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 ...

 

[noob2222]

Why wouldn't anyone believe AMD? It's not like they've ever been economical with truth before.

Could be entertaining to find out considering AMD fired the marketing team for BD BS. Wonder if they are a bit more cautious now that they know their jobs are on the line.

 

[jdwii]

Anyone here believe AMD's 25% cpu increase claim for trinity?

http://wccftech.com/amd-trinity-a8-es-pitted-llano-a83850-apu-3dmark-11-benchmark/

 

[bawchicawawa]

Anyone here believe AMD's 25% cpu increase claim for trinity?

Athlon to second generation bulldozer, 25% increase isn't a hard thing to believe considering the context...

 

[-Fran-]

http://wccftech.com/amd-trinity-a8-es-pitted-llano-a83850-apu-3dmark-11-benchmark/

Weird specs for the ES sample TBH...

IF those screenshot were true, then CPU intensive games are gonna run equal or just a tad better than current Llano A8 and GPU intensive ones might run faster (by quiet a lot, looks like), that's a good sign (again, if those are true).

Wonder how much more perf CPU wise could they squeeze from the final (or first batch) rev.

Cheers!

 

[esrever]

http://wccftech.com/amd-trinity-a8-es-pitted-llano-a83850-apu-3dmark-11-benchmark/

I wouldn't trust any of the bullshit this guy writes. In his bulldozer leaked benchmarks he has bulldozer beating the i7 in superpi...

 

[jimmysmitty]

http://wccftech.com/amd-trinity-a8-es-pitted-llano-a83850-apu-3dmark-11-benchmark/

Am I the only one wondering why it states the Trinity system is running a HD7950 in the 3DMark 11 specs? Hell I like this part:

The Trinity Accelerated Processing Unit which was used is part of the Quad Core A8 Series which feature the 2nd Generation Piledriver Core from AMD. The Trinity A8 sample also makes use of a DX11 – HD7000 Series IGP which results in more Graphics Processing power over its predecessor. The Trinity A8 Sample runs at a stock frequency of 3.2Ghz and was tested with the HD7950 GPU, Results of 3DMark 11 are posted below:

That alone states BS since the Llano was tested with the IGP but Trinity was tested with the HD7950 (still not even out yet). When its IGP vs IGP, we will talk.

 

[esrever]

he says thats piledriver which might or might not be an apu chip/

 

[jimmysmitty]

he says thats piledriver which might or might not be an apu chip/

You miss the part where he says Trinity. Over and over again. Why would he even compare it to Llano if its Pildriver but not Trinity? Trinity will be a APU first. Also why even compare it to Llano if there is not a proper on board GPU? Its like comparing a 2600K with the HD3K to a Phenom II with a HD6970 in gaming. Its not a legit or fair comparison.

We all know, and hope, that Llanos IGP would suck compared to a sull blown HD7950.

 

[esrever]

hmmm well I didn't really read it, just looked at the benches and his face was enough to turn me away.

 

[jdwii]

All i know is this is feeling like BD all over again with PD. I just don't see how its going to be 25% faster on the CPU side with trinity. Not with laptop clock rates its just something i'll have to see before i believe it. I do feel however the Graphics will be at least 30% better then Llano but the CPU i'm guessing the same or even in some cases a little slower, Unless they can clock the crap out of piledriver like 2.4Ghz stock with a nice turbo.

 

[de5_Roy]

i think cpu performance decrease is not unexpected. llano has mature cpu cores, die shrunk. those are true quad-cores. otoh, the trinity chip in the article looks like it has 2 pd modules, similar to a stripped-down-but-slightly-improved fx4100.
i also wonder why the trinity chip was benched with a 7950 since trinities can't (well... shouldn't be able to) hybrid cfx with high end 7950 ....

 

[noob2222]

hmmm well I didn't really read it, just looked at the benches and his face was enough to turn me away.

+1 rofl. every time I see that pic, I wonder if hes getting ready to offer some drugs for sale, looks stoned in the pic.

 

[truegenius]

lol
http://news.softpedia.com/news/Blind-Test-Shows-AMD-Machines-Run-Better-than-Intel-s-248202.shtml
hahaha...

 

[de5_Roy]

lol nice one.
more people should be aware of llano's performance.

 

[truegenius]

fx8 too which is supposed to be bad 8 thread cpu in price to performance ratio

 

[viridiancrystal]

lol
http://news.softpedia.com/news/Blind-Test-Shows-AMD-Machines-Run-Better-than-Intel-s-248202.shtml
hahaha...

That is quite surprising, and a good sign for AMD. APU's need to become more well known.

I still would like to see the Fx 8120 vs 2500k in multitasking. Game running on one monitor, skype, facebook, twitter and a livestream on another. something to that extent at least.

 

[-Fran-]

That is quite surprising, and a good sign for AMD. APU's need to become more well known.

I still would like to see the Fx 8120 vs 2500k in multitasking. Game running on one monitor, skype, facebook, twitter and a livestream on another. something to that extent at least.

Well, a more realistic test would be Steam + Teamspeak/Vent/RaidCall/Skype running along with the game plus AV software. Hell, now that I think about it, those should be added nowadays as well as standard testing software procedure! Even including an AV should be a good call 😛

On a second thought, no it won't, cause besides the games and OS, there wouldn't be any reference to performance, but if Intel causes that "regular multi-tasking while gaming" becomes noticeable, the way benchies are done today invalidates all the data by itself. We would need a true Hardware site to do the findings!

You know, on a side note, related to the softpedia link (thanks truegenius) and something Baron (not calling a flame war here, lol) noticed/say a while back. While programming at the office (i5 2400, 4Gb) and listening to music in youtube, when it's compile time or refresh-Eclipse time or other i-will-use-100%-CPU time, everything comes to a halt (windows freeze) or hard skips (youtube, streamming, etc) occur. My workstation has WinXP 32bits and a Quadro card in it (don't remember which one) and my home computer is in my sig, but in my house I also run steam and some sort of voice program (TS, Vent, Skype, etc) along with the programming stuff. Thing is, I've never noticed halts or skips on anything running in the BG when using the CPU to a 100%. From the Athlon X2 times (WinXP 32bits) and now (Seven Pro 64bits). The only difference is that at the office, I've got McAffe and Intel's Monitoring utility (don't recall the name of it) software running as well and here I've got Asus's ROG software, FRAPS, DAEMON Tools Pro (came with the Crosshair V, tee hee), Dropbox and AVG. Weird to say the least, now that I realize it.

Anyway, those blind tests sound like very good PR. Such a simple thing to do and you prove a thousand benchies wrong to those people, lol.

Cheers!

EDIT: Removed the Tarzan inside.

 

[esrever]

I have a hard time believing the people could tell the difference between a FX8150 and a 2700k unless hyperthreading was screwing something up. I can't even tell the difference between a i5 2500k and and phenom II x4 while gaming without fraps on.

 

[palladin9479]

I have a hard time believing the people could tell the difference between a FX8150 and a 2700k unless hyperthreading was screwing something up. I can't even tell the difference between a i5 2500k and and phenom II x4 while gaming without fraps on.

That was the whole point of their blind test, to both see for themselves and demonstrate to others that the actual "differences" between PC components today are so small, that you'd never notice them without running a benchmark.

 

[Chad Boga]

That was the whole point of their blind test, to both see for themselves and demonstrate to others that the actual "differences" between PC components today are so small, that you'd never notice them without running a benchmark.

Then the results should have been closer to 50/50.

A very strange result they managed to get.

 

[esrever]

I would have thought people would choose "no difference" 90% if the time. Its very weird that its quite a substantial majority for something like this.

 

[palladin9479]

Then the results should have been closer to 50/50.

A very strange result they managed to get.

Simple, test size. Use 10,000 people at multiple locations and the results would average out better. At least for the desktop machine.

If there were doing a definitive scientific study, then it would of been double blind with over a 1,000 subjects at a minimum. But they were not trying to derive data for a thesis and instead just making a point. Only us enthusiasts even care about the difference.

 

[de5_Roy]

couldn't they just right click on the computer icon and find out the system details? that alone should reveal most of the information one needs. sometimes game options show system information. i don't think keyboard usage/access was restricted.
that's the first thing an informed user would do on an unfamiliar pc.... i think.

 

[viridiancrystal]

couldn't they just right click on the computer icon and find out the system details? that alone should reveal most of the information one needs. sometimes game options show system information. i don't think keyboard usage/access was restricted.
that's the first thing an informed user would do on an unfamiliar pc.... i think.

This reins true. I always have this explainable urge to know what hardware i'm running on a new and unfamiliar computer.

 

[truegenius]

i think toms can try this test (blind test) to bring clear results to us (other than benchmarking)