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

[cytg111]

About the hilarious TDP discussion.

Zen idle: 93W
Intel idle: 106W

Zen load: 187W
Intel load: 191W

Zen delta: 94W
Intel delta: 85W

So tell me, who has the highest performance per watt?

I dont think you thought this one through. (struggling?)
From those numbers alone ... I would have to go with Zen.

 

[Insert_Nickname]

I take this opportunity to congratulate AMD management and marketing: Never in the history of AMD have so many enthusiast Intel cores been used to render an AMD CPU logo!

That almost cost me a new keyboard. I think this one is up there for quote of the year...

 

[majord]

SMT Yield is pretty huge across all official versions.. 1.6-1.8x on Skylake i3. If Zen gets higher SMT yield (percentage wise) , which given it's design is quite likely, then that would explain the use of Blender.

I dont think you thought this one through. (struggling?)
From those numbers alone ... I would have to go with Zen.

Depends if you're talking Platform or CPU.

 

[CatMerc]

About the hilarious TDP discussion.

Zen idle: 93W
Intel idle: 106W

Zen load: 187W
Intel load: 191W

Zen delta: 94W
Intel delta: 85W

So tell me, who has the highest performance per watt?

So Zen has better control of its power draw when idle?

 

[realibrad]

About the hilarious TDP discussion.

Zen idle: 93W
Intel idle: 106W

Zen load: 187W
Intel load: 191W

Zen delta: 94W
Intel delta: 85W

So tell me, who has the highest performance per watt?

Lol what?

Zen does, because at every point zen has lower power. Intel has a higher number. Why are you measuring delta?

If my Car gets better mpg at 10mph, and gets better mpg at 30mph and better mpg at 75mph, then I have the more efficient car. Just because my car uses 50% of the gas of your car at 10mph, and 95% at 75mph, does not somehow mean that my car is less efficient.

 

[rvborgh]

Just for giggles...

i gave it a go on my 48 core quad Opteron 61xx machine (SuperMicro H8QGi-F w/48 K10 cores overclocked to 3.0 GHz) ...

results:
24.61 seconds... while consuming 931 watts at the wall Samples was set at 200 for this run.
16.23 seconds... samples set to 128
12.86 seconds... samples set to 100

color me very impressed with Zen

 

[Atari2600]

About the hilarious TDP discussion.

So tell me, who has the highest performance per watt?

Its still an engineering sample. Hence their non use of turbo.

It would be hilariously misguided to compare the two in much detail.

 

[superstition]

Zen delta: 94W
Intel delta: 85W

So tell me, who has the highest performance per watt?

That's not bad considering Intel has a better process to work with.

 

[superstition]

If you start up Blender and don't change any settings, the sampling rate is 128. This is fairly close to the 140 we speculate they used in the live demo, so it could be in fact 128.

But the file, when opened without changing any settings, is 200.

 

[witeken]

So Zen has better control of its power draw when idle?

Lol what?

Zen does, because at every point zen has lower power. Intel has a higher number. Why are you measuring delta?

If my Car gets better mpg at 10mph, and gets better mpg at 30mph and better mpg at 75mph, then I have the more efficient car. Just because my car uses 50% of the gas of your car at 10mph, and 95% at 75mph, does not somehow mean that my car is less efficient.

Power does not depend solely on the CPU. We don't know what motherboard and other stuff AMD put in the test config, because >100W idle is ridiulcously high.

As a matter of fact, any modern CPU should dissipate close to 0W at idle. It gets power gated. Or do you think laptops also consume 100W at idle? I'm quite sure my laptop has about ~10 hours of battery life (which also includes the screen power consumption) when I'm just browing.

So idle CPU power consumption is 0W. Then we just have to look at how much power is added when the system is in load because that power will be coming from the CPU.

AnandTech also uses power delta, so it just shows how biased people are if you dismiss this calculation.

But here are some idle power consumptions:

The last Intel CPU to get >100W idle is Ivy Bridge.

So AMD probably attached a 50W light bulb to the Broadwell system to get similar idle powers so they wouldn't get humiliated?

 

[cytg111]

Depends if you're talking Platform or CPU.

I understand what was Implied(it was bolded out after all), point is it doesnt matter if it is 10 small smurfs burning the joules or what. it takes x watts to perform y work. From a to b. Now you could begin arguing that there was an a2 a3 or even an a4 in between there: The travelling salesman dont care.

 

[kraatus77]

About the hilarious TDP discussion.

Zen idle: 93W
Intel idle: 106W

Zen load: 187W
Intel load: 191W

Zen delta: 94W
Intel delta: 85W

So tell me, who has the highest performance per watt?

So now the goalpost has been moved to delta ? and it doesn't matter that zen is using less power at idle and load.

hilarious indeed.

 

[PliotronX]

That's not bad considering Intel has a better process to work with.

It's actually not a valid point. To demonstrate the inanity of including the delta:

Zen idle: 0W
Intel idle: 106W

Zen load: 94W
Intel load: 191W

Zen delta: 94W
Intel delta: 85W

Utterly pointless.

 

[Nothingness]

I take this opportunity to congratulate AMD management and marketing: Never in the history of AMD have so many enthusiast Intel cores been used to render an AMD CPU logo!

I'd be more interested to know how many AMD vs Intel CPU are in their design/validation clusters...

 

[witeken]

So now the goalpost has been moved to delta ? and it doesn't matter that zen is using less power at idle and load.

hilarious indeed.

If you are going to argue that a CPU uses 100W at idle, your are pretty crazy sir.

Power delta is the only comparison that is even close to being scientific in the absens of other data.

 

[Abwx]

About the hilarious TDP discussion.

Zen idle: 93W
Intel idle: 106W

Zen load: 187W
Intel load: 191W

Zen delta: 94W
Intel delta: 85W

So tell me, who has the highest performance per watt?

Perf/watt is comparable, Zen has higher dynamic power, but BDW-E is apparently more leaky (or Zen has better power managements features..) as hinted by the difference at idle.

The result is that at the plateform level Zen is slightly more efficient, and that s with a dual channel configuration for the 6900K as well, with more channels this would inflate the power numbers more than the perfs ones, in the handbrake demo the perf/Watt at the plateform level should exceed 10% in favour of Zen.

 

[kraatus77]

If you are going to argue that a CPU uses 100W at idle, your are pretty crazy sir.

Power delta is the only comparison that is even close to being scientific in the absens of other data.

it's obvious it wasn't idle power of cpu alone. probably full system from wall or psu. thus above 170w load power consumption.

you don't think full system consumes 0w without load, do you ?

 

[piesquared]

Power does not depend solely on the CPU. We don't know what motherboard and other stuff AMD put in the test config, because >100W idle is ridiulcously high.

As a matter of fact, any modern CPU should dissipate close to 0W at idle. It gets power gated. Or do you think laptops also consume 100W at idle? I'm quite sure my laptop has about ~10 hours of battery life (which also includes the screen power consumption) when I'm just browing.

So idle CPU power consumption is 0W. Then we just have to look at how much power is added when the system is in load because that power will be coming from the CPU.

AnandTech also uses power delta, so it just shows how biased people are if you dismiss this calculation.

But here are some idle power consumptions:

The last Intel CPU to get >100W idle is Ivy Bridge.

So AMD probably attached a 50W light bulb to the Broadwell system to get similar idle powers so they wouldn't get humiliated?

I think this is yours, you must have dropped it.

Trolling is not allowed
Markfw
Anandtech Moderator

 

[superstition]

But here are some idle power consumptions:

Interesting chart. Their 8320E is idling at the same level it consumes under full load?

 

[lixlax]

The idle numbers look like if they had a monitor attached in there as well.

 

[raghu78]

it's obvious it wasn't idle power of cpu alone. probably full system from wall or psu. thus above 170w load power consumption.

you don't think full system consumes 0w without load, do you ?

Thats the problem. He is making assumptions to suit his narrative that AMD cheated. Anyway its going to be 2-3 months at most before we know the final perf and perf/watt numbers.

 

[frozentundra123456]

Lol what?

Zen does, because at every point zen has lower power. Intel has a higher number. Why are you measuring delta?

If my Car gets better mpg at 10mph, and gets better mpg at 30mph and better mpg at 75mph, then I have the more efficient car. Just because my car uses 50% of the gas of your car at 10mph, and 95% at 75mph, does not somehow mean that my car is less efficient.

Well, the point is that there may be more efficient platforms for the intel system. I am sure AMD spent many hours optimizing their system, and not the intel one. But in any case, the numbers are basically equal *in this particular test chosen and set up by AMD*. All the AMD fans are talking about how great the benchmarks are, etc. but basically I am only taking two things from this demo.
1. The base clocks are pretty good (better than I expected), although we dont know ultimate overclocking speeds, which will be critical in this segment.
2. Power efficiency is also competitive.

Lets face it, all the other tests, whether powerpoint slides or demos, are still staged by AMD, and just like every manufacturer, most likely reflect best case scenario for their product.

 

[bjt2]

Serial code is code with a single thread in the language its written in.

But that is not a code without branches. Do you write much code without any if-else statements? Or what about for or while loops?

Ah. now i get it. You are saying that on single thread code, the bubble given by branch misprediction (but also cache miss and other interdependencies) are not exploited by a second thread.
But this is true also for INTEL, though...

 

[cytg111]

"All the AMD fans"? I dont see a single one. Eye of the beholder I guess.

 

[KTE]

Its easy to manipulate an experiment for a desired result.

Hence why independent repeatability is a key factor for result validity.

For those running away with the eulogies... Way too early days yet.


Gotta admit, the clocks are impressive. Every model above 3.4GHz base under full load would be very good.

Why no CPU Power or Single thread test, tho?

Currently they've only shown 8Cores+SMT favouring tests. MOAR CORES marketing.

How does that Handbrake version scale with cores/smt?

Altho bjts incredulous theory of MUCH lower power/FO4 and MUCH higher clocks than XV on 8core has failed. It's a nice sanity check.

Sent from HTC 10
(Opinions are own)