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

[ShintaiDK]

The CPU-Z benchmark was a hot topic here before. It might only reveal a rather small part of the whole picture.

The benchmark is what, 10 lines of code doing clock counting?

I am looking forward to we actually see something more relevant. GB4, CB15 etc.

 

[Dresdenboy]

The benchmark is what, 10 lines of code doing clock counting?

I am looking forward to we actually see something more relevant. GB4, CB15 etc.

It calculates Perlin noise. And from past analyses it emerged that it relies heavily on IDIV, favouring BD's many integer cores.

 

[Redentor]

The benchmark is what, 10 lines of code doing clock counting?

I am looking forward to we actually see something more relevant. GB4, CB15 etc.

Is GeekBench now a relevant benchmark? What!?

 

[MentalIlness]

http://www.guru3d.com/news-story/6-...marks-shows-50-faster-then-core-i5-7600k.html

The AMD Ryzen leaks and madness doesn't stop. Today some CPU-Z benchmark results have surfaced of what seems to be a six-core Ryzen 5 1600X processor. The 6 core and 12 threaded processor would have a base clock of 3.3 GHz with a 3.7 GHz Turbo. And the performance seems pretty nice.

Honestly I wasn't expecting to see a 6-core enabled part anytime soon as personally I am expecting an 8-core launch only.

If you look closely at the screenshots you can see summit ridge tagged, that's the real thing you guys. You cannot fake that unless you have master photoshop skills. The one problem that I have with these screenshots is that the two of them cannot be paired automatically together and thus the upper benchmark result set could even be an entirely different 6-core processor. Let's assume it's valid for now though, but please let's take things still with a little reserve and a grain of salt to be certain.

The single threaded (single core) performance of this Ryzen unit shows 1888 points. That is precisely on par with Core i7 6850K single core performance and that I did not expect (thus that is very positive). We know that the Intel Core i7-6850K clocks in at 6/12 (cores/threads) 3.6 GHz / 3.8 GHz (base clock / turbo clock). So compared 1:1 that is really fricken close on the per core performance. The Ryzen muli-threaded performance shows 12544 points, which is faster than a Core i7-6850K who reaches 10872 points in our own testing.

While it is not a fair comparison (4 versus 6 cores), we do expect these two processors to be priced in the same range and thus compared to a quad core Core i5 7600K the Ryzen R5 offers 50% more perf then the 8314 points that CPU scores. Obviously the 7600K has much faster turbo's hence that single threaded core performance locks in at a faster 2135 points.

Have a peek at the numbers, if all true then it is fairly okay to think that the IPC perf for AMD is spot, pretty darn good and close to comparable with what Intel is offering. And we wonder if THIS photo has anything to do with the above results shown.

 

[CentroX]

you are late to the party.

 

[Nothingness]

Between Canard PC, Passmark CPU Mark and 3DMark Firestrike Physics leaks it seems safe to say the $400 R7 1700X will offer 90% of the i7 6900K's performance overall. Zen seems to be very competent in Compression, Encryption, Integer Math, Floating Point while single thread is just a few percent behind Broadwell-E. Prime Numbers and Physics probably won't affect the overall too much as it's inline with Intel's high end quad cores.

If Prime numbers is indeed mostly testing IDIV instruction speed, then it definitely is a "don't care"

Is GeekBench now a relevant benchmark? What!?

GB4 is indeed now considered by many as relevant (at least much more than synthetic stuff from Sandra or CPU-Z).

 

[OrangeKhrush]

Incoming

Right on the Haswell:

i7 5960 @ 3.7Ghz = 148

 

[CatMerc]

Source?

 

[OrangeKhrush]

Source?

posted on OC.net, it was sourced to Baidu but the original link is down again NDA rules apply but again it seems to be a legitimate 6 core Ryzen benched and posting a score that is = to Haswell/Haswell E and about 9% off Broadwell.

 

[StinkyPinky]

I don't know about you guys, but I have bought my tickets.

Tickets to what you may ask? Why, to the hype train of course. First class. Choo choo, all aboooard.

 

[agouraki]

posted on OC.net, it was sourced to Baidu but the original link is down again NDA rules apply but again it seems to be a legitimate 6 core Ryzen benched and posting a score that is = to Haswell/Haswell E and about 9% off Broadwell.

it looks like its 4core/8thread for me on that task manager picture.

 

[revanchrist]

Incoming

Right on the Haswell:

i7 5960 @ 3.7Ghz = 148

Single Core: 146 cb
Multi core: 1130-1140 cb

Yup, big grain of salt.

 

[malitze]

it looks like its 4core/8thread for me on that task manager picture.

To me it looks like theres a two digit number (12) at the logical cores so should be a 6C/12T.

 

[Lovec1990]

it looks like its 4core/8thread for me on that task manager picture.

nope it looks like a 6c/12t

 

[agouraki]

nope it looks like a 6c/12t

you are right i was looking at the sockets/cores and thought its 4/8

 

[OrangeKhrush]

Single Core: 146 cb
Multi core: 1130-1140 cb

Yup, big grain of salt.

where do you get the MT score

 

[revanchrist]

where do you get the MT score

http://imgur.com/a/yFYSr

By stalking the post history of the OP from the Baidu Forum. He might have deleted his post, but his post history is still there.

 

[OrangeKhrush]

http://imgur.com/a/yFYSr

By stalking the post history of the OP from the Baidu Forum. He might have deleted his post, but his post history is still there.

It is possible, we calculated the SMT at 30% which is right on the money 1140/146 = 7.8 = 30% SMT

i7 5820K @ 3.7 = 1115

 

[Arachnotronic]

If Prime numbers is indeed mostly testing IDIV instruction speed, then it definitely is a "don't care"

The CPU-Z benchmark is pretty much worthless if you want to get a good idea of how robust your CPU is.

GB4 is indeed now considered by many as relevant (at least much more than synthetic stuff from Sandra or CPU-Z).

I have heard this from many people that I have talked to, and having read the GB4 whitepaper the workloads do seem pretty reasonable/interesting.

AMD also uses it and thinks it's great

 

[Lodix]

GB4 is now much more accurate than before.

 

[FIVR]

OrangeKhrush - "See guys, ignore the multitude of credible benchmarks showing Ryzen as competitive and focus on this cropped screenshot posted anonymously on chinese forums."


The rear guard action is strong with this one. You keep fighting that good fight, soldier. Never give up, even in the face of insurmountable odds.

 

[OrangeKhrush]

If my dependable source was correct.

146/153 =4.7% which is what he said about Ryzen to Broadwell its about another 5% onto that to SKL/KBL

 

[PhonakV30]

Intel Xeon X5650 specifications (6C/12T)
http://www.cpu-world.com/CPUs/Xeon/Intel-Xeon X5650 - AT80614004320AD (BX80614X5650).html

Xeon X5650 @ 4.3GHz CINEBENCH R15
https://www.youtube.com/watch?v=0r2c4nKPbVM
MT = 995
2x Intel Xeon X5650 (12C/24T , Default clock : 2666Mhz) = 1279

 

[Nothingness]

I have heard this from many people that I have talked to, and having read the GB4 whitepaper the workloads do seem pretty reasonable/interesting.

Yes, it is. The problem still is clock detection and fake results, but the benchmark itself is solid.

 

[BeepBeep2]

posted on OC.net, it was sourced to Baidu but the original link is down again NDA rules apply but again it seems to be a legitimate 6 core Ryzen benched and posting a score that is = to Haswell/Haswell E and about 9% off Broadwell.

Looks like the same source that Dresdenboy and I were speculating on a month ago that everyone denounced as "fake". I tried to say it may not be at Overclock.net and they were still locking threads and editing posts.

They were looking at Ryzen Cinebench score @ default as per Alva guy:

Guy in the middle is OCing champion as per users on reddit.

Alva is pretty well known.