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

[Sweepr]

I am just going to wait for the actual product reviews.

No you won't.

Goddamn it. 3.4 Ghz base. Beats a stock 6900k boosting to 3.5 Ghz while running at 3.4 Ghz. Man exciting times for the CPU market. I think AMD is back.

You are already proclaiming a winner before independent reviews, and blaming other users when leaked results don't match your expectations. Try to be honest before calling out other people.

 

[Glo.]

No you won't.



You are already proclaiming a winner before independent reviews, and blaming other users when leaked results don't match your expectations. Try to be honest before calling out other people.

Do you have any proofs that the leaked result is from Summit Ridge? Or is it because someone told you that is the case?

We have seen benchmarks already, comparing 6900K, and 8C/16T 3.4 GHz CPU from AMD. In case you missed it.
Stop spreading FUD.

 

[PhonakV30]

No you won't.



You are already proclaiming a winner before independent reviews, and blaming other users when leaked results don't match your expectations. Try to be honest before calling out other people.

Did you ever write a good post about AMD Products?

 

[kraatus77]

This thread is ruined. as usual.

 

[Sweepr]

Do you have any proofs that the leaked result is from Summit Ridge? Or is it because someone told you that is the case?

We have seen benchmarks already, comparing 6900K, and 8C/16T 3.4 GHz CPU from AMD. In case you missed it.
Stop spreading FUD.

You have no right to tell me what to post and what not to post. Grow a thick skin and deal with the leaked benchmarks, instead of attacking me for posting them. They might or might not reflect the actual product, but they are what they are. Basically Sandy Bridge level performance per clock in that particular Cinebench R15 score.

 

[Glo.]

Did you ever write a good post about AMD Products?

Good products do not need anyones acceptance. They are just good as they are. His opinion does not matter here. You are attacking the messenger, rather than the message.

What matters is that Sweeper without any proofs tries to prove that AMD is lying or is hopeless company that should be closed and go to pages of history.

You have no right to tell me what to post and what not to post. Grow a thick skin and deal with the leaked benchmarks, instead of attacking me for posting them. They might or might not reflect the actual product, but they are what they are. Basically Sandy Bridge level IPC looking at that particular Cinebench R15 score.

I am not telling you what to post, or not post. Learn to read with understanding. I have asked you a very simple question. What proof you have that the scores are factually from Summit Ridge 8C/16T CPU?

Maybe you just bought a shit someone tried to sell to you, because it confirmed your personal belief about particular brand?


Profanity is not allowed in the technical forums,
Markfw
Anandtech Moderator

 

[.vodka]

I am not telling you what to post, or not post. Learn to read with understanding. I have asked you a very simple question. What proof you have that the scores are factually from Summit Ridge 8C/16T CPU?

Maybe you just bought a shit someone tried to sell to you, because it confirmed your personal belief about particular brand?

Exactly.

Why did the leaker crop the CPU-z part of the screenshot in fritz chess?

It doesn't compute. If you're leaking, then leak properly. A leak like this is useless, it only prompts for further speculation, muddying the waters.

Anyway, CES is coming in two weeks' time. If the rumors are right AMD will reveal some more about Zen. SKUs, motherboards, probably another benchmark or two.

Between a leak of this quality and an AMD controlled benchmark, I take the AMD controlled benchmark which at the very least also tells me the conditions on which it was done, and so far can be replicated by other people with the same hardware.

 

[krumme]

Use the report function to handle threadcrapping

 

[clemsyn]

Why was is cropped? Still hoping for the best for AMD but it doesn't look good. I hope they get pricing right.

 

[PhonakV30]

Good products do not need anyones acceptance. They are just good as they are. His opinion does not matter here. You are attacking the messenger, rather than the message.

What matters is that Sweeper without any proofs tries to prove that AMD is lying or is hopeless company that should be closed and go to pages of history.

I am not telling you what to post, or not post. Learn to read with understanding. I have asked you a very simple question. What proof you have that the scores are factually from Summit Ridge 8C/16T CPU?

Maybe you just bought a shit someone tried to sell to you, because it confirmed your personal belief about particular brand?

He did "copy paste" from WCCFtech.they said It's from Chinese Forum.so Original article was from Wccftech.I can't find any source about Chinese forum.

 

[coercitiv]

You have no right to tell me what to post and what not to post.

Sweepr, what's with the thread title?

Summit Ridge (Zen) Benchmarks - Cinebench R15 & Fritz Chess (Page 138)

I'm speechless.

 

[.vodka]

Sweepr, what's with the thread title?


I'm speechless.

He likes to edit his threads as he pleases to fit his narrative, and he also likes to play moderator on his own threads.

He's not a moderator. I don't know why that behavior is allowed, it's not the first time he's done it.


Yeah, don't shoot the messenger. It is sometimes how the messenger delivers the message, you know. This message in particular is incomplete, which only triggers responses on both sides of the fence that wouldn't be otherwise.

Until the source of the leak is clarified and the missing details are revealed, this leak is useless.

 

[PhonakV30]

According to CpuMonkey

AMD FX-8350 : 665 / 4 Module = 166.25
(since Cinebech Is all about FP Performance so 4 Module = 8 Int + 4 FP )
AMD Ryzen : 1188 / 8 cores = 148.5
Intel Core i7-6900K : 1565 / 8 cores = 195.625
Intel Core i7-5960X : 1360 / 8 cores = 170

Ratio : 1.318 ( 6900K / Ryzen)

 

[Nothingness]

Granted there s no uarch that is favoured by compiler or special instructions the results should be in a same window, besides the FX8350 is at 13168 kn/s, given that Zen has SMT this would imply that the ST perf is about the same as the FX.

Just compare the ST perf of the FX to HW for instance, the difference is 41.5%, yet anothe hint that Zen has the same ST perf as the FX in this alleged review that show no CPUZ pic, or anything that would prove that it s a Zen used at stock settings.

I'm ready to believe that the leaked results are not correct, but that wasn't my point.

Also there s no branching issue overall, if we look at more recent chess games the FX beat even HW, so Fritzchss is a case favourable to Intel but not to the point of producing such paradoxal numbers.

http://www.hardware.fr/articles/946-5/performances-applicatives.html

Houdini and Stockfish are currently the best chess games.

Here is how it goes on Haswell:

Performance counter stats for './stockfish bench':

  2689,708339  task-clock (msec)  #  0,998 CPUs utilized  
  157  context-switches  #  0,058 K/sec  
  20  cpu-migrations  #  0,007 K/sec  
  4 595  page-faults  #  0,002 M/sec  
  8 494 677 184  cycles  #  3,158 GHz  
  <not supported>  stalled-cycles-frontend 
  <not supported>  stalled-cycles-backend  
  13 268 364 501  instructions  #  1,56  insns per cycle  
  1 686 175 899  branches  #  626,899 M/sec  
  83 973 894  branch-misses  #  4,98% of all branches  

  2,693845704 seconds time elapsed

As you can see the branch MPKI is 6.3, much higher than for Handbrake with its ~1.8 MPKI.

My point was that you simply can't claim the Handbrake (or any other integer) result can disprove the Fritz result Sweepr posted.

 

[Sweepr]

Unbelievable how a leak that puts Ryzen 20% above stock 7700K in CB R15 (even if it only matches the 4C/8T competitor in Fritz Chess) already causes such mass hysteria from the usual suspects. Can't even imagine what will happen if independent reviews put it below some people's expectations of better than 6900K performance at 95W TDP. I even said it would be a good product if these results are true, and made it clear multiple times that they might or might not reflect the actual product - it's in the nature of leaks and that's where the fun is. It's up to the user to sort out which of them to believe or not.

 

[TheELF]

Clock for clock, Zen should be on Haswell/Broadwell Level.

No, it should be 40% "+" over excavator.
https://forums.anandtech.com/threads/core-i3-6300-mini-review-atenra.2456159/#post-37868274
Aten-ras benchmark so no bias towards intel.
Excavator at 25% higher clocks @4.35Ghz and haswell @3.5Ghz is still ~34% faster.
At least for CB15 zen should be a good bit slower then haswell.

 

[Glo.]

No, it should be 40% "+" over excavator.

If there would be not more than 40% ST uplift, Zen at 3.4 GHz would not match/beat Broadwell-E 6900K in Blender and Handbrake benchmarks.

 

[TheELF]

dp

 

[PhonakV30]

I'm ready to believe that the leaked results are not correct, but that wasn't my point.


Here is how it goes on Haswell:

Performance counter stats for './stockfish bench':

  2689,708339  task-clock (msec)  #  0,998 CPUs utilized 
  157  context-switches  #  0,058 K/sec 
  20  cpu-migrations  #  0,007 K/sec 
  4 595  page-faults  #  0,002 M/sec 
  8 494 677 184  cycles  #  3,158 GHz 
  <not supported>  stalled-cycles-frontend
  <not supported>  stalled-cycles-backend 
  13 268 364 501  instructions  #  1,56  insns per cycle 
  1 686 175 899  branches  #  626,899 M/sec 
  83 973 894  branch-misses  #  4,98% of all branches 

  2,693845704 seconds time elapsed

As you can see the branch MPKI is 6.3, much higher than for Handbrake with its ~1.8 MPKI.

My point was that you simply can't claim the Handbrake (or any other integer) result can disprove the Fritz result Sweepr posted.

Does "stockfish" need to optimize for AMD SMT?

 

[Glo.]

Unbelievable how a leak that puts Ryzen 20% above stock 7700K in CB R15 (even if it only matches the 4C/8T competitor in Fritz Chess) already causes such mass hysteria from the usual suspects. Can't even imagine what will happen if independent reviews put it below some people's expectations of better than 6900K performance at 95W TDP. I even said it would be a good product if these results are true, and made it clear multiple times that they might or might not reflect the actual product - it's in the nature of leaks and that's where the fun is. It's up to the user to sort out which of them to believe or not.

I will ask one more time: Have you got any proof that the scores are actually from Summit Ridge 8C/16T CPU?

When people link WCCFTech leaking information about AMD GPUs everybody says it is rubbish, clickbait site, and it will not turn out true. But when they leak unnamed CPU scores, from Cinebench R15, and from Chess game, which comes from anonymous chinese forum, people jump to conclusions, and somehow, now WCCFTech is extremely credible source.

Proofs, my friend, proofs. Then we can talk. Right now, without CPU-Z shots its FUD. Nothing more, nothing less.

 

[krumme]

He likes to edit his threads as he pleases to fit his narrative, and he also likes to play moderator on his own threads.

He's not a moderator. I don't know why that behavior is allowed, it's not the first time he's done it.


Yeah, don't shoot the messenger. It is sometimes how the messenger delivers the message, you know. This message in particular is incomplete, which only triggers responses on both sides of the fence that wouldn't be otherwise.

Until the source of the leak is clarified and the missing details are revealed, this leak is useless.

It would be nice if someone then took the initiative and oportunity to create a zen bm thread. So we could have a collection of confirmed results as they keep comming in the next month or two.

 

[Sweepr]

Proofs, my friend, proofs. Then we can talk. Right now, without CPU-Z shots its FUD. Nothing more, nothing less.

Just so you know I managed to find the original source, it's a Baidu user and he says he will post more benchmarks in a few days. I will ask him to include a CPU-Z screenshot.

 

[krumme]

Btw it seems the thread title was changed back. Lol

 

[Sweepr]

Btw it seems the thread title was changed back. Lol

Of course, I'm open to justified criticism, as long as it doesn't include personal attacks.

 

[Glo.]

Just so you know I managed to find the original source, it's a Baidu user and he says he will post more benchmarks in a few days. I will ask him to include a CPU-Z screenshot.

Now we're talking.