AMD CPU speculation... and expert conjecture

[noob2222]

In other news. Upcoming x86 opterons doing the heavy lifting for mixed cpu servers spells ARM cores are weaker and x86 is not going anywhere soon.

Will be interesting to see how the os handles mixed architecture servers.

http://www.fudzilla.com/home/item/33401-amd-sheds-light-on-arm-based-seattle-part

 

[juanrga]

jdwii, I didn't add 30% to each benchmark and I am sorry you don't understand the difference between an average and a maximum value.

de5_Roy It was mentioned a pair of pages ago that next year (2014) start 20nm volume production and that 16nm FINFET chips had been tapped-out recently, therefore risk production was imminent. The first tests of 10nm bulk FINFET are being made.

noob2222, yes that was a ES kaveri, but no kaveri is not 37% slower than BD. That was your silly claim based in your own ignorance about what was tested and how was tested. I already corrected your misunderstanding two or three times before, but you insist.

See your edit just now. Your original silly claim was that a 200-250W chip requires LN2 for cooling, now you changed it to FX-9000 requires liquid watter for cooling. SSShhhh! Don't say to FX-9590 owners who are running at 5.3GHz on air.

Your claim AMD only ships FX-900 with watter colling is another of your lies!

FX-9590 with water colling kit: http://www.newegg.com/Product/Product.aspx?Item=N82E16819113351

Without it: http://www.newegg.com/Product/Product.aspx?Item=N82E16819113347

 

[jdwii]

"jdwii, I didn't add 30% to each benchmark and I am sorry you don't understand the difference between an average and a maximum value."

I'm sorry that now your trying to convince the readers at tomshardware that you never said 30% more performance in CPU tests. When a I5 2500K is easily 30% stronger then a A10 6800K and you where stating 30% more performance(not stating that its a up to number while putting down more realistic numbers) this means the future APU not on level with the 2500K like you mentioned besides instead of those certain circumstances stances. Also to state such a statement such as the next gen APU being on level with the I5 2500K you have to assume that means the majority of the time not in rare situations


"noob2222, yes that was a ES kaveri, but no kaveri is not 37% slower than BD. That was your silly claim based in your own ignorance about what was tested and how was tested. I already corrected your misunderstanding two or three times before, but you insist." And again your claim was that the next gen APU would be on level with the I5 2500K and you indeed did this by adding 30% to certain benchmarks which is nothing but a maximum gain(which should never be used over averages unless you have some kind of agenda)

Also i would like Noob2222 to say exactly what he meant by that statement i doubt he would ever say such a statement on purpose he has been on this site for quite some time(even the BD fourms i think), i'm sure he meant overall performance such as all 8 cores being used heavily and then in those cases i can see the 8150fx being 20-30% faster.

But instead of you using actual claims and being rational you resort to being hostile and using common fallacy in your argument would almost rather see hafijur return

 

[jdwii]

"Your claim AMD only ships FX-900 with watter colling is another of your lies!"
Or quite possibly he meant it by accident since Amd used to sell those as only water cooled(being more realistic on cooling and probably needed for reliability

Again you only look at one side of things another fallacy in your argument

Edit
Also i just now noticed that your statement was about the LN2 and i'm sure he meant that with the 250 Watt APU which was a absurd comment. However your statement calling him a liar(or his lies) is simply wrong and again a fallacy by which you attack the person. He was wrong not a liar.

I think the sky is gray does that make me a liar? No it makes me misinformed

 

[juanrga]

In other news. Upcoming x86 opterons doing the heavy lifting for mixed cpu servers spells ARM cores are weaker and x86 is not going anywhere soon.

Will be interesting to see how the os handles mixed architecture servers.

http://www.fudzilla.com/home/item/33401-amd-sheds-light-on-arm-based-seattle-part

LOL! After complaining here by discussing ARM stuff, now you post about ARM!

The funny part is that Fudzilla is making up things. Their source is listed at the bottom part and it is one of AMD blogs.

http://community.amd.com/community/amd-blogs/amd-business/blog/2013/12/12/amd-s-64-bit-seattle-arm-processor-brings-best-of-breed-hardware-and-software-to-the-datacenter

I copy and paste the relevant part about mixed x86-ARM racks from AMD:

All of this work on both hardware and software should shed some light on just how big ARM processors will be in the data center. AMD, an established enterprise semiconductor vendor, is uniquely placed to ship both 64-bit ARMv8 and 64-bit x86 processors that enable “mixed rack” environments. And thanks to the army of software engineers at AMD, as well as others around the world who have committed significant time and effort, the software ecosystem will be there to support these revolutionary processors. 2014 is set to see the biggest disruption in the data center in over a decade, with AMD again at the center of it.

Where it says anything about ARM cores being weaker? Where it says that x86 is required for "heavy lifting"? Nowhere. Mixed racks will be useful for doing the transition from x86 to ARM. Legacy software will run better in x86 chips, whereas new software is compiled to ARM.

In fact, the blog mentions that new ARM cores are high-performance and that AMD expect to conquer the server market with them:

Processors based on ARM’s 64-bit ARMv8 architecture will start to appear next year, and just like the x86 AMD Opteron™ processors a decade ago, AMD’s ARM 64-bit processors will offer enterprises a viable option for efficiently handling vast amounts of data.

The AMD Opteron processor came at a time when x86 processors were seen by many as silicon that could only power personal computers, with specialized processors running on architectures such as SPARC™ and Power™ being the ones that were handling server workloads. Back in 2003, the AMD Opteron processor did more than just offer another option, it made the x86 architecture a viable contender in the server market - showing that processors based on x86 architectures could compete effectively against established architectures. Thanks in no small part to the AMD Opteron processor, today the majority of servers shipped run x86 processors.

In 2014, AMD will once again disrupt the datacenter as x86 processors will be joined by those that make use of ARM’s 64-bit architecture. Codenamed “Seattle,” AMD’s first ARM-based Opteron processor will use the ARMv8 architecture, offering low-power processing in the fast growing dense server space.

To appreciate what the first ARM-based AMD Opteron processor is designed to deliver to those wanting to deploy racks of servers, it is important to realize that the ARMv8 architecture offers a clean slate on which to build both hardware and software.

ARM’s ARMv8 architecture is much more than a doubling of word-length from previous generation ARMv7 architecture: it has been designed from the ground-up to provide higher performance while retaining the trademark power efficiencies that everyone has come to expect from the ARM architecture. AMD’s “Seattle” processors will have either four or eight cores, packing server-grade features such as support for up to 128 GB of ECC memory, and integrated 10Gb/sec of Ethernet connectivity with AMD’s revolutionary Freedom™ fabric, designed to cater for dense compute systems.

But those are standard A57 cores. The true race will start with custom ARM64 cores, such as those developed by Nvidia, Google...

 

[juanrga]

"jdwii, I didn't add 30% to each benchmark and I am sorry you don't understand the difference between an average and a maximum value."

I'm sorry that now your trying to convince the readers at tomshardware that you never said 30% more performance in CPU tests. When a I5 2500K is easily 30% stronger then a A10 6800K and you where stating 30% more performance(not stating that its a up to number while putting down more realistic numbers) this means the future APU not on level with the 2500K like you mentioned besides instead of those certain circumstances stances. Also to state such a statement such as the next gen APU being on level with the I5 2500K you have to assume that means the majority of the time not in rare situations

One thing is what you don't understand what is being said/discussed, another very different thing is that you pretend to blatantly lie about what I said. This is the abstract of my article about Kaveri CPU performance:

Abstract: Chip-maker AMD will release a new family of APUs by the end of 2013. This new family will be named Kaveri and will substitute the current Richland family of APUs. In this article I will consider Kaveri performance using all the information that AMD has disclosed as well as my best guesses on the parts that remain unknown. I estimate that the the top Kaveri GPU will have a CPU clocked at 4 GHz with a maximum performance of 128 GFLOP and a GPU clocked at 900 MHz with 922 GFLOP, giving a total of 1050 GFLOP for the whole APU. Combining all this data, I predict that the CPU of the top Kaveri APU will be about 26% faster than top Trinity APU and about 17% faster than top Richland APU. This would put the multi-threaded performance of the CPU of the new quad core Kaveri APU at the same level than an Intel quad core i5 or a six-core AMD FX with traditional software. The new Kaveri APU will show its real strength with HSA software, which will exploit the performance of both the CPU and the GPU. With HSA enabled software, Kaveri has the potential to be much faster than an Intel i7 or an octo-core AMD FX. Some developers are finding accelerations of up to 500% when enabling HSA. A collection of APU and CPU benchmarks and scores is given.

http://www.brightsideofnews.com/news/2013/11/4/what-to-expect-from-kaveri-a-detailed-predictive-analysis.aspx

http://juanrga.com/en/AMD-kaveri-benchmark.html

I never said 30% faster than A10-6800k Richland APU. You did!

In some tests, Kaveri will be as fast as 2500k

In other tests Kaveri will be slower than a 2500k

And in another tests Kaveri will be faster than a 2500k

 

[The Q6660 Inside]

All they did was hand out an R9 BF4 bundle, $500 voucher on next gen consoles and a laptop.. QUIT STALLING ALREADY 😛

 

[noob2222]

@juan

My own ignorance by downloading the cosmology benchmark and running it on my 8120 at 2.1 ghz and scoring 5000+ pts vs that ES only scoring 3300? A benchmark that scores the same with 2, 4, or 8 cores enabled and the results themselves sais "per core". Yet "somehow" I am wrong because of something you said. Run the program yourself you tool. Its free to download and you will see you are quite the liar on this one.

Either that ES is not kaveri or its roughly 37% slower than my 8120. You still actually believe it is kaveri so you have your own answer.

As for your crusade to promote ARM.

Lawrence Latif is a blogger and technical communications representative at AMD.His postings are his own opinions and may not represent AMD’s positions, strategies or opinions.

And the comment on custom cores are going to be superior to AMD's own base core really makes it an awesome idea for AMD to abandon x86. Great so in otherwords your saying AMD will have the slowest ARM and x86 processors on the market and at the same time claim 20% of the market share for themselves.

 

[etayorius]

Juanrga, any idea if Kavery will support CrossfireX with the APU GPU and say... a R9 270X? I assume this may only be available in MANTLE supported games.

 

[Ags1]

Mantle specifically supports assymetric multi-GPU configurations. The scenario I've heard quoted is doing totally different kinds of processing on the iGPU and dGPU.

 

[juanrga]

noob2222,

The problem is not that you have no idea of what was measured and how. The problem is that you are corrected again and again and again, but you insist on spreading the same nonsensical "Kaveri will be 37% slower than Bulldozer" FUD.

Regarding AMD servers. That disclaimer that you copy-paste is a standard legal disclaimer that appears in any blog from AMD. Pay attention to this entry in the same bussines blog

http://community.amd.com/community/amd-blogs/amd-business/blog/2013/10/23/the-trend-towards-integration-of-industrial-devices

or this one from the gaming blog

http://community.amd.com/community/amd-blogs/amd-gaming/blog/2013/10/17/the-four-core-principles-of-amd-s-mantle

The first was written by the AMD director of AMD Embedded Solutions division, and the second by the AMD director of ISV Gaming and Alliances at AMD, but both contains the same legal disclaimer "His postings are his own opinions and may not represent AMD’s positions, strategies or opinions..." :sarcasm:

Your unability to understand a standard legal disclaimer is not the problem here. The problem is that you used a news to support your idiotic rant, but when the source (AMD) of the news is saying exactly the contrary to what you believed, then you pretend now that the source (AMD) is not valid anymore. Hypocrisy at its best!

 

[juanrga]

tourist, both configurations: dGPU+dGPU and GPU+iGPU

Juanrga, any idea if Kavery will support CrossfireX with the APU GPU and say... a R9 270X? I assume this may only be available in MANTLE supported games.

Unfortunately, I cannot sure you anything, but some time ago some benchmarks were leaked in the Internet with some new sites believing that Kaveri was a 13CU part. As I said in my article it was a hybrid crossfire configuration with a discrete Radeon R-series. The leaked benchmarks didn't use MANTLE. Again take this with a grain of salt. Maybe you want wait until full details are revealed at CES 2014.

 

[jdwii]

Kaveri sounds perfect for my future mini-ITX build. Can anyone reliably say what kind of GPU performance to expect, relative to current cards? I'm hoping at the very least it's more powerful than my laptop's GTX 560m

Kaveri iGPU graphics performance will be between a HD7750 DDR3 and a HD7750 GDDR5.

Kaveri iGPU compute performance will be better thanks to hUMA.

We know Kaveri will have greater IPC compared to Phenom, but how about actual Multithreaded performance? how much faster will Kaveri be compared to Phenom in this regard?

If Kaveri can Match or get very close to the i5 2500k performance without HSA/MANTLE, it will be a win... if it´s 10% slower or more to the i5 2500k then it will be a fail.

I really want to get Kaveri and pair it with a R9 270X or 280X, if it can get very very close to the i5 performance, with a 10-15% OC should be quite good.

In my BSN* article I predicted Kaveri would match an i5 2500k without HSA/MANTLE. Since, I got benchmark data of a sample of real Kaveri compared to a 2500k. I am still analyzing the data, but at this time I can say that the data seems to confirm that Kaveri matches an i5 2500k without HSA/MANTLE.

I will write an update to my article soon.

"adding 30% to every benchmark around and claiming its "RL results" How often is any cpu intel or amd improve a set amount across the board? "
Not the only one who noticed it

Juan is going to be VERY disappointed. Its almost going to be fun watching him explain this away. I mean, seriously, I foresee him getting run out of here faster then JF-AMD was post BD...

Seems like this happens several times before a new product is released you have over hype and then disappointment. We had someone from the BD saying 50% more performance compared to the X6, Now we have someone stating as fast as a 2500K even though a 2500K is faster by 35% on average and even using upto statements almost twice as fast as a 5800K(which is a good 7% slower compared to a 6800K)

 

[jdwii]

http://www.anandtech.com/bench/product/675?vs=288

Also when i looked at all of those test's and did percent differences it came to the fact that under games the 2500K was 47~ Faster in gaming benchmarks and the FP unit has not been improved much under steamroller and the next gen APU will be clocked lower then even the 5800K in the comparison.

Now we Look at CPU intensive comparisons and we get 45% less performance with the 5800K compared to the I5 2500K.

So again except in those certain situations the 2500K will remain faster on average compared to the next gen APU in CPU oriented tasks. Unless Amd can improve performance by 45%.
Man that took me a long time to do the math from all those Benchmarks and their still only approximate

I remember when Amd picked those certain situations and bragged about a item before.

 

[gman97005]

Hi guys.
Take that:
http://cdn3.wccftech.com/wp-content/uploads/2013/12/AMD+Desktop+Q1+2014+_VTB_Page_20.jpg
And that:
http://www.benchmark.pl/uploads/backend_img/a/fotki_newsy/201306/PM/amd-apu-trinity-richland-intel-core-haswell-pcmark-porownanie-1.jpg

About ~10% better than Richland. Quite nice.

The fact that Kaveri does it with a lower clock speed is what catches my attention, there's some serious magic going on there and knowing the way that the Richland Core scales with fast RAM like DDR 2133 i can already imagine what Kaveri will do with faster clock speeds, i think the Intel fanboys are just jealous.
...

Intel Improves 10%, improves GPU = Intel Sucks
AMD Improves 10%, improves GPU = Quite Nice

I never said that Intel sucks. Also for Kaveri that's nice achievement, because it's 10% lower clocked than Richland.

And I know that PCMark is not good benchmark, but at least it's something.
Oh, we also have this: http://browser.primatelabs.com/geekbench3/223722

The fact that Kaverti does it with a lower clock speed than Richland is what catches my attention, i can see Kaveri scaling with
fast RAM just like Richland but better, it's a beast and i think the Intel fanboys are just jealous..

 

[noob2222]

@juan

You idiot. That cosmology ES is not kaveri. The only way to get the results of your system on the cosmology website is to run it through their program. Anyone on here can verify what I have said. You on the other hand insist on trying to insult me instead of pulling your head out of your back side and face the truth. That retarded "blog" took an 8 core benchmark, divided by 8 (the number of cores in 8xxx) then calculated the clock speed then multiplied by 4 to come up with their fake ass BD numbers.

they cut the BD and PD numbers in half in order to try and justify posting anything.

Don't try to tell me I'm clueless when you cant even offer anything other than your opinion.

http://www.cosmologyathome.org/show_host_detail.php?hostid=199535

48 core intel scoring 8181 integer pts.

4 core "kaveri" scoring 3311.

http://cosmologyathome.org/show_host_detail.php?hostid=187215

that means in your world if cosmology is a multicore benchmark, at 48 cores and 1.8 ghz kaveri will score 39,000+ or 4 times faster than a 2.7 ghz xeon. But according to your favorite blog kaveri is only 32% faster than PD not 4x xeon.

ITS FAKE, WHAT DONT YOU GET ABOUT THE TRUTH.

If I'm a liar. Prove it. Download cosmology and run your own tests.

http://www.cosmologyathome.org/

lets see your results, quit making up bs.

 

[Reynod]

noob play nice and stop biting the other inmates ... or your not coming on the shuttle up to Elysium to get those warts removed buddy.

 

[juanrga]

jdwii, You make me laugh by quoting anandtech database. Not only you missed my explanations of why it is not a reliable source, but you also misses the remarks made by others about it.

noob222, insulting me doesn't change the fact that you continue being plain wrong. The correction to your "Kaveri is 37% slower than Bulldozer" nonsense is the same than lots of pages ago. I will not spend time explaining to you the same again.

As another reader commented here, both of you are trying hard to make Kaveri look as the poor processor launch in the history... but Kaveri is a clear winner for AMD 😀

 

[Ags1]

Palladin, Yuka, Gamerk, 8350rocks and others on the thread who tried out my benchmark app, I have built a new version with an EXE launcher for windows. It lets me control the command line and ensure the app runs at high priority by default. Also, some of your CPUs were not identified correctly - this version (on windows of course) takes a second shot at the CPU name by reading the registry.

http://www.headline-benchmark.com/download.jsp

 

[Ranth]

As another reader commented here, both of you are trying hard to make Kaveri look as the poor processor launch in the history... but Kaveri is a clear winner for AMD 😀

I'd rather keep expectations low, that way you won't be massively disappointed when it "fails"(aka Not living up to the high expectations), and the chance of being pleasantly surprised is a lot higher...

 

[etayorius]

AnandTech tends to give 10-15% more performance to Intel CPUs, that`s fact.

 

[noob2222]

@juan

AMD only making low end cpus from here on out is the problem. Kaveri can't compete with the 6350 and 8xxx cpus other than in hsa enabled programs.

APU isn't a good replacement for HEDT. This would be like AMD only releasing duron cpus during their athlon days.

Tourist, your correct. Kaveri is a gpu primarily and a cpu as an afterthought. Its not going to be a 2500k. Juans numbers were only taken from the best available richland benchmarks, not the average, and in some cases not even realistic.

He didn't even include what his richland numbers were to show how fake his calculations are.

http://www.hitechlegion.com/reviews/processors/35202-amd-6800k-6700?start=5

http://www.hardwarecanucks.com/forum/hardware-canucks-reviews/61510-amd-richland-review-a10-6800k-a10-6700-benchmarked-11.html

http://techreport.com/review/24954/amd-a10-6800k-and-a10-6700-richland-apus-reviewed/9


X264. Add 15% and its still slower than the i5 2400 but "somehow" his is as fast as the 2500k.

http://www.phoronix.com/scan.php?page=article&item=intel_4770k_linux&num=3

john the ripper. You can see the intel numbers but how the fk did the 5800k go from 2456 to 4310 on kaveri. Thats 75% faster than trinity. Richland was only 7% faster than trinity. Where did 4310 come from. Who knows.

http://openbenchmarking.org/prospect/1306286-SO-AMDA1068015/8c6727bc7043bf3ed817729f76e07a9ba0e72153

 

[blackkstar]

http://postimg.org/image/b6fmqohpp/

Someone was asking for numbers? Found this, make no claims on the legitimacy of it. Take with salt, etc.

 

[jdwii]

jdwii, You make me laugh by quoting anandtech database. Not only you missed my explanations of why it is not a reliable source, but you also misses the remarks made by others about it.

noob222, insulting me doesn't change the fact that you continue being plain wrong. The correction to your "Kaveri is 37% slower than Bulldozer" nonsense is the same than lots of pages ago. I will not spend time explaining to you the same again.

As another reader commented here, both of you are trying hard to make Kaveri look as the poor processor launch in the history... but Kaveri is a clear winner for AMD 😀

First of all these are reliable benchmarks and actual real world benchmarks with real applications not theatrical performance dreams.
Second your the one insulting other's
Also again i'm being realistic(not magically adding 30% to everything) about performance me running on Amd hardware only i would love to see amazing things happen and they are actually but not in the way your describing.

 

[noob2222]

@blackstar

Impressive overclock if thats on air, though the 6800k has been seen 5.0-5.2 ghz. About 28% faster clock for clock than richland so the number looks possible.

Will be interesting to see if this is a best case benchmark.