• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

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

Page 159 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Status
Not open for further replies.
superstition said:
Even if you were to, it wouldn't rebut my statement unless it can be proved that the games that show that my statement seems true were designed in a manner that sabotages performance on Intel. And, even if that were the case it would also have to be proven that your citations aren't examples where performance was sabotaged on AMD.

The bottom line here appears to be that, in contrast with the hyperbole, Piledriver is able to keep up with Sandy in games when they are developed in a manner that utilizes its strengths. That is not a "major disaster" in terms of the architectural design. It just means Piledriver hasn't improved since 2012 which isn't surprising since it's a 2012 architecture. Yes, Sandy is more efficient overall but that doesn't make Piledriver a "major disaster" considering that AMD hardly had the development resources Intel enjoyed.
Click to expand...
Yea, it was a great success. Look at how it helped them gain market share and take over the server market, oh, wait, I must be confused, that wasnt bulldozer was it? You also forgot to mention high power consumption and the fact that the cherry picked benchmark you managed to dig out doesnt include intel's latest architecture, and in any case appears to be gpu limited, since the top six cpus show identical performance. In any case, this is off topic in a Zen thread.
 
frozentundra123456 said:
In any case, this is off topic in a Zen thread.
Click to expand...
It's not off-topic. This topic is about benchmarking. Benchmarking is about comparing one architecture to another. My posts have been about the Blender benchmark and how it relates to Piledriver, including the first post today that included the Desert of Kharak result.
frozentundra123456 said:
Yea, it was a great success. Look at how it helped them gain market share and take over the server market, oh, wait, I must be confused, that wasnt bulldozer was it?
Click to expand...
1) Servers and gaming are not the same market. My post specifically discussed gaming. Obviously, AMD suffered more in the server market where performance per watt is more important. People have been discussing gaming performance, though.

2) Market success and architectural design quality are two different issues.

frozentundra123456 said:
the cherry picked benchmark you managed to dig out doesnt include intel's latest architecture
Click to expand...

1) As I already said, not only is that not the only gaming result that demonstrates my point I haven't seen anyone other than myself make the complaint about cherry-picking a benchmark in relation to the Ryzen Blender hype — in terms of how it provides an insufficient picture of performance due to not leveraging AVX/XOP.

2) Why should a 2012 architecture be compared with the latest Intel architectures to demonstrate that it's not a massive failure?
frozentundra123456 said:
and in any case appears to be gpu limited, since the top six cpus show identical performance.
Click to expand...
It's clearly not GPU limited, certainly not the point of not being able to demonstrate my point.

I'm going to say this once, too. If you or anyone else wish me to respond to anything you post in the future, make it polite.
 
Arachnotronic said:
You got it. Ryzen is going to be fun to play with 🙂
Click to expand...

Just remembered; I have an AMD GPU, the venerable 7850. If you're running NVidia, is there a way to explicitly disable DCLs? I know that there's a "Thread Optimization" setting in the driver settings program, but no idea if that's the ticket.

If there isn't, we could try an older game (Direct3D 9 + ENB for draw call measurement), like Fallout New Vegas or Skyrim, since DCLs are a D3D11-only feature.
 
I have New Vegas and Skyrim and would be happy to test it if/when I get my Summit Ridge. I have all kinds of goofy 290s and 390s around the house, plus a 270. No guarantee that any of them work for display now though . . .

edit: downside is that I'm currently limited to 1600x1200 max res so if you want hi res testing, I'm no good.
 
Last edited:
DrMrLordX said:
I have New Vegas and Skyrim and would be happy to test it if/when I get my Summit Ridge. I have all kinds of goofy 290s and 390s around the house, plus a 270. No guarantee that any of them work for display now though . . .

edit: downside is that I'm currently limited to 1600x1200 max res so if you want hi res testing, I'm no good.
Click to expand...

The more the merrier. The ideal would be to cross reference between a bunch of different processor architectures. I'll make a dedicated thread about it when we've got lads with Zen to compare against, see if we can get a community effort going.

With graphics settings, they're to be set to the minimum allowed by the engine. Object and shadow distances up, texture and shadow resolution down. Draw calls are unaffected by texture detail 'n' such, so we ought to mitigate the (lack of) GPU performance as much as poss. I'll make a wee archive with all the .ini settings 'n' a bare-bones ENB when the time comes.
 
http://pc.watch.impress.co.jp/docs/column/kaigai/1036983.html

3.jpg.html
 
MajinCry said:
The more the merrier. The ideal would be to cross reference between a bunch of different processor architectures. I'll make a dedicated thread about it when we've got lads with Zen to compare against, see if we can get a community effort going.
Click to expand...

Feel free to set it up a bit early if you want me to do some testing with Steamroller. This 7870k is good for 4.7 GHz when necessary. Not sure if you'd want me to use the iGPU or a Hawaii card for testing, though. I think I can get my 390 to play ball.
 
DrMrLordX said:
Feel free to set it up a bit early if you want me to do some testing with Steamroller. This 7870k is good for 4.7 GHz when necessary. Not sure if you'd want me to use the iGPU or a Hawaii card for testing, though. I think I can get my 390 to play ball.
Click to expand...

Aye, slap on the 390. The more GPU grunt, the better. As for CPU clocks, 3.4Ghz is a good baseline since that's what we've seen Zen have.
 
bakyt115 said:
http://pc.watch.impress.co.jp/docs/column/kaigai/1036983.html

3.jpg.html
Click to expand...
Interesting to note that the microcode rom can output 6 uops cycle, vs 4 of previous design (also skl seems have the classic INTEL 4-1-1-1 burst decode). From the video of september 1st, we can also know that in the uop cache, for microcoded instructions, only a pointer to microcode rom is stored. The microcoded instruction is expanded in the microop queue at a rate of 6 uops/cycle. This saves space in the uop cache and don't block the decoders when hitting a microcoded instruction. I wonder if the decoders can output 4/instr cycle even if microcoded... INTEL and previous AMD archs can't... And the decoders will be blocked for all the time in which the microcode ROM spits uops... With Zen apparently this is not the case...
 
Last edited:
Why is the Handbrake demo from New Horizon so much overlooked? Rysen was actually ~16% faster in this test. Is this irrelevant?
 
MajinCry said:
Aye, slap on the 390. The more GPU grunt, the better. As for CPU clocks, 3.4Ghz is a good baseline since that's what we've seen Zen have.
Click to expand...

Okay, sounds good. Let me know when you have something thrown together and I'll redownload New Vegas and/or Skyrim and see what I can do.
 
If it performs like that with the SMT and uop cashe nerfed what will happen when it is fully functional? 🙂
 
Last edited:
Just compared to the piledriver apu i have, Zen will do extremely well. I am going to start putting money aside for a new zen based system about end Q2 or beginning Q3 2017.
I now use about 4 virtual machines (not at the same time) and although my A10-6700 can manage it fine, in the near future i will be running virtual machines with more demanding software. I am interested to see how well (dated) games will run in a virtual machine. And i think with zen that will run really well.
I have seen what the I7-6700 can do, and it seems zen can match an I7-6700 easily.
I like it when my virtual machine is as snappy as the host os is.
I ordered a 520W psu, and if all goes well i will have my rx480 card around half of next week.
In the near future, my A10-6700 based hardware will become the media pc.
 
krumme said:
Then 8c zen+ will be fast enough for the next gen consoles.
Click to expand...

What clocks will 8c zen+ have when it is clocked low enough to fit in the consoles tdp budget? 🙂

edit : and here comes the bug creep it seems .. here is to hoping that nothing serious turns up 🙂.

(the way AMD has slow played the hand so far, crippling smt and tracecache could be another iteration of that slowplay .. who knows.)
 
Last edited:
Status
Not open for further replies.

TRENDING THREADS

Back
Top