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

[Abwx]

- 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

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

This tells you everything you really need to know.

Thoses numbers mean nothing, they are not the bench results at all, it s just deliberate misleading from the OP, he used the CPU framerate and made it look in the OP as if it was AOTS numbers, the highest FPS with a 480 and standard mode is 70 FPS on average with the fatstest submission registered.

http://www.ashesofthesingularity.co...X 480 Graphics","gameVersion":"1.30.21168.0"}

 

[The Stilt]

People are making snap judgments from a benchmark that is wholly unreliable and not particularly static in terms of predictable performance.

If someone actually had this on their hands, they'd benchmark a well known bench and not this. If they really have the ES on hand and used this bench, they really had to cherry pick pretty hard to put it in a negative light.

So the main purpose for system manufacturer / OEM evaluation is to leak benchmarks?
This benchmark was obviously leaked because of negligence, since it got pulled so fast. AMD wouldn't be able to react so fast, especially with a leak from a person working for Chinese / Taiwanese company. Most likely the user noticed his error due all the fuss and removed his results from online. If AMD had asked the makers of AotS to withdraw the results, most likely they would have banned or removed the user completely to prevent similar, further mishaps. Except the user is still there.

I've personally made a similar error in the past

 

[beginner99]

OTOH, let's say I actually needed an 8-core CPU -- Intel's cheapest is $1089 USD. That gives AMD room to price Zen between $350-699 and still undercut the 6900K to the point where they aren't competitors. Thus, for me Zen was never going to live up to gaming expectations but for someone who wants a multi-threaded CPU for [insert whatever tasks] a $545 8-core Zen would cost 1/2 of a 6900K. Is that a FAIL? I don't know, I don't buy $1090 CPUs for productivity but on paper if 6900K isn't 2x faster, 8-core Zen priced < $600 has a market.

I like your posts but I want to add my standard post about CPU price vs. total ownership price of the end product which includes software. And in the market you describe (which IMHO is a pretty small niche, too small to maintain CPU R&D and process costs) CPU price matters very little. Yes, the CPU might be half-price but if the end-product (workstation) is running 2-3 software packages each costing $5000 per seat, that CPU price difference won't matter much at all. Add to that the user using the workstation costs you $50/hrs upwards, paying 100% more for the CPU while only saving 5% of users time is still totally worth it over a multi-year period.

 

[Arachnotronic]

So the main purpose for system manufacturer / OEM evaluation is to leak benchmarks?
This benchmark was obviously leaked because of negligence, since it got pulled so fast. AMD wouldn't be able to react so fast, especially with a leak from a person working for Chinese / Taiwanese company. Most likely the user noticed his error due all the fuss and removed his results from online. If AMD had asked the makers of AotS to withdraw the results, most likely they would have banned or removed the user completely to prevent similar, further mishaps. Except the user is still there.

I've personally made a similar error in the past

Really appreciate your insight on this stuff, The Stilt.

 

[jpiniero]

This tells you everything you really need to know. Assuming the 4770k is locked at 3.4Ghz, it still supposedly is 38% faster per clock. If the 4770k is at 3.9Ghz, it is 59% faster per clock. We know neither of these to be the case. The bench is unreliable and not speculation based off of these supposed numbers are no better than a rumor.

Now that I have had a chance to look at it, the 6700K is likely overclocked to around 4.8. Stock would get in the mid to high 80s (CPU framerate and if you had a more powerful GPU). Broadwell-E 6-core gets into the 90s.

 

[Sweepr]

Thoses numbers mean nothing, they are not the bench results at all, it s just deliberate misleading from the OP, he used the CPU framerate and made it look in the OP as if it was AOTS numbers, the highest FPS with a 480 and standard mode is 70 FPS on average with the fatstest submission registered.

http://www.ashesofthesingularity.com/metaverse#/ladders/benchmark/overall/Low_1080p?viewType=myself&filters={"gpu":"Radeon (TM) RX 480 Graphics","gameVersion":"1.30.21168.0"}

So according to your logic the CPU benchmark results mean nothing, and the only relevant metric are the GPU bound overall results. Would be a good way to hide the less than stellar showing from Summit Ridge ES here, wouldn't it? Meanwhile in other posts you try hard to convince us that running The Witcher 3 + WinRAR at the same twice is a great CPU benchmark. Thanks for the laughs.

Using your link:

- 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

- 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

 

[hrga225]

Well,after examining comparison done by guru3d and wccf,and AotS benchmark leaderboard,I have to say we are back to square one - we know nothing(or almost nothing) of general performance of Zen CPUs.
Game bencmark is GPU bound(which should be if we are honest),CPU score in bencmark is inconsistent,and from Guru3d graphs it looks like there is some bottleneck on platform.
Like forum member itsmydamnation said in one of threads if we could only get someone to leak spec 403.gcc,but it is what it is.

 

[Abwx]

So according to your logic the CPU benchmark results mean nothing, and the only relevant metric are the GPU bound overall results. Would be a good way to hide the less than stellar showing from Summit Ridge ES here, wouldn't it?

It s you that are trying to hide the numbers, CPU framerate means nothing because it s mainly a measure of RAM bandwith of the system, or are you going to explain us that SKL is 60% faster/clock than HW..?.

Using your link:

- 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

- 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

Back with the CPU framerate, are you not tired to hide the bench numbers...?..

Here are the numbers, the 6600K use false fullscreen contrary to the two others, but i didnt expect you to be accurate, btw, it does better than intel own 8C/16T :


- Core i5-6600K 4C/4T - 3.5-3.9 GHz (2015 Skylake)
Average: 69.8 FPS
Normal batch: 77.9 FPS
Medium batch: 70.8 FPS
Heavy batch: 62.5 FPS

Core i7-5960X 8C/16T 3.2 GHz

Average: 65.5 FPS
Normal batch: 78.3 FPS
Medium batch: 67.6 FPS
Heavy batch: 54.9 FPS

- Summit Ridge 8C/16T ES - 2.8-3.2 GHz (2016/2017 Zen)
Average: 54.3 FPS
Normal batch: 60.8 FPS
Medium batch: 55.5 FPS
Heavy batch: 48.1 FPS

http://www.ashesofthesingularity.co...-details/8e1f0605-d5d9-49d1-afc0-e6fbf9e51262

http://www.ashesofthesingularity.co...-details/5d5ab818-bc9e-4340-8a33-9e243a097a1f

And next time post the links of whatever data you re using, just like i did, because me i ve nothing to hide, isnt it...

Meanwhile in other posts you try hard to convince us that running The Witcher 3 + WinRAR at the same twice is a great CPU benchmark. Thanks for the laughs.

The laugh is for the ones denying the results of benchmarks, with a FX you can multitask all the way whatever the apps, with i5s even the most recent you cant, indeed you are left changing the goalposts as a mean to elude the numbers...

 

[Sweepr]

Amazing that a 4C/4T 6600K is still almost 30% faster than the 8C/16T Summit Ridge ES on average in the RX 480 benchmarks you're picking.

Back with the CPU framerate, are you not tired to hide the bench numbers...?..

Keep using irrelevant GPU average framerate, meanwhile I'll stick with CPU framerate to compare CPUs till newer benchmark results appear.

Updating the OP as we speak, it's worse than I thought based on your links.

- Summit Ridge 8C/16T ES - 2.8-3.2 GHz (2016/2017 Zen)
Average: 58.9 FPS

- Core i7-5960X 8C/16 - 3.0-3.5 GHz (2014 Haswell-E)
Average: 109.8 FPS

 

[krumme]

So a few things here:
1. Software support, AVX 512 is going to be way more fragmented then AVX is, how long before anyone cares?

2. Load store bandwdith, unless they double L/S (one of the most expensive parts of a core for both power and size) to be 4 times that of Zen then unless you can keep your work in registers 512bit avx is in the same place Zen is for 256bit.

3. What do you mean by "full speed". >Haswell FP pipes are all FMA and have longer latency then Zen so if the data is in registers even with back to back execution on Zen 256bits will execute quicker, Intel has 2 main pipes for FP that are 256bit, Zen has 4 pipes (effectively 2add, 2mul) that are 128bit. The execution time isn't an issue (speed), throughput in 256bit ops per clock will be a fair bit lower for Zen then >haswell because of load/store bandwidth, but are you going to buy an 8-10 intel chip anytime soon? I know im not dropping the $2500 a i7-6950 costs in OZ. The 8 core @ $1600 is far more reasonable

Its worth remembering TAM, how much TAM increase does intel get by adding large amounts of IPC, how much TAM does intel get by displacing exsisting FPGA/ASIC/DSP/etc by deploying new instructions/features into Core. What Intel does isn't necessarily the best thing for general computing, its the best thing for Intel to make money*. AMD has such a low Server TAM at the moment that all they need to do is general computing well to see a massive increase in TAM.

* I have no problem with this

The slim downed fpu vs Intel stuff is imo about the most important decision and hope for this product to stay compettitve in some segments. Its seems like a no brainer from a business perspective but amd typically have acted driven solely by mostly meaningless tech or about as worse; tech introduced to early.
I hope the decision to go 128 is because it gives them a compettitive edge not to go 256bit and avx512. And not because of lack of ressources. It probably isnt though. Lol.

 

[krumme]

With the rumors of 48core Zen+ cpu's for 7nm i wouldn't expect 256bit ops then either as that would be a ~300% increase in throughput the cache fabric would need to provide.

The product must also be fit for consoles on 7nm imo it they think business and not have "free hands". It sets some limitations for size and cost.

Jaguar was extended from 64bit ops in bobcat to 128 as i remember. As we can see in doom they now use tons of asynch to ofload cpu i guess. Is there any cost benefit here in keeping the same wideness on the programming side?

 

[Abwx]

Amazing that a 4C/4T 6600K is still almost 30% faster than the 8C/16T Summit Ridge ES on average in the RX 480 benchmarks you're picking.

Keep using irrelevant GPU average framerate, meanwhile I'll stick with CPU framerate to compare CPUs till

That s the results of the benchmark, so it s you that are picking datas that have no relevance with the CPU actual perfs, you should add a 4790K so people can see that you re using whatever fill your needs to mislead people, the 6600K not only has a different setting but is highly overclocked as most of the systems at AOTS page, Guru3D has their running at 3.2 while we can be confident about ZEN not being ocked..

You know how to read, what is written in the page i linked..?.

http://www.ashesofthesingularity.com/metaverse#/ladders/benchmark/overall/Low_1080p?viewType=myself&filters={"gpu":"Radeon (TM) RX 480 Graphics","gameVersion":"1.30.21168.0"}

It is written "Benchmark" and the higher score is 68 FPS and 6600 as total score.

 

[cytg111]

What is wrong with you two? (read : edit posts before, you know...)

At any rate, Zen looks to be AMDs Nehalem moment and given the incremental advances we've seen from Intel since Sandy I suspect it will be easier for AMD to catch up.. with Moore dead and all that.

 

[Sweepr]

That s the results of the benchmark, so it s you that are picking datas that have no relevance with the CPU actual perfs

Talking about yourself and your overall framerate results? Too bad they still can't hide the masive advantage a plain 4C/4T 6600K still has over 8C/16T Summit Ridge ES here:

GPU:
RX480 + 6600K - Average: 65.0 FPS
RX480 + Summit Ridge ES - Average: 54.3 FPS

CPU:
6600K - Average: 84.1 FPS
Summit Ridge ES - Average: 58.9 FPS

http://www.ashesofthesingularity.co...-details/03d8944d-b6ef-4c32-8dab-37e71eebe47e

you should add a 4790K so people can see that you re using whatever fill your needs to mislead people.

Name calling doesn't change the fact that the numbers are not in your favor. This is not SemiAccurate.

 

[Abwx]

Using overall framerate results to compare CPUs is just bad faith. Too bad that the numbers are not in your favor, either way.

So let s use the CPU framerate :

- Core i7-6700K 4C/8T - 4.0 GHz (2015 Skylake)
Average: 102.8 FPS
Normal batch: 114.8 FPS
Medium batch: 110.7 FPS
Heavy batch: 87.4 FPS

- Core i7-4790 3.60GHz

Average: 62.5 FPS
Normal batch: 68.3 FPS
Medium batch: 64.5 FPS
Heavy batch: 56.0 FPS


SKL has that much better perf than a HW.?.

Yet you call this the most accurate comparison for CPU perf, Lol, bad faith you said..?..

Beside the 6600K in your exemple does not use the correct setting, do your homework as you re just using irrelevant datas.

The above figure are from the already posted link wich is the relevant version of the game contrary to your misleading postings :

http://www.ashesofthesingularity.com/metaverse#/ladders/benchmark/overall/Low_1080p?viewType=myself&filters={"gpu":"Radeon (TM) RX 480 Graphics","gameVersion":"1.30.21168.0"}

 

[Sweepr]

So let s use the CPU framerate :
SKL has that much better perf than a HW.?.

6700K might be overclocked here. Still doesn't change the fact that a stock 4C/4T Haswell/Skylake-S is ahead of 8C/16T Summit Ridge in this benchmark. Using results from the links provided by you (above):

GPU:
RX480 + 6600K - Average: 65.0 FPS
RX480 + Summit Ridge ES - Average: 54.3 FPS

CPU:
6600K - Average: 84.1 FPS
Summit Ridge ES - Average: 58.9 FPS

http://www.ashesofthesingularity.co...-details/03d8944d-b6ef-4c32-8dab-37e71eebe47e

Yet you call this the most accurate comparison for CPU perf, Lol, bad faith you said..?..

Better than overall framerate comparisons including GPU performance? No doubt.

 

[Abwx]

6700K might be overclocked here. Still doesn't change the fact that a stock 4C/4T Haswell/Skylake-S is ahead of 8C/16T Summit Ridge in this benchmark.

It is ahead of the overclocked 6700K and of the 6950K as well, and also ahead of Guru3D 3.2 5960X, so much for your "fact"...

 

[Sweepr]

It is ahead of the overclocked 6700K and of the 6950K as well, and also ahead of Guru3D 3.2 5960X, so much for your "fact"...

It isn't. Didn't want to embarass you, but it's in your own links (again):

- Core i7-6700K 4C/8T - 4.0 GHz (2015 Skylake) + OC
Average: 107.3 FPS
http://www.ashesofthesingularity.co...-details/6096dd00-7e11-4368-b588-97413922f5c7

- Core i7-5960X 8C/16 - 3.0-3.5 GHz (2014 Haswell-E)
Average: 109.8 FPS
http://www.ashesofthesingularity.co...-details/5d5ab818-bc9e-4340-8a33-9e243a097a1f

 

[AtenRa]

Ok what the hell ??

When i run my system with 2154MHz ram and 1614MHz ram, Game Framerate also got decreased.

2154MHz ram
Franerate = 57,7fps
CPU Framerate = 87,6fps

1614MHz ram
Franerate = 53,4fps
CPU Framerate = 75,9fps

That means that CPU and Memory ALSO affecting game Framerates.

But if you look at the Core i5 6600K vs Core i7 5960X both with RX 480, Core i7 5960X has lower Game Framerates than Core i5 6600K but the CPU Framerate is higher.


Core i5 6600K
Franerate = 69,8fps
CPU Framerate = 82,4fps

Core i7 5960X
Franerate = 65,5fps
CPU Framerate = 109,8fps

So Core i7 5960X should have higher Framerates than 6600K because of its higher CPU Framerates.

 

[Sweepr]

65 vs 69 FPS is close, could be down to different RX 480 models, GPU clocks, stock vs OC, etc.

 

[Abwx]

No it isn't. Didn't want to embarass you, but it's in your own links (again):

- Core i7-6700K 4C/8T - 4.0 GHz (2015 Skylake) + OC
Average: 107.3 FPS
http://www.ashesofthesingularity.co...-details/6096dd00-7e11-4368-b588-97413922f5c7

- Core i7-5960X 8C/16 - 3.0-3.5 GHz (2014 Haswell-E)
Average: 109.8 FPS
http://www.ashesofthesingularity.co...-details/5d5ab818-bc9e-4340-8a33-9e243a097a1f

According to CPU framerate SKL is more than 60% faster than a HW, so as already pointed you are picking the numbers that fill your need in function of the comparison, if CPU framerate is accurate for ZEN/SKL then it should be accurate for SKL/HW as well, wich is obviously not the case but nevermind, it suit your usual bias.

So Core i7 5960X should have higher Framerates than 6600K because of its higher CPU Framerates.

As pointed by your test the CPU framerate gain more from faster RAM bandwith than the game FPS, the 5960X has higher CPU framerate because it has a quad memory channel, SKL use DDR4 while HW use DDR3, that s the reason of the vastly different CPU framerates numbers.

The 6600K is obviously highly ocked, here a result that seems more in line with the CPU ratings and also with the relevant settings :

http://www.ashesofthesingularity.co...-details/c3ea33e6-b8ef-4723-bec0-6bc51b0c971c

 

[Sweepr]

According to CPU framerate SKL is more than 60% faster than a HW, so as already pointed you are picking the numbers that fill your need in function of the comparison, if CPU framerate is accurate for ZEN/SKL then it should be accurate for SKL/HW as well, wich is obviously not the case but nevermind, it suit your usual bias.

Wrong again. Average CPU FPS using RX 480, same version of the benchmark @ 1080p High:

i5-4690K (3.5 GHz): 63.4 FPS
i5-6500 (3.2 GHz): 59.2 FPS

www.ashesofthesingularity.com/metaverse#/personas/b9adb998-fbed-4274-ba39-41c59177083d/match-details/95779a26-f54e-497f-948e-ccbe43e6ac53
www.ashesofthesingularity.com/metaverse#/personas/56353b2f-d80e-4fdc-9b81-06c885c534f7/match-details/3e964390-5a99-4122-8bab-41d7e0ecee75

 

[Abwx]

Wrong again. Average CPU FPS using RX 480, same version of the benchmark @ 1080p High:

i5-4690K (3.5 GHz): 63.4 FPS
i5-6500 (3.2 GHz): 59.2 FPS

www.ashesofthesingularity.com/metaverse#/personas/b9adb998-fbed-4274-ba39-41c59177083d/match-details/95779a26-f54e-497f-948e-ccbe43e6ac53
www.ashesofthesingularity.com/metaverse#/personas/56353b2f-d80e-4fdc-9b81-06c885c534f7/match-details/3e964390-5a99-4122-8bab-41d7e0ecee75

Your two exemple have different settings than the one used for Zen, you can check at WCCF.

 

[Sweepr]

Your two exemple have different settings than the one used for Zen, you can check at WCCF.

It's an Intel vs Intel comparison, and they are valid to prove this claim:

According to CPU framerate SKL is more than 60% faster than a HW

False.

 

[Abwx]

It's an Intel vs Intel comparison, and they are valid to prove this claim:



False.

Lol...

I posted the numbers, incredible the denial...

So let s use the CPU framerate :

- Core i7-6700K 4C/8T - 4.0 GHz (2015 Skylake)
Average: 102.8 FPS
Normal batch: 114.8 FPS
Medium batch: 110.7 FPS
Heavy batch: 87.4 FPS

- Core i7-4790 3.60GHz

Average: 62.5 FPS
Normal batch: 68.3 FPS
Medium batch: 64.5 FPS
Heavy batch: 56.0 FPS

http://www.ashesofthesingularity.com/metaverse#/ladders/benchmark/overall/Low_1080p?viewType=myself&filters={"gpu":"Radeon (TM) RX 480 Graphics","gameVersion":"1.30.21168.0"}