[Coolaler] AMD Piledriver FX Engineering Sample Benchmarks

[james 1]

Link: http://translate.google.com/transla...%E7%94%A8!?s=a5f4ee8fe8128f8b2210d3c77e775853

There is a thread over on oc: http://www.overclock.net/t/1291114/coolaler-amd-piledriver-fx-vishera-engineering-sample-benchmarks

 

[dac7nco]

That's pretty depressing.

 

[pelov]

I worked it out to a ~4.5% IPC bump from the 8150 for an 8350.

http://hardforum.com/showpost.php?p=1039012872&postcount=141

That's *if* that ES sample is indeed the FX8300 which is the 95W TDP 4-module part.

The 8350 should be just a hair over the stock 2600K scores in cinebench 11.5, so the multi-threaded gains look to be pretty good. The single-threaded gains will probably not be as good due to AMD opting to go with the higher base clocks over the higher clocking single module turbo speeds.

 

[Smoblikat]

Clock for clock its almost identical to FX81XX, this sucks

 

[Ventanni]

I'd like to see a little more than just Cinebench scores.

 

[ShintaiDK]

Oh it happend again...

 

[OVerLoRDI]

Clock for clock its almost identical to FX81XX, this sucks

Well we don't really know what spin this is. Retail version might be better. Just trying to be optimistic..

 

[pelov]

Clock for clock its almost identical to FX81XX, this sucks

Uh... it isn't what the fanboys were expecting. I saw some people claiming a 15% IPC bump, which is nuts and would only happen if the original design had a serious error. This just wasn't the case, rather it had a lot of little errors The clocks seem to be pretty good, though. Still not at that 4.5ghz range that they initially planned for, but I'd reckon there's still a 10-20% increase in performance from Zambezi > Vishera if you take into account the IPC bump and clock speed gains.

I'd like to see a little more than just Cinebench scores.

Yea, I do too. It's difficult to judge it's performance with just Cinebench, especially just the multi-threaded benchmark. Something like Queens would be useful in order to see the branch prediction improvements and their tweaking of the L1.

 

[Olikan]

MOAR benchmarks

 

[Sweepr]

Looks like a single digit IPC increase. Coupled with the increased clock it should look a bit better against IB than BD did against SB back in 2011 (if Intel doesn't release higher clocked IBs), but considering there is no Steamroller-based FX on AMD 2013 client roadmaps I'd say 4-core Haswell is its main competitor (not IB). When it arrives in Q4 it will be a few months away from Haswell. Haswell should bring an IPC boost over IB (which is already ~5% faster CPC than SB, BD's main competitor), OC improvements and AVX 2.0.

 

[moonbogg]

Those cinebench scores look a little suspicious to me anyway. But, thats beside the point. The below text told me everything I need to know right away about aMd's new chip.

"CPU-Z-ah ....... Han bowl of cream, I do not know!"

 

[pelov]

WTB useful benchmarks

 

[pantsaregood]

Did the CPU-NB come factory clocked at 2.2 GHz? If so, there's a significant part of the 5% IPC increase. This is completely unimpressive, just like I expected. What exactly did AMD change, anyway? It isn't even a die shrink.

 

[pelov]

The 8150 is also at 2.2ghz.

 

[Hypertag]

I expected to see actual analysis and rational discussion on that topic, but instead all I saw was even more disturbed AMD fanboys than I see here.

Honestly, this post just kind of scares me. This guy is literally crying because AMD's new processor will be 5% better than its old processor, which still means it fails to compete with AMD's 1090t from April of 2010.

 

[Ventanni]

Honestly, this post just kind of scares me. This guy is literally crying because AMD's new processor will be 5% better than its old processor, which still means it fails to compete with AMD's 1090t from April of 2010.

Lol I don't think he's "literally" crying. I think, like many, he's just disappointed.

 

[MaxPayne63]

Is there no NDA on this, or is the risk of a pink slip nonexistant? Can't help but notice what I would assume is a serial number for the ES.

 

[Rvenger]

Guys remember that the PD CPU being tested is the lowest tier 8 core. There are actually 2 more steps above this CPU. So compare to a 2500k at most IMO. If they manage to fix the power consumption problem while gaining an overall 10% in singlethreaded performance, I think it will be good, especially for upcoming Windows 8.

 

[Olikan]

...lol, it's fake...( or a bug in ES )

PD have worst IPC than BD in that winRAR benchmark

 

[nyker96]

Is there no NDA on this, or is the risk of a pink slip nonexistant? Can't help but notice what I would assume is a serial number for the ES.

The only risk here is for AMD to have another flop!

Intel = tick, tock ...
AMD = flip, flop ...

;[

 

[piesquared]

Looks like a nice performance boost good job AMD, and the positive feedback around the web even outweighs the intel fanboys' and shills' propaganda, that's a good sign also. It will be interesting to see how the resonant clock mesh affects clocks and power consumption also. I don't expect intel's investor group to do anything but bash any product that might hurt their own wallet and would rather see all competition cease to exist, but the fact is that this is looking promising.

 

[Nemesis 1]

Looks like a nice performance boost good job AMD, and the positive feedback around the web even outweighs the intel fanboys' and shills' propaganda, that's a good sign also. It will be interesting to see how the resonant clock mesh affects clocks and power consumption also. I don't expect intel's investor group to do anything but bash any product that might hurt their own wallet and would rather see all competition cease to exist, but the fact is that this is looking promising.

Link to all this positive feed back on web or I call trolling. I was hearing befor this 20-25% IPC what happened . What was it again that you said it would be a more than 1 topic. Intel can increase clock speads at will if they desire to .But thats really not IPC is it?

 

[Nemesis 1]

Guys remember that the PD CPU being tested is the lowest tier 8 core. There are actually 2 more steps above this CPU. So compare to a 2500k at most IMO. If they manage to fix the power consumption problem while gaining an overall 10% in singlethreaded performance, I think it will be good, especially for upcoming Windows 8.

Ok . Good post . 1 question tho . whats windows 8 going to do for AMD pile ? Also did you notice that this 3rd tear cpu was using 124 watts. Intels are 95 watts . So If intel wanted to they could bring out a 125 watt. IB./ For some reason everyone wants to compare to intels last generation yet again . really a sad lot.

 

[moonbogg]

CPU-Z-ah ....... Han bowl of cream, I do not know!

 

[Vesku]

I think the refresh will be closer to what I was expecting out of original bulldozer. A bit behind Nehalem in single threaded but able to better run 8 threads.

My main concern is how long it's taking to revise, it might be released this year.. maybe.

Doesn't seem like desktop Trinity will be in the retail components channel until October and that's something that has shipped to OEMs. Luckily for them IB hasn't raised the bar that much from SB.