Zen is faster than Broadwell-E clock for clock

[rattacko123]

AMD has released some performance comparisons between the 6900k and a Zen 8 core both running at 3.0Ghz

Quote

Using Blender to measure the performance of a rendering workload (a Zen CPU mockup of course), AMD ran an 8-core / 16-thread Zen processor at 3.0 GHz against an 8-core / 16-thread Broadwell-E processor at 3.0 GHz (likely a fixed clocked Core i7-6900K). The point of the demonstration was to showcase the IPC improvements of Zen and it worked: the render completed on the Zen platform a second or two faster than it did on the Intel Broadwell-E system.

How the Zen Architecture works:

 

Source: http://www.pcper.com/reviews/Processors/AMD-Zen-Architecture-and-Performance-Preview?utm_source=twitterfeed&utm_medium=twitter

[iDeFecZx]

So for the AMD CPU to be as fast, the Intel one need to be handicapped?

[TheRandomness]

Was this supposed to be in the news and reviews section? Also, AMD has released meaning it's probably not as good as that. 

[Katsunaka]

It's easy to compare the high end Zen chip to a handicapped 6900k to throw more gas on the hype fire.

 

Lets gimp a 6900k, compare to what will be likely the high end Zen chip and post some more hype.

 

I gave up on AMD a long while ago, hype leads you nowhere. 

[zMeul]

2 minutes ago, iDeFecZx said:

So for the AMD CPU to be as fast, the Intel one need to be handicapped?

what did you expect?! it's AMD who we're talking about

no benchmark done by AMD should be trusted

[TheRandomness]

Just now, Wikiforce said:

Some Rumors say it will only cost 99$

 

No. Just no. 

[Nena Trinity]

6 minutes ago, TheRandomness said:

Was this supposed to be in the news and reviews section? Also, AMD has released meaning it's probably not as good as that. 

AMD claimed a FX-8370 4.3GHz gets 638, yet mine gets 680...

[TheRandomness]

Just now, Nena360 said:

AMD claimed a FX-8370 4.3GHz gets near 650, yet mine gets 680...

AMD claimed a few years ago* Program optimisations might've helped with that c:

[Nena Trinity]

Just now, TheRandomness said:

AMD claimed a few years ago* Program optimisations might've helped with that c:

Cinebench gets many updates?

[Paragon_X]

Why you say that the 6900K is handicapped, they both run on 3.0Ghz. even the Engineering sample was running at 3.2/3.7 Boost (same as stock 6900). Its clearly an IPC benchmark. and the results are exciting, maybe some competition atlast.

[BrightCandle]

At least we have some basic confirmation on SKUs of 8c16t and its hitting its performance levels in something real like. The clock for clock comparison is nice but its 1 app and AMD has a history of showing its best results for marketing purposes (AoS for its graphics cards for example) and also they have changed the Intel CPU from reference which might suggest they can't get the clockspeeds.

 

There are 3 factors to determine whether anyone cares about this product. When it releases, what it costs and what clockspeed it comes at. If its too late then Intel will have Skylake-E out and this will be behind on IPC already, if its too expensive its not going to appeal to the market that would buy it and its unlikely business customers will switch in droves and the clockspeed is going to determine whether it actually competes with Broadwell-E or not.

 

Still its a little better than we hoped, its slower core for core than a CPU Intel has had out for a year now but its got the cores to compete with what Intel has out but not what it will likely have out before it releases. I think its still too late but the price/performance might change its fortunes.

[TheRandomness]

3 minutes ago, Nena360 said:

Cinebench gets many updates?

Why wouldn't it have an update (however small) in a few years?

[Nena Trinity]

Just now, TheRandomness said:

Why wouldn't it have an update (however small) in a few years?

I guess it may but I got my Cinebench R15 shortly after FX-8370 was released and never swapped the .exe file

[Mira Yurizaki]

I'm more concerned with Zen's single threaded performance than anything. For what most people use computers for, single threaded performance is still better to have. And I'm also wondering what Zen's TDP and power consumption is. AMD can compete all it wants on IPC, but if it's perf/watt is lacking, it's going to be sidelined at the server market and maybe in the laptop market.

 

But I wouldn't be surprised if AMD sells the Zen they were demoing for the same as an i7-6700K just to undermine Intel.

[Morgan MLGman]

12 minutes ago, Nena360 said:

I guess it may but I got my Cinebench R15 shortly after FX-8370 was released and never swapped the .exe file

My FX-8350 @4,53 gHz  scored around 720 points in Cinebench More than my current i7-3770 @3,9gHz with Turbo to 4,3. By a few points, but still.

[AresKrieger]

I highly doubt the ludicrous claim made in the title of this post as even by there own processing goals it would be below haswell in terms of ipc

 

Also trusting any company whether it be intel, amd, or Nvidia to do proper testing for marketing is insanity

[Dabombinable]

23 minutes ago, Nena360 said:

AMD claimed a FX-8370 4.3GHz gets 638, yet mine gets 680...

I still find it funny that when you add the omitted power consumption-the FX 8370 and 9590 just look like shit in comparison.

[Atmos]

22 minutes ago, Katsunaka said:

It's easy to compare the high end Zen chip to a handicapped 6900k to throw more gas on the hype fire.

 

Lets gimp a 6900k, compare to what will be likely the high end Zen chip and post some more hype.

 

I gave up on AMD a long while ago, hype leads you nowhere. 

 

25 minutes ago, iDeFecZx said:

So for the AMD CPU to be as fast, the Intel one need to be handicapped?

 

You guys are 100% missing the point.

Its not a handicapped 6900k, they're not saying "Hey, this product we're showcasing here is just as fast as the 6900k!" What they're saying, and I quote here is "...The point of the demonstration was to showcase the IPC improvements of Zen"

All of amd's previous cpu launches have had awful ipc compared to intel. Clock per clock, core per clock, they lost every time. Now they have a product with what they're claiming is a higher ipc, and that's massive. Okay, maybe not massive, but it sure as shit should give intel a kick in the pants and actually force them to advance their own damn chips on the consumer end. That's what this release is about; not anything like saying that its better than a 6900k, they never claimed it to be except when compared clock per clock, core per core. 

 

That also means that, if this cpu is capable at running at the same speeds as a 6900k normally, then it would be faster than it. However, the way that they decided to display this information, I think its very likely going to mean that the consumer chip from amd is going to have a lower clock speed than the 6900k.

[AresKrieger]

25 minutes ago, Wikiforce said:

Some Rumors say it will only cost 99$

 

If those rumors are are true then the performance is terrible, AMD will not sell products for drastically under market value. So clearly that's a load of crap since AMD wants to up their performance not downgrade it.

[Dabombinable]

1 minute ago, Atmos said:

 

 

You guys are 100% missing the point.

Its not a handicapped 6900k, they're not saying "Hey, this product we're showcasing here is just as fast as the 6900k!" What they're saying, and I quote here is "...The point of the demonstration was to showcase the IPC improvements of Zen"

All of amd's previous cpu launches have had awful ipc compared to intel. Clock per clock, core per clock, they lost every time. Now they have a product with what they're claiming is a higher ipc, and that's massive. Okay, maybe not massive, but it sure as shit should give intel a kick in the pants and actually force them to advance their own damn chips on the consumer end. That's what this release is about; not anything like saying that its better than a 6900k, they never claimed it to be except when compared clock per clock, core per core. 

 

That also means that, if this cpu is capable at running at the same speeds as a 6900k normally, then it would be faster than it. However, the way that they decided to display this information, I think its very likely going to mean that the consumer chip from amd is going to have a lower clock speed than the 6900k.

With the shit Intel, AMD and Nvidia have pulled in the past, do not believe the benchmark at all and wait until Zen is in full production and reviewed.

[zMeul]

27 minutes ago, Paragon_X said:

Why you say that the 6900K is handicapped, they both run on 3.0Ghz. even the Engineering sample was running at 3.2/3.7 Boost (same as stock 6900). Its clearly an IPC benchmark. and the results are exciting, maybe some competition atlast.

that CPU has a base clock of 3.2Ghz and a boost clock of 4Ghz: http://ark.intel.com/products/94196/Intel-Core-i7-6900K-Processor-20M-Cache-up-to-3_70-GHz

AMD's benchmarking was done at 3Ghz, by literally gimping the Intel CPU - lower freq translates into lower IPC

 

so what is AMD trying to prove here, that their ZEN based CPU has same IPC (or better) than a i7 6900K?

because that's false and it's a marketing move to distract from the real performance of ZEN - ZEN isn't at Broadwell-E level

[ThinkWithPortals]

31 minutes ago, iDeFecZx said:

So for the AMD CPU to be as fast, the Intel one need to be handicapped?

 

27 minutes ago, Katsunaka said:

It's easy to compare the high end Zen chip to a handicapped 6900k to throw more gas on the hype fire.

I think you're both missing the point.

The point was not that AMD is gimping the 6900K to make the new Zen octacore look better. The point is that the new AMD chip with 8 cores at 3GHz beats the new Intel chip with 8 cores at 3GHz, meaning that the AMD chip supposedly has better IPC as the clockspeed and core count are the same. They're clocking down the 6900K to make the results more comparable.

 

Inb4 I get called an AMD fanboy for this

[Katsunaka]

1 minute ago, Dabombinable said:

With the shit Intel, AMD and Nvidia have pulled in the past, do not believe the benchmark at all and wait until Zen is in full production and reviewed.

Effin A Right.

 

Hype is nothing. All of them use it to get people started on forums. No point in getting hyped up until you get a tracking number from UPS/Fedex. 

[zMeul]

5 minutes ago, ThinkWithPortals said:

I think you're both missing the point.

The point was not that AMD is gimping the 6900K to make the new Zen octacore look better. The point is that the new AMD chip with 8 cores at 3GHz beats the new Intel chip with 8 cores at 3GHz, meaning that the AMD chip supposedly has better IPC as the clockspeed and core count are the same. They're clocking down the 6900K to make the results more comparable.

no, you are missing the point

because if you downclock, you get lower IPC and that is getting in the range of previous generation of Intel CPUs, like Haswell-E, for example - and that is what AMD needs to hide

 

also, Kabilake is just around the corner and will probably see an earlier release than ZEN - that's another IPC increase over Broadwell-E

 

so ... who's AMD kidding!?

[Paragon_X]

2 minutes ago, zMeul said:

that CPU has a base clock of 3.2Ghz and a boost clock of 4Ghz

AMD's benchmarking was done at 3Ghz, by literally gimping the Intel CPU - lower freq translates into lower IPC

 

so what is AMD trying to prove here, that their ZEN based CPU has same IPC (or better) than a i7 6900K?

because that's false and it's a marketing move to distract from the real performance of ZEN - ZEN isn't at Broadwell-E level

IPC was their main problem in the previous generation and no it does not lower the IPC. IPC means Instructions per Cycle (1 hertz) that means whether the clockspeed is lower or higher it does not make a difference in the actual IPC count.