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

Status
Not open for further replies.

Sweepr

Diamond Member
May 12, 2006
5,151
2
131
#1
Everyone has been asking that this thread be updated, and it has not by the op for months, so a new thread was created today before this one was updated, and that one stands. This one is locked.
Markfw
Anandtech Moderator

Link to new thread
https://forums.anandtech.com/thread...-prices-and-discussion.2499879/#post-38751437

- Performance


21/02/2016 Update:






https://videocardz.com/66216/firefly-was-a-great-show


- Price

According to the snapshots of (already unavailable) Amazon pages, the Ryzen 7 1800X is likely to cost 499 USD, while Ryzen 7 1700X might be 100 USD cheaper (399 USD). The bundle on Amazon includes Corsair H100i v2 cooler, but you look down to ‘bundle details’, there’s a price for each item separately.

More importantly, the release date was confirmed (by Amazon) for March 2nd. For now, let’s treat this as a rumor of high-credibility.

https://videocardz.com/66222/amd-radeon-7-1800x-to-cost-499-usd-ryzen-7-1700x-399-usd






https://videocardz.com/65913/how-fast-is-ryzen




http://wccftech.com/amd-ryzen-7-1700x-processor-tested/




http://wccftech.com/amd-ryzen-1600x-cinebench-r15-performance-confirmed/


23/12/2016 Update (Unofficial):




19/12/2016 Update: Summit Ridge 8C/16T Cinebench R15 MT & Fritz Chess Scores Leaked (Rumor)



Comparison #1:

Core i7-7700K (Stock) 971* cb
Core i7-7700K (OC) 1108* cb
Core i7-6900K (Stock) 1565 cb
Core i7-6950X (Stock) 1863 cb
AMD RYZEN (8/16) 1188 cb

*http://www.bit-tech.net/hardware/cpus/2016/12/05/core-i7-7700k-performance-overclock-preview/3

Comparison #2:





Comparison #1:

Core i7-7700K (Stock)
35.52
Core i7-7700K (5GHz)
41.44
Xeon E5-2670 (Stock)

41.88
Core i7-6900K (Stock)
47.80
Core i7-6950X (Stock)
51.50
AMD RYZEN (8/16)
36.86

Core i7-6900K 30% faster @ FritzChess, 31% faster @ Cinebench R15.


14/12/2016 Update: AMD Demonstrates Ryzen Performance At New Horizon Event, 95W TDP Confirmed




http://www.tomshardware.com/news/amd-zen-ryzen-new-horizon,33194.html


OLDER LEAKS - Straight from Ashes of the Singularity's CPU benchmark:

#COMPARISON 1

- Summit Ridge 8C/16T ES - 2.8-3.2 GHz (2016/2017 Zen)
Average: 58.9 FPS
Normal batch: 65.8 FPS
Medium batch: 62.8 FPS
Heavy batch: 50.5 FPS

http://www.ashesofthesingularity.com/metaverse#/personas/50728c1c-7384-447b-99a9-1a7c330cce99/match-details/bfba4b4a-4b1e-4ab3-8f2f-2375321ea68b

- Core i7-980 6C/12T - 3.33 GHz (2010 Westmere)
Average: 58.6 FPS
Normal batch: 65.2 FPS
Medium batch: 59.7 FPS
Heavy batch: 52.3 FPS

http://www.ashesofthesingularity.com/metaverse#/personas/388c81f2-8199-4446-b54b-675f5a0aa833/match-details/45e74e9d-4b0c-4557-8225-a3b8664dd4a8

- Core i7-4770 4C/8T 3.4-3.9 GHz (2013 Haswell)
Average: 66.0 FPS
Normal batch: 74.5 FPS
Medium batch: 69.5 FPS
Heavy batch: 56.6 FPS

http://www.ashesofthesingularity.com/metaverse#/personas/2fb047a6-39a4-4039-8156-ee759c81bd89/match-details/4baad586-6271-42bd-84c5-6884cfb3341d

- Core i5-6600K 4C/4T - 3.5-3.9 GHz (2015 Skylake)
Average: 82.4 FPS
Normal batch: 87.6 FPS
Medium batch: 85.5 FPS
Heavy batch: 75.2 FPS

http://www.ashesofthesingularity.com/metaverse#/personas/498d1fda-24d1-427f-91cb-591c867d25bf/match-details/8e1f0605-d5d9-49d1-afc0-e6fbf9e51262

- Core i7-6700K 4C/8T - 4.0 GHz (2015 Skylake) + OC
Average: 107.3 FPS
Normal batch: 125.7 FPS
Medium batch: 113.8 FPS
Heavy batch: 89.2 FPS

http://www.ashesofthesingularity.com/metaverse#/personas/045e4b2f-dd7b-4187-aafa-cde9dfe20eb6/match-details/6096dd00-7e11-4368-b588-97413922f5c7

- Core i7-5960X 8C/16 - 3.0-3.5 GHz (2014 Haswell-E)
Average: 109.8 FPS
Normal batch: 127.9 FPS
Medium batch: 119.7 FPS
Heavy batch: 89.7 FPS

http://www.ashesofthesingularity.com/metaverse#/personas/b02c74ea-e3d0-479f-9a11-0b0be0b732c7/match-details/5d5ab818-bc9e-4340-8a33-9e243a097a1f

The Haswell score is 100% at stock and based on the latest version of the benchmark included in the search engine (1.24.20823.0). The benchmark likes cores/thread and scales with more than 4C/8T:



3.0 GHz Haswell-E beating 4.0 GHz Skylake-S.


#COMPARISON 2

- Summit Ridge 8C/16T ES - 2.8-3.2 GHz (2016/2017 Zen)
Average: 31.5 FPS
Normal batch: 36.5 FPS
Medium batch: 33.8 FPS
Heavy batch: 26.2 FPS

- Core i5-4670K 4C/4T 3.4-3.8 GHz (2013 Haswell)
Average: 52.6 FPS
Normal batch: 56.9 FPS
Medium batch: 54.4 FPS
Heavy batch: 47.5 FPS

https://forums.anandtech.com/threads/first-summit-ridge-zen-benchmarks.2482739/page-4#post-38414245

#UPDATE 13/08

AMD ZEN Engineering Sample AOS - Further Performance Analysis





http://www.guru3d.com/news-story/amd-zen-engineering-sample-aos-further-analysis.html


# Comparison 3

Intel Xeon E5-2603 v3 (Haswell) @ 1.60 GHz
- Single-Core Score: 1804

https://browser.primatelabs.com/v4/cpu/117877

AMD Zeppelin (Zen) @ 1.45 GHz
- Single-Core Score: 1141

https://browser.primatelabs.com/v4/cpu/105227
 
Last edited by a moderator:

2is

Diamond Member
Apr 8, 2012
4,287
1
106
#2
I guess this is good for those people who can't wait to upgrade their system to one that's as fast as a 5+ year old Intel machine.
 

mikk

Platinum Member
May 15, 2012
2,417
41
126
#3
Looks very slow compared to Skylake. How does Ashes of the Singularity scale over many CPU cores/threads? How fast is an FX-8350 in this benchmark?
 

Joepublic2

Golden Member
Jan 22, 2005
1,114
0
76
#4
I really hope that's an early chip sample that isn't hitting its clock targets or has some faster paths disabled in the microcode or something, because if it's representative of the retail product, it's terrible.
 
Mar 10, 2006
11,719
122
126
#5
I don't know why anybody expected AMD to match/beat Intel. The R&D gap is massive, and Summit Ridge is a perf/watt optimized server part.

If you want a gaming chip, get a Skylake 6700K.
 
Aug 11, 2008
10,457
67
126
#6
Very disappointing if those are the final clocks. Maybe that is why all they hyped was the IPC gains.
 
Mar 10, 2006
11,719
122
126
#7
Very disappointing if those are the final clocks. Maybe that is why all they hyped was the IPC gains.
No kidding. Pretty obvious when ALL they would talk about is the 40% IPC number to drum up hype that they would not be at iso clocks to previous generation products.

Anyway, I know there will be some who say "this isn't final" blah blah but at this point I have zero reservations just recommending Intel CPUs to anybody and telling them to not bother to wait for Zen.
 
Jan 12, 2005
15,829
3,968
126
#8
I wouldn't base too much off a single benchmark running on an engineering sample.
 
Mar 10, 2006
11,719
122
126
#10
True, but I wouldnt bet on final silicon being significantly better either.
Yep, at this point they've got to have silicon running at or close to final clocks, esp if they are going to launch in just 4 months.
 

NostaSeronx

Platinum Member
Sep 18, 2011
2,372
159
126
#11
Majority of the improvement can be construed into basically being a lower latency/more modern platform.

It has an integrated northbridge that isn't locked to something that is 5.2 GT/s and on a older process node. It has access to PCIe 3.0 which has been tweaked since its inception that it is now lower latency than Hypertransport 3.1. The L1/L2/L3/Memory most likely have lower latency and higher bandwidth.

http://www.ashesofthesingularity.co...-details/5f60c37c-b017-4695-a136-17da0aedab42
^-- FX-8310 + RX-480

http://www.ashesofthesingularity.co...-details/bfba4b4a-4b1e-4ab3-8f2f-2375321ea68b
^ Zen ES + RX-480

The extra ALU pipes help, but aren't really a huge boost. Since, the AGLUs in Bulldozer if given Arithmetic datapaths would do the same thing. While for much lower area cost and design costs(from scratch not needed.) At least with Bulldozer, we would be getting physical cores with independent L1Ds. So, data thrashing on the L1D is deferred to the L2 cache and Memory.
 

plopke

Senior member
Jan 26, 2010
215
4
101
#12
I might be the only one but if they would get 2500K performance and cheap , I would still be sold it is 8C/16T still in the end :p.

PS yes it is a bit underwhelming.
 

raghu78

Diamond Member
Aug 23, 2012
3,999
34
136
#13
Given how poorly AMD have been executing in the past I would be surprised if AMD got even Sandy bridge IPC and 4 Ghz clocks in turbo. If they do that it would be a success. Anyway I think Vega and Zen will be huge disappointments. Right now AMD looks like they won't last for long as they cannot design competitive products and cannot find a way to make money.
 
Mar 10, 2006
11,719
122
126
#14
Given how poorly AMD have been executing in the past I would be surprised if AMD got even Sandy bridge IPC and 4 Ghz clocks in turbo. If they do that it would be a success. Anyway I think Vega and Zen will be huge disappointments. Right now AMD looks like they won't last for long as they cannot design competitive products and cannot find a way to make money.
They are making good money off of semicustom, but seriously they need to stop trying to compete with NVIDIA and Intel at the same time in their respective markets.
 

Qwertilot

Golden Member
Nov 28, 2013
1,414
37
106
#15
Well, to some extent that competition is 'free' once they are doing the semi custom stuff anyway. More so with the gpu's than zen perhaps.
 
Mar 10, 2006
11,719
122
126
#16
Well, to some extent that competition is 'free' once they are doing the semi custom stuff anyway. More so with the gpu's than zen perhaps.
Not really, there's a lot of extra work that goes into taking the core GPU IP and building dedicated ASICs aimed at gaming. I would imagine that Vega using HBM2 doesn't fall out "freely" from console SOC dev.
 

know of fence

Senior member
May 28, 2009
555
0
71
#17
- Core i7-980 6C/12T - 3.33 GHz (2010 Westmere)
Average: 58.6 FPS
Normal batch: 65.2 FPS
Medium batch: 59.7 FPS
Heavy batch: 52.3 FPS

http://www.ashesofthesingularity.co...-details/45e74e9d-4b0c-4557-8225-a3b8664dd4a8
The Westmere comparison uses a different video card! (780 ti compared to a RX 480) .. And of course there are differences with ram, clocks and other stuff. Someone needs to dig for a better juxtaposition.
With Zen Sandy Bridge performance is expected, also 4.0 GHz turbo on a 8 core chip is completely unrealistic.

Here is Zen beating the i5-2500K
http://www.ashesofthesingularity.co...-details/5e39e1eb-d6c3-47e6-8794-461fcb914e86

What is a "CPU-Framerate" anyway?
 
Last edited:

dealcorn

Senior member
May 28, 2011
247
0
76
#18
Does the Jim Keller pixie dust come with any sort of a guarantee?
 

SAAA

Senior member
May 14, 2014
398
0
56
#19
Majority of the improvement can be construed into basically being a lower latency/more modern platform.

It has an integrated northbridge that isn't locked to something that is 5.2 GT/s and on a older process node. It has access to PCIe 3.0 which has been tweaked since its inception that it is now lower latency than Hypertransport 3.1. The L1/L2/L3/Memory most likely have lower latency and higher bandwidth.

http://www.ashesofthesingularity.co...-details/5f60c37c-b017-4695-a136-17da0aedab42
^-- FX-8310 + RX-480

http://www.ashesofthesingularity.co...-details/bfba4b4a-4b1e-4ab3-8f2f-2375321ea68b
^ Zen ES + RX-480

The extra ALU pipes help, but aren't really a huge boost. Since, the AGLUs in Bulldozer if given Arithmetic datapaths would do the same thing. While for much lower area cost and design costs(from scratch not needed.) At least with Bulldozer, we would be getting physical cores with independent L1Ds. So, data thrashing on the L1D is deferred to the L2 cache and Memory.
New platform after 5 years of the same old is a good thing indeed, the rumors about low idle power and all the PCIe lanes (36?) are what I'm looking forward the most from AMD now.
Better than Intel's quads for gaming? Impossible of course, heck quad core Zens have more chances to catch latest i7s than the octa core only for higher possible clockspeeds.
BTW if even Ashes who scales so well with threads shows a ~3GHz octa core Zen as only 43% better than ~ 4GHz FX-83xx then we have a problem.

With Zen Sandy Bridge performance is expected, also 4.0 GHz turbo on a 8 core chip is completely unrealistic.
Yeah, but it looked like it was their target: going from ~4GHz Piledriver to similar clocked Zen, in order to get all the IPC advantage into actual performance. We'll see if that 4GHz is impossible, probably not for overclockers, but I doubt with 95W TDP it's going to be that high at stock. Unless it's a process issue running 4+ should be doable with decent cooling.

What is a "CPU-Framerate" anyway?
Pretty sure it means the framerate the cpu runs the game at, think AI, logic etc vs the visual framerate from the GPU (that in the end is the one you see).
 
Last edited:

Sweepr

Diamond Member
May 12, 2006
5,151
2
131
#20
Very disappointing if those are the final clocks. Maybe that is why all they hyped was the IPC gains.
Matches the latest leaks. Heck, even the OPN from the benchmark is the same:

So the OPN is "1D2801A2M88E4_32/28_N". Using the already known schema, "D" stands for desktop, "28" for the base clock frequency (2.8GHz), the first "8" of "88" for the number of cores. And more clearly, "32/28" stand for turbo boost frequency (3.2GHz), and the base clock again (2.8GHz). This matches the information given for the 8C DT ES in an AnandTech forum posting recently:

"The most exciting part is core clock. The 8c/95W variant's base clock is 2.8GHz, all core boost is 3.05GHz and maximum boost is 3.2GHz."
 
Aug 11, 2008
10,457
67
126
#21
The Westmere comparison uses a different video card! (780 ti compared to a RX 480) .. And of course there are differences with ram, clocks and other stuff. Someone needs to dig for a better juxtaposition.
With Zen Sandy Bridge performance is expected, also 4.0 GHz turbo on a 8 core chip is completely unrealistic.

Here is Zen beating the i5-2500K
http://www.ashesofthesingularity.co...-details/5e39e1eb-d6c3-47e6-8794-461fcb914e86

What is a "CPU-Framerate" anyway?
Actually, in Ashes, would not a 480 be expected to be faster than a 780TI
 
Jan 26, 2014
74
0
36
#22
I seriously hope that all of that is fake and this is an early sample of the chip or something.
 

moonbogg

Diamond Member
Jan 8, 2011
9,756
28
126
#23
Christ almighty. Looks like my Skylake-X platform is going to cost:

1 left arm
1 right knee cap
1 center eye ball (rare, but necessary)
1 liver from stripped zebra cat (whatever)

Ticket me, infract me, ban me, gank me with a hoolahoop. I'm pissed. AMD sucks. They need to just die so someone else can maybe take their spot.
A point in time comes when there is nothing left to do but lay down and die. As creatures, we do this. Businesses are no exception to the laws of life and death. AMD is on life support, breathing through a tiny straw, costing those who love them a great deal of emotional pain. Its easier to just let go. AMD, let it go. Say good bye. Its time to pull the plug.
 
Last edited:
Mar 10, 2006
11,719
122
126
#24
I seriously hope that all of that is fake and this is an early sample of the chip or something.
My guess is that final retail silicon won't be much better. At this stage of the development process, it's too late for sweeping changes. I think this is it for zenver1.
 
Last edited:

Bryf50

Golden Member
Nov 11, 2006
1,429
0
91
#25
I don't know how Zen will perform and it's not a product I will likely buy. This thread is just the usual suspects doom and glooming. If it was any other company the same people would be quick to poke holes in these results for obvious reasons. It's sad that you don't even have to read the OP. Just look at the title and the poster and you already know the contents.


Personal attacks are not allowed here.
Markfw900
 
Last edited by a moderator:
Status
Not open for further replies.


ASK THE COMMUNITY

TRENDING THREADS