AMD Ryzen (Summit Ridge) Benchmarks Thread (use new thread)

[CatMerc]

How can a product, which delivers more than it originally promised be underperforming?
Zen, like every other x86 design has it's weaknesses and compromises but still?

Don't you already know the exact performance of Ryzen?

 

[Glo.]

https://www.cpubenchmark.net/high_end_cpus.html

According to this: 6900K 17000 pts in PassMark. 5960X 15.4K pts. in PassMark. Ryzen Sample: 15000 pts.

But yes, you have to value also difference in memory performance. Both Intel CPUs are quad channel, Ryzen Sample is dual channel and has slow memory in the platform.

 

[Agent-47]

Amd showed a blend test without numbers or anything just said they finished the test about the same time...such a reliable benchmark..

Need to wake up from whatever bubble you are in

Intel's blender test results were replicated by third-party reviewers using the parameters provided by AMD.

 

[Agent-47]

https://www.cpubenchmark.net/high_end_cpus.html

According to this: 6900K 17000 pts in PassMark. 5960X 15.4K pts. in PassMark. Ryzen Sample: 15000 pts.

But yes, you have to value also difference in memory performance. Both Intel CPUs are quad channel, Ryzen Sample is dual channel and has slow memory in the platform.

are you referring to https://s28.postimg.org/fp76snv3h/MWSnap_2017_02_10_21_18_27.jpg

It shows a Pentium beating out a SL at lower blocks.. enough said
These benchmarks needs to be carried out by the same reviewer for a proper comparision, otherwise you will get to a case where 7700k at 5G having the same ST performance as a 5820K at 4.8G

<redacted "benchmarks without a source">
Markfw
Anandtech Moderator

 

[lobz]

Who told you that Ryzen can beat the i7 6900k?
Amd showed a blend test without numbers or anything just said they finished the test about the same time...such a reliable benchmark..

everything outside Amd own benchmarks points Ryzen underperforming badly...Canard benchmarks, sandra sisoftware benchmarks etc

e-tards should be your show title on comedy central

Insulting other members is not allowed.
Markfw
Anandtech Moderator

 

[lobz]

https://www.cpubenchmark.net/high_end_cpus.html

According to this: 6900K 17000 pts in PassMark. 5960X 15.4K pts. in PassMark. Ryzen Sample: 15000 pts.

But yes, you have to value also difference in memory performance. Both Intel CPUs are quad channel, Ryzen Sample is dual channel and has slow memory in the platform.

which will have even much less than marginal impact on real world performance, let's all not forget that when we compare total passmark scores

 

[Det0x]

<redacted "benchmarks without a source"
Markfw
Anandtech Moderator

Holy sh** o_0

This looks very impressive, especially when we keep in mind that some of these other "baselines" were running at or close to ~ 5ghz while the engineering sample Ryzen was at ~ 3.4-3.8ghz with slow memory

 

[majord]

Can't be real. Wen know Ryzen can beat 6900k in MT.
So to loose this badly in St but beat in MT, AMDs SMT implementation must be so much superior than Intel's that it just impossible.

I'd just like to point out that higher SMT yield (that is, % extra performance SMT vs No SMT) doesn't actually indicate a 'superior implementation' by default. More typically indicates the design is more throughput orientated.

 

[zinfamous]

6950X outscores Ryzen by 30% in multithreaded. A complete stomp.

Call the press! An $1800 CPU with 2 more cores and 4 more threads utterly stomps a $500 CPU! Someone think of the children!

Maybe your math is different, but for me, 30% greater performance for ~320% greater cost is a failed argument of support.

 

[zinfamous]

sm625, after so many years of reading, I had to register to write this down, how incredibly epic of a fail this was, what you've done with these benchmarks.
according to all the numbers you've provided - against all your intentions - Ryzen is one of the greatest achievements of AMD to date, being financially constrained, on an inferior process with smaller cores, well.... WOW at these scores!!! Yep, even if it's actually @ 3.8

You are the greatest lurker of all times!

consider that AMD was operating at less than 1/10th the R&D budget of Intel--and that includes their RTG sector, if I'm not mistaken. Granted, one can honestly consider the work on BD/PD as sort of the foundation of Zen.

 

[vissarix]

LMAO at the people going on full damage controll...why be worried guys? Ryzen is a clear success..at 3.8ghz it has almost the same ST of an i3 4160 running at 3.6ghz..

 

[realibrad]

Call the press! An $1800 CPU with 2 more cores and 4 more threads utterly stomps a $500 CPU! Someone think of the children!

Maybe your math is different, but for me, 30% greater performance for ~320% greater cost is a failed argument of support.

That's worse price scaling than some of the women I have dated.

 

[zinfamous]

Why is sm625 calling it a turd? At those clockspeeds thats good results by ryzen.

alternative facts are the new order. Better join up, comrade.

 

[Agent-47]

I have taken the liberty to normalize the passmark results. the ST, FP and INT results puts ryzen (assuming 3.8G) JUST south of SL/KL. but the PN and Physics does not add up. Even there is too much variance among intel CPUs (??) illustrating the need to the benches to be done by the same reviewer.

EDIT: ignoring the variance in PN and Physics, given that the ST/FP/INT performance of ryzen is virtually identical, it can say that SMT in AMD is not too awesome in physics and PN. I donot think passmark uses AVX2.

 

[DrMrLordX]

If everything goes according to rumour, in a month we will know.

Sooner than that, hard launch is on the 28th. Just 17 more days . . .

edit: agreed that the Passmark data as displayed is basically useless.

 

[zinfamous]

LMAO at the people going on full damage controll...why be worried guys? Ryzen is a clear success..at 3.8ghz it has almost the same ST of an i3 4160 running at 3.6ghz..

I haven't yet seen a benchmark of Ryzen running @ 3.8ghz

 

[Glo.]

https://videocardz.com/65825/first-amd-ryzen-7-1700x-benchmarks-are-here
VideoCardz take on the topic of benchmarks. Single threaded performance is exactly on level of Haswell/Broadwell.

 

[Asterox]

Holy cow, that is bad bad bad in writes and latency department.

I am convinced single threaded passmark was certainly running at 3.8Ghz. Do not know about multi threaded tests though, but it should be enough to establish that rest of tests were not running on higher clock than that.

Where is your absolute proof, be realistic you or we dont now any accurate details.

This is ES Zen CPU, Turbocore is most likely not working the right way, therefore excluded from normal use.

Passmark Single Core

- i7 2600K=TurboCore up to 3.8ghz, score 1920

- ES Zen for now logically stick up for 3.4ghz, Score 2046

 

[lobz]

LMAO at the people going on full damage controll...why be worried guys? Ryzen is a clear success..at 3.8ghz it has almost the same ST of an i3 4160 running at 3.6ghz..

you have no idea if it was 3.8 or turbo-disabled 3.4

 

[Jan Olšan]

I recall something about The Stilt stating that the turbo of Ryzen is fully controlled by SMU and the "outside" only sees the base clock, which prevents stuff like motherboard forcing max turbo for all cores.

So with that, maybe it is possible that the turbo above 3,4 GHz was actually used by those Passmark benches, even though the program didn't detect it for the above reason.

Edit: this sample is 1700X anyway, 1800X should still be some 5-6% faster.

 

[jpiniero]

you have no idea if it was 3.8 or turbo-disabled 3.4

I was wondering if you could tell if XFR was enabled.

I looked at the Sandra result and compared it to a random 6900K score that seemed like it had turbo disabled (3.2 ghz):

MM Integer: 718 vs 525 for the Ryzen "38_34"
MM Long Int: 211 vs 159
MM Quad Int: 3299 vs 3589
MM Single Float: 654 vs 522
MM Double Float: 374 vs 300
MM Quad Float: 13352 vs 13271

Not sure what to make of it beyond perhaps some have AVX(2) optimizations and some don't.

 

[CentroX]

Memory is running on 2400 mhz on a 300 motherboard.

 

[ultima_trev]

Videocardz has displayed CPU passmark scores as well albeit he throws in his stock i7 6800K.

Asides from prime and physics (which I guess are AVX heavy), Ryzen seems to be only slightly behind Haswell-E. It's essentially an i7 5960X with half the memory channels and AVX performance, which probably won't benefits mainstream applications for a while.

https://videocardz.com/65825/first-amd-ryzen-7-1700x-benchmarks-are-here

Good job, AMD!

 

[lolfail9001]

Can anyone with BCLK overclocked Skylake chime in? BCLK OC bios should make AVX on Skylake work at same speed as Ryzen theoretically does (never switch into 256 bit mode, that is), so it could help to verify that PN/Physics test are actually AVX based and there's nothing else fishy going on with them.

 

[Arachnotronic]

Can anyone with BCLK overclocked Skylake chime in? BCLK OC bios should make AVX on Skylake work at same speed as Ryzen theoretically does (never switch into 256 bit mode, that is), so it could help to verify that PN/Physics test are actually AVX based and there's nothing else fishy going on with them.

https://www.cpubenchmark.net/cpu_test_info.html

Physics Test
The Physics Test uses the Bullet Physics Engine to perform a benchmark of how fast the CPU can simulate the physics interactions. The test repeats the first several seconds of the simulation as many times as possible within the test duration.

I believe the Bullet Physics Engine uses AVX2.