AMD Piledriver rumours ... and expert conjecture

[Reynod]

We have had several requests for a sticky on AMD's yet to be released Piledriver architecture ... so here it is.

I want to make a few things clear though.

Post a question relevant to the topic, or information about the topic, or it will be deleted.

Post any negative personal comments about another user ... and they will be deleted.

Post flame baiting comments about the blue, red and green team and they will be deleted.

Enjoy ...

 

[cgner]

Likewise, I doubt they will simply cancel CPU production altogether or even cut off half a generation.

 

[-Fran-]

AFAIK, Vishera was the last AM3+ CPU and then whatever Steamroller spits, it was going to be AM4 or some sort of FM2 (or 3) socket.

Cheers!

 

[mayankleoboy1]

that OBR guy has released the PD 'benchmarks' .
Result : 4% increase in IPC.

overall increase in percent ~10-15%. Which is a result of higher MHz.

 

[obsama1]

I checked it out. Barely any single-threaded performance improvements, but the improvements are good. But that OBR guy is so biased. Total Intel fanboy.

 

[dattimr]

that OBR guy has released the PD 'benchmarks' .
Result : 4% increase in IPC.

overall increase in percent ~10-15%. Which is a result of higher MHz.

He shows the single-threaded Cinebench scores on the summary, but they do not appear anywhere on the screenshots.

Either way, only a [insert your own adjective here] would test the 8150 with 8 GBs of RAM, and the 8350 with 2048 MBs -- take a close look at the CPU-Z screenshots for the 8150, and the BIOS screenshot for the 8350. Moreover, even if that is indeed Vishera, and if it is using the retail stepping, the 8350 can easily hold its own against the 2600K, 2700K, and even the 3770K.

 

[dattimr]

I checked it out. Barely any single-threaded performance improvements, but the improvements are good. But that OBR guy is so biased. Total Intel fanboy.

Except for the fact that he seems to think that SuperPi is a useful single-threaded benchmark, along with CPUMARK 99 (!) and Cinebench scores that aren't shown on the screenshots. Also, nevermind the fact that the 8350 is running Windows 7 x64 with plentiful 2048 MBs of RAM.

 

[dattimr]

Actually, mind you that, in order to simulate an 8350, he is overclocking an ES by means of some beta BIOS -- those are not the scores for an actual 8350.

 

[jdwii]

Actually, mind you that, in order to simulate an 8350, he is overclocking an ES by means of some beta BIOS -- those are not the scores for an actual 8350.

For some reason this moron always does this, Use the same system spec's, then use good single threaded benchmarks such as Cincebench, Itunes, Also use Wprime and make it use only one core and then run hand brake and 7-zip to check multithreading performance. Why cant people benchmark the right way again with no BIAS.

 

[cgner]

For some reason this moron always does this, Use the same system spec's, then use good single threaded benchmarks such as Cincebench, Itunes, Also use Wprime and make it use only one core and then run hand brake and 7-zip to check multithreading performance. Why cant people benchmark the right way again with no BIAS.

:lol: so true. I guess because there are few non-fanboiz left in this world.

 

[davemaster84]

AMD Rumored to Discontinue Performance CPU’s after Piledriver – Will Focus on APU development
http://wccftech.com/amd-rumored-di [...] velopment/

Rumor: AMD CPU canceled their Vishera?
http://technewspedia.com/rumor-amd [...] r-vishera/

Phil Hughes ‏@AMDphil
https://twitter.com/AMDphil/status/238044301588848643
Quote :


we don't comment on specific rumor/speculation but in terms of Vishera and other 2012 products,
there have been no changes

Actually what Phil is really saying in other words is that those rumors are nonsense and that AMD plans for vishera are intact.

 

[davemaster84]

Actually, mind you that, in order to simulate an 8350, he is overclocking an ES by means of some beta BIOS -- those are not the scores for an actual 8350.

Yeah I agree with ya, there's no way to properly simulate a processor which architecture is brand new and unknown to the public. Lets just wait in the next few days for the real benchmarks.

 

[-Fran-]

Leaks -> grain of salt... Or a ton of it, actually.

Cheers! 😛

 

[Smeg45]

I've done it. Chucked out my 965 and switched to Intel, an i5 3470. The difference is huge. Literally. From overall smoothness to no more dips and stuttering to better minimum FPS. No more AMD for me unless they really become (Ha!) competitive. I shudder to thing of a new PC in 2012 with FX or x4. Big big big difference.

 

[viridiancrystal]

well, if there are no more delays, then the chips will probably start shipping soon, so be prepared for "leaks"

 

[davemaster84]

I've done it. Chucked out my 965 and switched to Intel, an i5 3470. The difference is huge. Literally. From overall smoothness to no more dips and stuttering to better minimum FPS. No more AMD for me unless they really become (Ha!) competitive. I shudder to thing of a new PC in 2012 with FX or x4. Big big big difference.

Congrats in your new rig, it's true that there's a very relevant difference between phenom II and Ivy, but not only in performance but in prices. I tought about going to Intel too but with the cost I'll rather get another 680 to pair my current , anyway cheers.

 

[dattimr]

I've done it. Chucked out my 965 and switched to Intel, an i5 3470. The difference is huge. Literally. From overall smoothness to no more dips and stuttering to better minimum FPS. No more AMD for me unless they really become (Ha!) competitive. I shudder to thing of a new PC in 2012 with FX or x4. Big big big difference.

Those remarks are quite interesting. When you talk about "overall smoothness," are you referring to general usage? Because I have to use an i3 530 3.0 GHz with HT at my job, and it compares very unfavorably to my Phenom II 970 3.5 GHz. The i3 is not an Ivy Bridge (not even a Sandy Bridge, actually), but it most certainly struggles whenever I start compiling a project on Visual Studio 2010 while doing anything else on the PC, for instance. That machine generally becomes unresponsive whenever I am running five or six programs at the same time, whereas my Phenom II setup can handle pretty much any workload that I have thrown at it so far -- 2 AAA games running at the same time, plus MS Word, Skype, 20+ Chrome and IE9 tabs, several PDF files open in Adobe Reader, Visual Studio 2010 and Eclipse, etc. Also, one of my best friends has just upgraded from a Core 2 Quad to an i5 2600K, and he has overclocked it well above 4.0 GHz. There is no way to tell the difference between his machine and mine in general usage, but perhaps he gets better FPS in games. I can't compare our gaming experiences because our GPU setups are very different, though.

 

[king smp]

Do realize that when people spend larger sums of money on an upgrade they will rationalize that there is a big improvement even when the difference is not actually that noticeable.
I went from a Core2 Duo 3ghz to a Phenom II x 4 925 @3.5ghz to a Phenom II x 4 965 BE 4ghz to a Phenom II x 6 1100T BE @4ghz with a Xeon equilavent to a I5-2400 as a second rig and really going from a dual core to quad core was very noticeable and the other upgrades were an improvement but really we are talking a few minutes faster encoding times and better benches
as far as running multiple browsers and office apps it isnt anything you would notice

 

[dattimr]

http://translate.google.com/translate?hl=de&sl=auto&tl=en&u=http%3A%2F%2Fwww.planet3dnow.de%2Fcgi-bin%2Fnewspub%2Fviewnews.cgi%3Fid%3D1346580414

Interesting investigation. Apparently, there is a good chance that people haven't been dealing with Vishera all along, but rather with another revision of Zambezi. Some of the blue-ish die shots that appeared recently, which were supposedly from Vishera, might be fake.

 

[de5_Roy]

I've done it. Chucked out my 965 and switched to Intel, an i5 3470. The difference is huge. Literally. From overall smoothness to no more dips and stuttering to better minimum FPS. No more AMD for me unless they really become (Ha!) competitive. I shudder to thing of a new PC in 2012 with FX or x4. Big big big difference.

fyki, you have managed to hurt the feelings of the members belonging to the following clubs:
■'i don't notice the difference',
■'i do big multitasking',
■'i run pov ray bench all day',
■'every time someone chucks an amd cpu, i throw away a kitten',
■'intel cpus are overpriced', and
■'smoothness (in gaming) is overrated like those intel quad cores'.
😛 😗

 

[Smeg45]

Those remarks are quite interesting. When you talk about "overall smoothness," are you referring to general usage? Because I have to use an i3 530 3.0 GHz with HT at my job, and it compares very unfavorably to my Phenom II 970 3.5 GHz. The i3 is not an Ivy Bridge (not even a Sandy Bridge, actually), but it most certainly struggles whenever I start compiling a project on Visual Studio 2010 while doing anything else on the PC, for instance. That machine generally becomes unresponsive whenever I am running five or six programs at the same time, whereas my Phenom II setup can handle pretty much any workload that I have thrown at it so far -- 2 AAA games running at the same time, plus MS Word, Skype, 20+ Chrome and IE9 tabs, several PDF files open in Adobe Reader, Visual Studio 2010 and Eclipse, etc. Also, one of my best friends has just upgraded from a Core 2 Quad to an i5 2600K, and he has overclocked it well above 4.0 GHz. There is no way to tell the difference between his machine and mine in general usage, but perhaps he gets better FPS in games. I can't compare our gaming experiences because our GPU setups are very different, though.

The original i3 530 is not a quad and the architecture is aging, so it won't cope with something rendering or compiling and a few other programs open at once, an x4 has four real cores so Windows can shuffle around more threads at will. Simple put, more cores, more stuff you can do.

To everyone who has replied, basically my minimum FPS shot up 10-20% at least in a lot of games (still testing) and I don't notice any more large dips and/or stutters. So if it hits 50, it stays at 45-50, and doesn't drop to 35 or so like my old 965. Specifically in Crysis 2, the 965 would frequently hit 35-45, the 3470 hits 45-60+, with the minimum a solid 42. I only saw it briefly dip to 38 or so during a level transition. For the Witcher 2, its dead stable at over 50-60 with my 7950, every setting maxed except for Ubersampling. It just feels *slick* and doesn't fumble. After gaming for 2.5yrs with the 965 I notice a definite difference.

 

[cgner]

fyki, you have managed to hurt the feelings of the members belonging to the following clubs:
■'i don't notice the difference',
■'i do big multitasking',
■'i run pov ray bench all day',
■'every time someone chucks an amd cpu, i throw away a kitten',
■'intel cpus are overpriced', and
■'smoothness (in gaming) is overrated like those intel quad cores'.
😛 😗

LOL! Daym streit ya

 

[sarinaide]

http://www.obr-hardware.com/2012/09/preview-amd-fx-8350-piledriver-last.html someone ban this crunt sack from the internet not only is he extremely biased and cannot hide it, he also talks crap to trafic attention.....but its like they say "feed the trolls". Why would Steamroller, Excavator be Server grade only, why bolt on a high end Radeon core for servers why make all the changes to improve single threaded performance on server grade chips.....this guy is a first class prick of the highest order.

I have seen a FX 8350 sample and there are things I am happy with something I am not pleased with but it was to be expected being the same process as Zambezi, in fact Zambezi with slight revision, unlike Vishera which will be a new process completely.

 

[truegenius]

^
😗
by mistake

now i corrected that 😛

 

[sarinaide]

^ Wrong thread maybe?

 

[sarinaide]

^ Burn him at the stake......Heretic 😛