Overclock.net - An Overclocking Community

Overclock.net - An Overclocking Community (https://www.overclock.net/forum/)
-   Hardware News (https://www.overclock.net/forum/225-hardware-news/)
-   -   [Tom's Hardware] Intel Brings Its Own Benchmark to Refute AMD's '2X' EPYC Claim (https://www.overclock.net/forum/225-hardware-news/1726782-toms-hardware-intel-brings-its-own-benchmark-refute-amds-2x-epyc-claim.html)

WannaBeOCer 05-31-2019 10:22 AM

[Tom's Hardware] Intel Brings Its Own Benchmark to Refute AMD's '2X' EPYC Claim
 
1 Attachment(s)
Source: https://www.tomshardware.com/news/in...eon,39544.html

Quote:

Intel is firing back at AMD CEO Lisa Su's Computex claims that the company's upcoming 7nm EPYC Rome data center processors are twice as fast as Intel's Xeon Scalable 8280 processors in a popular benchmark. But Intel says that AMD didn't configure its Intel test system correctly, and also didn't use the most relevant processors for comparison testing. Now Team Blue has released benchmarks to back its claims.

But Intel doesn't want to just be more competitive. It wants to prove that it will continue to lead even after AMD's 7nm Rome processors come to market. To that effect, Intel also included test results with its Xeon Platinum 9000-series (Cascade Lake-AP) that come armed with as many as 56 cores, 112 threads, and 12 memory channels crammed into a package that dissipates up to 400W. These new behemoths, which are essentially two Skylake-SP CPUs in a single socket, only come in OEM servers, so they aren't available on their own like AMD's Rome chips will be.

AMD did use a different compiler for its EPYC setup than it did for the Intel system, but that's expected. Intel's tuning did expose greater performance from its chips than AMD did, but that appears to be because AMD didn't use the optimizations. Intel's optimizations are available publicly on its website, meaning that AMD either wasn't aware of the optimizations or chose not to use them. Aside from the lack of optimizing the competitors' platform, which may be unintentional, it doesn't appear that AMD went out of the way to skew its test results.

tpi2007 05-31-2019 10:47 AM

So, they're saying:

1. "We don't lose by that much.";
2. "And when compared against the upcoming 2S 48C/96T system that's a non upgradeable, soldered to the board special OEM order we expect to tie (up to 1.5% higher) with a much more power hungry setup.";
3. "And if we dial it up to 11, our 2x400w 2S 56C/112T system that is also the same OEM affair can be up to 23% faster.".

I get why they have to say this, but the outcome is a little embarrassing.

rdr09 05-31-2019 10:53 AM

Is there a benchmark that will uncover which gets exploited the most?:)

looniam 05-31-2019 11:21 AM

in a totally unrelated note:


Hwgeek 05-31-2019 11:43 AM

Quote:

Originally Posted by tpi2007 (Post 27986970)
So, they're saying:

1. "We don't lose by that much.";
2. "And when compared against the upcoming 2S 48C/96T system that's a non upgradeable, soldered to the board special OEM order we expect to tie (up to 1.5% higher) with a much more power hungry setup.";
3. "And if we dial it up to 11, our 2x400w 2S 56C/112T system that is also the same OEM affair can be up to 23% faster.".

I get why they have to say this, but the outcome is a little embarrassing.

Funny part is that 2x400w 2S 56C/112T is actually 4s 28C to match 2s Rome :-).

Intel just made a bet that they could just invest in Fabs and keep continuing down-scaling their same monolithic design to make more CPU's per wafer, but 10nm showed them that they can't make good yields with their massive 28c design and they are stuck trying to fix the 10nm processes till this days, if they made a choice to move to chiplet design then they could compete AMD's Servers cheap more easily.
.

tpi2007 05-31-2019 12:03 PM

Quote:

Originally Posted by Hwgeek (Post 27987060)
Funny part is that 2x400w 2S 56C/112T is actually 4s 28C to match 2s Rome :-).

Intel just made a bet that they could just invest in Fabs and keep continuing down-scaling their same monolithic design to make more CPU's per wafer, but 10nm showed them that they can't make good yields with their massive 28c design and they are stuck trying to fix the 10nm processes till this days, if they made a choice to move to chiplet design then they could compete AMD's Servers cheap more easily.
.


Their response reminds me of Monty Python's dead parrot sketch, here in adapted form:

Quote:

Spoiler!


startekee 05-31-2019 12:08 PM

I am confused. Is this saying that Intel's 56c setup is faster than AMD's 64c?

WannaBeOCer 05-31-2019 12:28 PM

Quote:

Originally Posted by startekee (Post 27987100)
I am confused. Is this saying that Intel's 56c setup is faster than AMD's 64c?

With optimizations which probably Rome would also benefit from when looking at Zen's performance when running Clear Linux. Maybe in a few years we'll be able to see an optimized AMD distro.

https://www.phoronix.com/scan.php?pa...ux-10way&num=4

"Intel does not guarantee the availability, functionality, or effectiveness of any optimization on microprocessors not manufactured by Intel."

Hwgeek 05-31-2019 12:40 PM

Quote:

Originally Posted by tpi2007 (Post 27987094)
Their response reminds me of Monty Python's dead parrot sketch, here in adapted form:

You killed me with that, ROFL!.

cssorkinman 05-31-2019 12:41 PM

Wow things are really getting bad if Intel is complaining about a compiler that favors AMD..... ;)


All times are GMT -7. The time now is 06:58 AM.

Powered by vBulletin® Copyright ©2000 - 2019, Jelsoft Enterprises Ltd.

User Alert System provided by Advanced User Tagging (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
vBulletin Security provided by vBSecurity (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.

vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.