AMD "Piledriver" refresh of Zambezi - info, speculations, test, fans

[FlanK3r]

if yes, it will be great! Finally superpi in xtremecooling under 10s

[Oliverda]

if yes, it will be great! Finally superpi in xtremecooling under 10s

[cesariuth]

New FX-8350 Vishera...



http://www.donanimhaber.com/islemci/...islemciler.htm

[Darakian]

What does 3C 2012 translate to? Third quarter?

[FlanK3r]

I think yes, so some between June and September

[Oese]

So this years Xmas most likely Damn i need at least leaks ASAP!

[haylui]

Dr. Su offered some positive early indications for Trinity. She said design wins are "tracking ahead of Llano," and that chips are already shipping to PC makers, with products due by the middle of the year.

source: http://techreport.com/articles.x/22452/2

Looks like the chip already to be shipped at least.

[GeneralDeath]

source: http://techreport.com/articles.x/22452/2

Looks like the chip already to be shipped at least.

Nice read line up looks great Question is the performance and time (release) as it looks after a quick read that 2013 is going to a good year for AMD. Still interested in seeing some results for PileDriver (increase frequency and IPC)

[cesariuth]

amdtrinityfusiondetay_2_dh_fx57.jpg

http://www.donanimhaber.com/islemci/...slemcileri.htm

The source talk about SI graphics core next????

[informal]

The GPU should be VLIW4 instead of VLIW5 that is in Llano. VLIW4 approx. has ~1.1-1.15x better performance than VLIW5 at similar stream proc. count and clock . So 384SP in Trinity should be equivalent to ~422-440SP in Llano(at same clock). Since top Llano's GPU runs at 600Mhz and Trinity's at 800MHz,it's obvious how much faster will Trinity based APU be in GPU tasks: 422/400x800/600=~1.40x or 40% faster.

[cesariuth]

The GPU should be VLIW4 instead of VLIW5 that is in Llano. VLIW4 approx. has ~1.1-1.15x better performance than VLIW5 at similar stream proc. count and clock . So 384SP in Trinity should be equivalent to ~422-440SP in Llano(at same clock). Since top Llano's GPU runs at 600Mhz and Trinity's at 800MHz,it's obvious how much faster will Trinity based APU be in GPU tasks: 422/400x800/600=~1.40x or 40% faster.

How would you explain the dual graphics technology with VLIW 4 (two different arquitecture for discrete graphics) for Trinity?

[informal]

How would you explain the dual graphics technology with VLIW 4 (two different arquitecture for discrete graphics) for Trinity?

Because not all 7000 series are GCN based... But honestly I don't know how AMD plans to pair VLIW4 in Trinity with re-branded 7570 or lower models that are still VLIW5 based. Maybe parts can work in hybrid crossfire even though they are different.

[Smartidiot89]

How would you explain the dual graphics technology with VLIW 4 (two different arquitecture for discrete graphics) for Trinity?

VLIW4 was confirmed for Trinity during AMD Fusion Developer Summit 2011. GCN won't make an appearance in Trinity, AMD even explicitly stated GCN for Kaveri during Analyst Day.

[cesariuth]

VLIW4 was confirmed for Trinity during AMD Fusion Developer Summit 2011. GCN won't make an appearance in Trinity, AMD even explicitly stated GCN for Kaveri during Analyst Day.

then there is no dual technology grafics for Trinity?

[informal]

Of course there is... But we don't know with which models will it be able to pair up with.

[AbortRetryFail?]

Fudzilla
- Monday, 21 November 2011

AMD claims a significant improvement in the "dual graphics" department as the new A8-, A6- and A4-series Trinity parts should benefit greatly when paired up with Turks Pro GPU, aka Radeon HD 6570.

[LowRun]

Although public details about Piledriver are scant, Cyclos has announced that AMD licensed its resonant clock mesh technology in order to reduce the chip's clock distribution power.

http://www.presence-pc.com/actualite...edriver-46763/

http://www.theinquirer.net/inquirer/...ant-clock-mesh

[tbone8ty]

interesting, hopefully this is curb high power consumption at load for bulldozers

nice to see their modular design is starting to benefit from such things. hopefully it works out along with all the the other piledriver tweaks

[EniGmA1987]

I hope even more so that Global Foundries gets its process in line. Tweaks are good and all but if the actual production process is terrible we still will have lower than hoped for clock speeds and higher than hoped for power consumption.

[Tomasis]

according to dobaher (rumors?) that BD 8140 (also other as 4, 6 cores) is coming.

new revision? I hope for improved OC

[EniGmA1987]

according to dobaher (rumors?) that BD 8140 (also other as 4, 6 cores) is coming.

new revision? I hope for improved OC

Well if it is called the 8140, Id guess it will OC about the same and have similar power draw, but launch at a slightly slower clock speed. If it was going to be better AMD would have called it the 8160.

[Tomasis]

you are right.. I want see new tech from piledriver (mesh) implement on BD early, maybe the hope is out anyway.

[tbone8ty]

These are 95w instead of 125w so maybe glofo is getting better

[tbone8ty]

http://en.wikipedia.org/wiki/List_of...icroprocessors

are these confirmed B3 steppings? for FX-4170 and FX-6200

hmm maybe not

http://www.cpu-world.com/CPUs/Bulldo...70FRW4KGU.html

any reviews yet?

[BeepBeep2]

http://en.wikipedia.org/wiki/List_of...icroprocessors

are these confirmed B3 steppings? for FX-4170 and FX-6200

hmm maybe not

http://www.cpu-world.com/CPUs/Bulldo...70FRW4KGU.html

any reviews yet?

I think AMD are just milking B2 with the pointless FX-4170 and FX-8140 releases. The extra 100 MHz won't be worth a single $ over the 8120. We'll never see a B3 stepping out of Bulldozer IMO between how hard it failed (if not in performance to some, performance per watt is dismal...), production capacity with APU/CPU balance and available wafer slots @ GloFo.

We can blame the manufacturing process all we want but it's been 6 months since retail production began for FX-8100 series, 9 months since AMD Llano retail production and apparently it has only improved 3%. To be honest, AMD and GloFo need about 10x that.

One reason I am reluctant to blame the manufacturing process is Llano's performance per watt. At a high vcore (1.4v+ stock vcore) it can still beat Phenom II performance per watt, memory bandwidth and performance per clock and include a GPU with significantly less package pins (not sure how they pulled that off!)... Bulldozer couldn't, at all.

I think it's right for AMD to skip a new stepping here and get the already reworked and finished Piledriver core on the market ASAP. I'm sure they already have early engineering samples of 8-core FX-8300 CPUs, and Trinity will probably ramp up retail production before the end of March.