• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

*** AMD "Zen" thread (inc AM4/APU discussion) ***

I don't think we can read very much into a single game benchmark, but it is a pretty positive result nontheless. Significantly more performance than an FX-8350 (which has eight cores) whilst being clocked over 1GHz lower. If engineering samples are going out at 2.8/3.2GHz, then 3.5GHz and up is not unlikely for release. Nice.

Though as others are pointing out, nobody should actually read much into a single leaked ES result.
 
DragonQ said:
If you compensate for the clock speed difference, Zen is faster than the rest in that benchmark figure you showed. It'll be a while before we know how well Zen clocks though. I have no idea what the hell it's testing though, does it include the IGPs or are they using a separate GPU?
Click to expand...

Assuming this is Zen desktop CPU not the APU then they must have used an external GPU.

Edit:

If we normalise the frequencies to 3.6Ghz (assuming the Zen chip is capable of hitting that frequency) the results look like this with the Zen seeing a 28% increase in performance putting it 14% ahead of the I7 4790. Interestingly the hyperthreading gives the 4790 a 16% advantage over the i5.

ylKUdGb.jpg
 
Last edited:
AthlonXP1800 said:
After searched for my CPU frame performance on ATOS standard 1080p benchmark database, surprised to see Zen ES is about 20% SLOWER than my 4 years old 3770K CPU.

Here is list of CPU framerate how powerful Intel CPU performance over the last 5 years compared against Zen ES:

Broadwell-E 6950X 10C/20T: 127.9
Skylake 6770K 4C/8T: 98.8
Devil's Canyon 4790K 4C/8T: 87.7
Haswell 4770K 4C/8T: 69.8
Ivy Bridge 3770K 4C/8T: 69.7
Sandy Bridge 2600K 4C/8T: 70.3
Zen ES 8C/16T: 58
Sandy Bridge 2500K 4C/4T: 47.9

Zen ES running at 3.2GHz turbo CPU performance sit on Sandy Bridge level between 2500K and 2600K, really very poor for 8C/16T. AMD probably will set 3GHz as final base clock in in the next few months that would bring CPU performance closer to 2600K. Look like I definitely will get Kaby Lake 7700K after saw poor Zen ES CPU score in first benchmark, it would be a downgrade from my 3770K CPU if I go for Zen CPU.
Click to expand...

Errm, how about a single link to the 'results' you've found. Because again, since you ignored it before, the benchmarks gives ZERO indication of actual clock speed being run, none at all, and the benchmark scales exceptionally well with clock speed.

http://www.ashesofthesingularity.co...-details/413d8164-4e1a-45d6-8daa-1d0969aac24b

The slowest one shows just 72fps, the fastest one

http://www.ashesofthesingularity.co...-details/8effedd3-5358-44f7-b682-c24ee07a644e

shows 125fps, the middle one(with a cpu score not just gpu score) shows 97fps.

This was to show the range of a more comparable chip with 8 cores and 16 threads.

Incidentally in the database I can see a grand total of TWO 6950 results, neither of which give a result of 127fps, one is 110fps, the other is 138fps... again showing the clock speed differences and no there is no indication that 110fps is 'stock' either, it could be, it could simply be overclocked less than the 138fps result.

EDIT:- For some reason it defaulted to showing only GTX 1080 results, there are 9 entries, the highest is still 138fps, the lowest is actually 45fps, multiple around the 70-80fps range, several at 122fps, one higher than that. It's pretty easy to conclude that >90fps are significantly overclocked, 70-80 is likely the stock 3Ghz range and 45fps is a weird outlier that can likely be discounted.


https://forums.overclockers.co.uk/showpost.php?p=29868418&postcount=204

As shown in the quoted text in that post, the only 6950k result I saw in that thread had a whopping 64fps cpu score when running at the stock 3Ghz clock speeds.

EDIT2:- 2700k scores, ranging from 48fps to 92fps..... what made you pick 70fps as representative of all 2700k? What made you pick any of these numbers and why didn't you link to any of these results? This is before you even get into what game settings can effect CPU performance because I honestly don't know and a huge portion of these results have different settings.
 
Last edited:
I'm skeptical about Zen if i'm honest.

Yeh AMD is better than an I5 in a game that's biased towards AMD.

Real world overall performance based on the AOTS results I'd expect it to be on par with an I5 at best, if not worse.

It could get better it's a sample after all, but to be honest after the hype and expectations of the 480 to what was actually released (It clocks to 1500!!!) I'm going to remain pessimistic about this rather than optimistic.

I'm personally waiting for Zen before I upgrade but I refuse to get caught up in the hype this time.
 
The boost clocks are nonsense you won't get those frequencies on all the cores, they only boost when 1 or 2 cores are being used and there's enough thermal room to ramp up the clock rates. Take the base clocks and work from there.
 
Im optimistic about it, most of the bits ive read over the past few hours paint these results in mostly a positive light.

Its clocked low as its an Engineering Sample, release clocks are expected to be higher.

Its performance if scales as expected, puts it in direct competition with Skylake 6700 or so.

My main concern is AMD price it badly, and judging by the prices of their GPU's recently, i dont hold much hope they release it cheaply.

Also the AOTS benches apparently dont scale well with more cores, not sure whats available right now to check the DX12 performance of it.

In all honesty we wont know until final release versions hit reviewers and they can lob top end GPU's at it to compare how it performs really.
 
It won't be cheap, especially if the performance is there. I paid a small fortune for my S939 setup. Was well worth it, though, as the performance warranted the price.
 
At 2.8 / 3.2 GHz it's faster than Intel at 3.4 / 3.8 GHz but slower than 3.6 / 4.0Ghz, but the whole thing is meaningless..
If the game is using 8 threads then clock for clock the performance is about the same as Intel, if more than 8 threads it's slower.
It's one of those things where anyone can interpret it how they like to push a conclusion, which is exactly what some are already doing.
The fact is with this chart we know absolutely nothing.
 
AthlonXP1800 said:
http://wccftech.com/amd-zen-es-benchmarks/

First Zen ES Ashes of the Singularity benchmark leaked.

http://i.imgur.com/sIaA6DX.jpg

Zen ES has 2.8GHz base clock and 3.2GHz turbo clock but the performance is slower than Haswell 4790 CPU, it not at skylake level AMD claimed months ago.

Zen ES are seemed underwhelmed for 8C/16T, if the final retail are similar or slight improved peformance at higher base and turbo clock then guess I will wait for Kaby Lake.
Click to expand...

So considerably slower than 4790K in one of the few types of games that actually favours more cores...
 
humbug said:
At 2.8 / 3.2 GHz it's faster than Intel at 3.4 / 3.8 GHz but slower than 3.6 / 4.0Ghz, but the whole thing is meaningless..
If the game is using I threads then clock for clock the performance is about the same as Intel, if more than 8 threads it's slower.
It's one of those things where anyone can interpret it how they like to push a conclusion, which is exactly what some are already doing.
The fact is with this chart we know absolutely nothing.
Click to expand...

Far too much reasoned thought in this post.
 
All we can say for sure is that these samples at 3.2Ghz are slower than Intel at 4Ghz, normalised clock rates they look the same. We don't know how many cores are in use, until we do we know nothing about performance, we do know it's more than 4 given the lower clocked Zen is faster than the higher clocked 4 core intel
 
Martini1991 said:
Depends if it performed better when all cores are in use, if not, why would you want more cores but lesser performance overall? If it's 10% slower per core, then that's fine, but if it's 10% slower 8 core versus a 4 core i7 when they're going full whack, then that's not good enough.

Either way, I don't think a single Singularity result is enough to start worrying over.
Click to expand...

Yeah, that's what I was getting at. 10% slower in games due to their love of IPC (at the moment anyway), but then 40-80% faster in properly multi-threaded programs like the Adobe suite, for the same money as an i7 6700k.

If that turned out to be the situation with Zen vs Sky/Kaby lake then I'd very likely go for the more cores.
 
Last edited:
mmj_uk said:
So considerably slower than 4790K in one of the few types of games that actually favours more cores...
Click to expand...

What results are you looking at?

The ones I'm looking at show it (Zen) being considerably faster than a 4790K with what one would expect to be release clocks.

The game does not scale beyond 8 threads. Indeed, the Zen is handicapped as unless the game or application scales to more threads, latency and therefore frame times increase.

Hence why a 4790K or 6700K are both significantly faster than a 6950X / 6900K (even with a little OC'ing) in AotS.

Back of beer mat calculations show an IPC increase of ~45% over Piledriver (FX8350), suggesting that AMD's performance claims have been conservative if anything, and it should easily compete with Skylake in gaming, and the Intel -E platforms in heavily threaded workloads.
 
Last edited:
LFaWnyI.png

XA3NPwx.png

Posted on SA forums.

The shader count for the Zen APU looks to be 768 shaders.

That means if Zen can get even to Haswell levels of single thread performance,that would be a 4C/8T chip with a 768 shader IGP and much lower power consumption levels than the FX8350 too.

It is also apparent,we might be seeing the consumer versions later than the ones for server looking at that roadmap or a soft launch for Zen at the end of the year.
 
Martini1991 said:
We've been seeing over exaggerations to the IGP performance for years now.
The bottleneck has always been the bandwidth.
Click to expand...

I'm not an expert on IGPs but just comparing hype to benchmarks I'd be inclined to agree with you.

As for Zen, well it looks good (huge bucket of salt) but only in the sense that it will be providing some competition to Intel and keeping prices more honest. However, and maybe this is unfair, it does feel like because AMD have been relatively nowhere consumer CPUs are now about a year behind in development from where they should be.
 
You must log in or register to reply here.
Back
Top Bottom