you don't seem to want to believe, it will leave you impressed, in fact you would never bother to even regard a i3 with HD4000 anywhere in the same zip code as a cheaper A10 because the truth is no matter how high the intel numbers are fabrication or truth what is cold hard truth and TR did it on latency test is that HD experience is just not in the league of AMD nor should it be hinted at being.

i think....i can test hd4k and apus but my main reason for not going amd was simple - power consumption. personal preference. i also considered apu-powered laptop but when i went digging into price, power numbers and software support, changed my mind.
frame time should be measured as well, but not all sites have adopted this type of method. i'd like toms to include frame time variance in their future igpu tests, those may be able to paint a better picture of igpu performance.

 

[jdwii]

Crap x 2 = 2ce the crap

Intel HD graphics are terrible... I could care less what anyone else says... You know no one is going to be honest on graphics performance... I'd rather have a single A10 APU (heck, even an A8) for gaming than a single i7 4770k for gaming.

That power issue that intel is have, I'm pretty sure its true... have you seen how many power phases those friggin z87 boards have?!?

Amd's last major mess up was Bulldozer and the constant delays of parts now they're pretty good on releases(getting rid of that crappy Globalfoundries helped i'm sure), With Intel its this and Sandy i think.

 

[de5_Roy]

But, as we go larger, despite process, we lose the smaller, aimed more at gpu divy of power. Or more efficient, as in GCN or latter iterations, coming soon.
As well as catch up from better power resourcing, and also better cpu push/IPC/PPC whatever you want to call it.

Spoiler

If removing power from cpu to sacrifice to gpu, when most of the low hanging fruit has been taken up for power management, against someone who already has the gpu edge, is incurring power management benefits, and has better gfx solutions, I dont see it, unless what weve seen so far is garbage, and HSW is much better graphically than what weve seen in reality, and all without the true testing , in depth testing.
When it comes to this real true testing, all Im saying is, this is the future, or a huge part of it, SFF and mobile gfx capability, as seen by Intels very approach, where higher resolutions show them failing miserably, the tech sites better wise up and start treating this market segment accordingly

i couldnt understand the first part. hope i didn't get it wrong.
imo neither trinity nor ivy bridge have a significant power efficiency advantage against their respective predecessors. afaik, intel's igpu causes power drain (in laptops) when the cpu has to do minor tasks and vice versa due to llc being connected to both. trinity doesn't have good turbo boosting, it's igpu doesn't have good power management either. at least the turbo is being addressed with richland. amd doesn't have power management benefits, just more aggressive throttling. intel's software inhibiting the igpus' raw performance is true. even if there was some kind of in depth testing, intel's solution will be handicapped by intel's drivers - this has been said many times before and is proven. it's due to intel's lack of expertise and experience in gpu field. i think the reason intel sometimes can come close to amd's performance is due to amd's negligence (also proven). amd's igpu performance is also held back for various reasons, one of them being yearly cadence that amd follows doesn't allow the drivers and software utilities to mature before being phased out. another being driver performance not benefitting entry level as much as it does high end.

 

[jdwii]

http://www.kitguru.net/components/graphic-cards/zardon/asus-f2a85-v-pro-amd-a10-5800k-w-hd7660d/21/

True APU result, Intel HD made worse than it is - AMD bias

am i supposed to believe an amd biased site now? neither kitguru nor hardwareheaven provide intel/amd gfx driver versions (or any driver versions), windows 8, updated softwares. xbitlabs' test systems actually have higher bw ram not underspecced like the other two's. in your own posted links, hd4k can keep up with 7660d rather well in borderlands 2, f1 2012 and dota benches either in min. fps or avg. fps or both. so the xbitlabs' f1 2012 @1080p lowQ is not that far fetched. shows that intel can make improvements as well, even though kg and hh both used more intense gfx settings.

Spoiler

http://www.hardwareheaven.com/reviews/1583/pg9/amd-a10-5800k-apu-performance-review-gaming-performance.html

Better results but not accurate.

Here is the absolute kicker, this has basically made me view xBit as a micky mouse site, these results are about right and we have tested the 2012 results with i7, i5 and 13 HD4000 and all APU's and the results at Ultra settings at 1080P with 2xMSAA gets you around 39-43 Average FPS on a HD7660D and around 21-24 on HD4000 (regardless of the CPU) basically xBit fucked up there own review.

Complete and utter balls up. If they got F1 2012 with a in game Benchmark tool wrong I shudder to even contemplate how wrong everything else is. Since F1 2012 is a game I play very often and part of Apex racing league I can record my game for you, record the frames and frame times, record max, min and average, show you the game is on ultra settings on the apu where intel including haswell is barely playable on medium at 1080.

rofl. reminds me that time when techreport fell off amd fanboy favor when they used cinebench to test intel and amd cpus....
if xbitlabs !@#$ed up their own review, the other benches where 7660d leads might be invalid as well.

The sites i use when i first look at a new product are often toms hardware,anandtech(and i don't care they are biased as well) and techpowerup if they have a review. then hardwarecanucks, overclockers club

i've never cared for Xbitlabs they influenced their results often by only choosing certain benchmarks or games same as others. Its like somone only reviewing Bulldozer on handbrake and 7-zip of course its going to look good.

 

[Blandge]

I still don't see why everybody thinks Anandtech is so bias towards Intel.

 

[bobbybamf12]

Okay this is how its going to go down. AMD going to say that steamroller going to get 30% performance increased. Then when the product comes out you will see that they just up the clock speed and tweak some things. AMD lost all their credibility after bulldozer and they shouldn't even put a percentage on how much better steamroller is then piledriver. All i care about is ipc! I couldn't careless about a cpu hitting 4.5ghz stock. I can overclock the cpu myself, of course it would be sweet hitting 4.5ghz on 125 watts but still won't make the cpu worth it. Hopefully there focus is on make a huge jump in ipc rather then adding more mhz to clock speeds cause that will only get them so far.

 

[griptwister]

Okay this is how its going to go down. AMD going to say that steamroller going to get 30% performance increased. Then when the product comes out you will see that they just up the clock speed and tweak some things. AMD lost all their credibility after bulldozer and they shouldn't even put a percentage on how much better steamroller is then piledriver. All i care about is ipc! I couldn't careless about a cpu hitting 4.5ghz stock. I can overclock the cpu myself, of course it would be sweet hitting 4.5ghz on 125 watts but still won't make the cpu worth it. Hopefully there focus is on make a huge jump in ipc rather then adding more mhz to clock speeds cause that will only get them so far.

o' rlly? What is the base for your predictions? Keep in mind there is a new process and there is a new architecture and they probably want to compete with Intel on single threading because that's where they have to improve. Keep in mind AMD has a smart group of people working with them. They recognized their issues with single GPUs and were able to fix those, who knows, they might fix their single threaded performance due to high demand. That's the only thing holding them back.