AMD ZEN Engineering Sample AOS - Further Performance Analysis

that would mean that:

• AMD still hasn't figured out the process quirks

{AND/OR}

• Zen is so powerful that they purposly chose to leave performance on the table
• its power bill can support entire village for a month

 

It's interesting how consistent these results are with the promised 40% IPC improvement.

 

I guess many people seriously underestimated how awful fx cpus are, including myself, though I had one.

 

I will happily take a 4/6core hyper threaded APU with a 460 equivalent gpu could be a monster at the right price.

 

Did anyone notice it says 32 nm on CPU-Z, I thought ZEN was 14 nm?

 

CPU-Z with 32 nm = FX 8370, not Zen.

 

Thanks, I'm not always the sharpest tool in the shed

 

Please, read carefully the result of that benchmark: The I7@3.6GHz is just 24% faster than I5@3.4GHz.
Is possible that an Intel 8 threads CPU be just 24% faster than a 4 threads one that run at lower speed? Obviously not. The difference would be arround 100%.
The answer is very clear: That test was made with the 4 processors using just four threads. (Furthermore, it is the most honest way to do the test, and see the real performance difference between all cpu.)
In this scenario, and considering the clock speed, the Zen has the highest FPS/GHz of all. The Zen performance is close to the Skylake.

 

I7 4C/8T cpu ain't 100% faster than I5 4C/4T cpu on raw performance . 100% faster needs 8C/8T cpu.

Generally Hyperthreaded I7 4C/8T chip performance is pretty close to what a 5C/5T cpu would be, if one would exist.

 

I don't know about OC, but I expect a 3.6-4.0 final frequency. Which with close-to-Haswell IPC would be amazing for a 8/16 chip at around $350-$400.

 

What do you even mean by this? That doesn't make any sense. What you're saying implies that Zen is released/close to being released and that this in-fact is a retail sample, not engineering sample. What exactly do you not understand about engineering sample? that is experimental.

...wtf are you....what? no seriously what? my brain hurts reading your comment, wtf are you even trying to say?

Like, the only thing i can even remotely come to with your statement, is MAYBE you are saying that since this EXPERIMENTAL processor is clocked at 2.8/3.2Ghz, that somehow that means AMD screwed up/doesn't know what they are doing, or is so power hungry that they can't get the Mhz up?

What kind of logic is that?

These threads are so idiotically illogical i just...don't even understand how people are coming up with this stuff.

 

You are a bonafide F@#$TARD... This entire article is purely based on theory... And you claim it as "proof"? :bang::3eyes::stewpid:

 

No no. Chaython is right. All his life's accomplishments have brought him to this moment. Where he can look Jim Keller in the eye and laugh and laugh and laugh...

 

Actually looks very impressive given non-final bios, drivers or OS support and arguably on a "weaker" MB chipset than Intels.

If we calculate theoretical FPS, this sample is getting 18.125 FPS per GHZ (58 FPS/3.2 boost clock) vs Haswell @ 16.35 FPS per GHZ (65.4 FPS/4 GHZ boosted). All on RX480 GPU.

If the MB quality is as good as an Intel Z97 chipset, and AMD can get the clocks up, it may be faster than Haswell or close to Skylake which is actually very awesome.

I only wish can AMD makes CPU that can be used on an Intel chipset MB like the OLD days (remember AMD K and K2 series CPUs?) and someone plug in a Fury or faster GPU to see how the cpu scales.

 

You're right.... Sorry, I was most definately wrong... What a tool I am!... :tool:

 

I dont understand why people bash AMD so much for an ES Sample, people doesn't understand that competition is good? AMD was the only one driving Intel prices down in the past, I refuse to pay 1000€ for a cpu, Intel right now is making big money for cpu's that doesn't even cost 100 bucks to make just because they have no competition, I really hope AMD success with their Zen line to force Intel to lower their ridiculous prices, people fight with each other like if they where the ones making money when in reality an Intel fan has to pay a premium price because there's no competition, we the consumers need companies to have competition to drive prices down and to push innovation, and I think AMD is doing much better than it was doing the past 10 years, remember that when a new line of hardware comes in it takes time to improve, they refine the process and they bring better performance with it, Intel has been doing it with their lines for a long time, highly refined architectures but they haven't change the architecture that much, zen will get better with time, and at least now it seems that they are back in the right track.

 

Well I was doing some testing on my own with my 4790k in ashes. I am starting to think whoever ran the benchmark had vsync turned on. When I run at 1080p with vsync, on extreme settings my CPU average FPS is around 60 when I down clock to 3.2 GHz.

With that said when I turn vsync off and my fury goes under max load, the CPU gets put under a greater load (90%+) as well and all of a sudden my CPU FPS jumps into the 80s/90s.

The 480 should be capable of a rather high framerate in the first batch of this ashes test at 1080p, but with vsync on it isn't going to be under 100% load I imagine. That may be the bottle neck your referring to.

 

Its not only the 4 channel memory advantage for the 5960X. The 5960X and the 8370 are given a 400Mhz advantage. Remember the Zen's base clock is 2.8Ghz and its really a guess how long the Zen boosts to stay under the 95watts.

 

Hello Hilbert,

someone noticed that the leaked benchmarks were performed in "GPU mode". Can you confirm whether you used the same? If not, could you run the 5960x and 480 in the same mode to compare.

thanks

 

you also typo..lol