AMD Piledriver rumours ... and expert conjecture

Page 177 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Status
Not open for further replies.
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 ...
 
http://www.extremetech.com/computing/130255-new-bulldozer-deep-dive-sheds-fresh-light-on-amds-troubled-cpu


Not a bad read, we already now this stuff but its restated!

It's a good summary of Anand's article.

And like you said, there's nothing new, but it's good to remind that to everyone: PD won't be a magic touch to the BD uarch, but we will see improvements at least.

Just let them be an upgrade for me and I'll be happy, hahaha.

Cheers!
 
Point [missed] completely.

Enthusiast clockers push for the numbers, highest overclock attainable, partularly with LN2 clocks... its the point of competitive clocking. i.e just because it doesn't suit your needs or its not in favour of your position doesn't mean its irrelevent...in fact to overclockers it is very relevant.

PS: HD Radeon 7970 pushing 1800+ on the core for the highest GPU overclock.

But it is irrelevant to an enthusiast rig or high performance computing workstation, *that* is the point. These extreme overclockers should do what they do with all cores.
 
who needs an AMD HS/F, it AMD only because my dumb ass mailed off the wrong brackets last month with a CPU + HS/F deal.
it was for an AMD chip but I included Intel brackets just in case.
too bad they are the wrong ones...
:??:
lightly used and no bent fins or anything...

http://www.newegg.com/Product/Product.aspx?Item=N82E16835233025

Not bad but then again I have hoarded a lot this past three years when it comes to coolers. :pt1cable:
Maybe that it could make a neat little mod project one day.
 
I'm sorry Mal, but i'm siding with sarinaide this time, I will have to say that i'm completely confused why everyone thinks the 600 series from nvidia is so great when they can't even get them out the door. This is the reason why Amd is taking market share away from Nvidia on the discrete video card market. To all those Nvidia fans i must say the future does not look that bright for them when Amd APU's are starting to get rid of 100$ video cards and below. Not just Amd either even the Intel 4000HD graphics is making it harder for Amd/Nvidia to compete in the lower-end spectrum, i'd say if you have a Intel ivy their is no reason to buy a 6450/5450 anymore since the onboard graphics are just as good if not better.
The 6XX has the nvidia label on it which is good enough for the mentally ill (his name translates to that from french)
 
Just for Fun let me try to restate some of the issues with Bulldozer in regrades to the performance, from most important to least important based on Cient/Work station performance not server.

Longer Pipeline= Higher clock speed which is something Amd failed to do
Branch prediction/Prefetcher
L2 Cache
L1 Cache(WTH did they share it?)
2x ALUs and 2x AGUs for Bulldozer vs 3x ALUs/AGUs for the phenom per core. Plus 2 IPC per core for BD vs 3 IPC for the Phenom BUT the Bulldozer can handle these operations more efficiently then the Phenom.
CMT only has 80% scaling per core when the Phenom had 93% scaling(so good!)
The Longer Pipeline which is not that big of a deal since the Branch Prediction was supposed to overcome this, but it does hint at what Amd was trying to do and it does cause some small latency issues.
How windows handles CMT(based on windows 8 vs 7 benchmarks this is only a 5-10% increase in performance and usually less then 5%)
You can easily see this in these articles below

This article seems like the L2/L1 cache is being part of the problem.
http://www.extremetech.com/computing/100583-analyzing-bulldozers-scaling-single-thread-performance/2

Then if we take a look at this article we can see CMT only can scale 80% so its multithreaded performance is also lower then a TRUE 8 core Phenom would be.

http://www.legitreviews.com/article/1741/11/

Then the Bulldozer is only clocked at 3.6Ghz which is lower then the Phenom 975/980 and is only 8% higher then the 1100T.


So what did Amd improve in Piledriver?
Clock mesh should improve Clock speeds/Power consumption, Global foundries most likely improved their 32nm die as well meaning Piledriver might be made on a newer stepping and this might make Piledriver a little smaller then Bulldozer was and global foundries should have more Piledriver processors at launch vs Bulldozer at launch, which i hope means Amd will have lower prices.
L1 cache has also been tweaked
The Prefetching and branch prediction have been improved as well
Scheduler has also been improved

This most likley leading to a modest 15-20% boost in performance compared to the bulldozer at stock with IPC being around 7-10% better

What Amd left out and we hope they improve with "steamroller"
die shrink even 28nm would be nice
L2 Cache/L1 cache
2x ALUs and 2x AGUs for Bulldozer vs 3x ALUs/AGUs for the phenom per core. Plus 2 IPC per core for BD vs 3 IPC for phenom
Branch prediction/Prefetcher
CMT only has 80% scaling so i would like to see a 10 core steamroller

So this concludes what i think is personally wrong with Bulldozer and some of the improvements they made with Piledriver and dreams i have for Steamroller.

Now What did Amd improve on when it comes to the Bulldozer vs the Phenom?
Number one is support for newer instruction sets(this is easily a performance boost in some areas), Memory controller is on par with the original I7 series.FPU is better since their is only 4 of them in BD vs 6 in the Phenoms and Bulldozer still beats the Phenom FPU is some cases, L3 cache speeds/size. Turbo core,(scratching my head right now), I guess higher clock speeds with the 4170fx.
Since the Phenom had higher scaling and bulldozer had lower performance per cycle the bulldozer is usually only 10% faster in multithreading while being 10-15% slower in single core tasks with having a 9% higher clock speed with turbo on.

 
honestly who cares about the FX-8150 and it's world record overclock only on two cores..

show me an 4m/8c clocked stable around 6.0GHz and I will be impressed.
but some chip knocked down to two cores, I don't care..


You do realize the irony in what you just stated right?

You say you don't care about something at two cores, yet the things you claim to use an Intel CPU in only run on one to two cores. (Gaming!!! orz 101010101).

So whats more important to you, single core performance or total CPU performance?
 
gigabyte's new trinity motherboard
http://semiaccurate.com/2012/06/03/trinity-board-shown/
two pcie slots makes me wonder if hybrid tri-fire (or amd triple gfx) is possible...


WTH??? makes me scratch my head and ask what they were trying to accomplish. APU sockets are about low power computing, multi-GPU power gaming setups make about as much sense as a 4.2L V8 in a Civic.
 
AMD struggles with multi-GPU as it is...............

We've already been over this. You've gotta be screwing with the settings pretty badly to have it remotely matter.

Phenom II x4 970BE @4.0~4.2Ghz (depending how I'm feeling that day).
2 x 580 Hydro's

There is no struggling involved here. And if my 970 can do it, then I don't see why a 8150/8120 can't, and no reason why a PD variant couldn't. An APU shouldn't be doing high end CFX / SLI anyway, kinda defeats the purpose.
 
what's funny is that I had the 975BE @ 4.2GHz pushing SLi N560GTX-Ti Hawks....
well needless to say it's now 2500K @ 4.2GHz pushing SLi N560GTX-Ti Hawks....
BIG DIFFERENCE.
😛 😛


It shouldn't be, unless your doing something on purpose, which knowing you and UK I would think so. Display capabilities are almost entirely GPU bound with the CPU's only job being to order the vertices and send them to the GPUs. Texture data is sent to both GPUs simultaneously via DMA, CPU isn't even involved. Things like z-ordering and lighting are all done in the GPU now, stencils and what little the CPU used to do has been taken over by shaders.

Only way to hit a CPU limit in multi-GPU setups would be to attempt a crazy high resolution without using AA and as few shaders / lighting effects as possible. Basically you want to lighten the load on the GPU as much as possible while still keeping it high enough to tax the CPU. In actual real-use scenarios, 1920x1080 AA / bells & whistles, there won't be a difference between the two.

I'm now waiting for UK to post the same posts he did last time he made this claim. Of course he won't post my reply where I blew barn sized holes into his argument.

And lolerskates at the idea of Mal / Recon being remotely unbiased or neutral.

That would require being even handed and recognizing both the good and bad of each product. Not jumping on the hate-AMD bandwagon that's been all the rage lately.
 
And yes... i run without SLi, as my new motherboard is..... just... a pain..
Games run much smoother on the SB based rig...

This statement is an oxymoron. You can not state that AMD can't do SLI, then claim your not using SLI.

Also can not use "smoother" as that's subjective and based on perception. If you wanted to see the AMD system skip / lag then you'll see it, whether it exists or not. If you wanted to see a single card i5 run "smoother" then a dual card PII, then you'll see it regardless of the actual truth. Just like people claim to recognize single images at 8.3ms.

I'm living breathing proof that a PII @4.0Ghz or higher (got mine to 4.5 but I don't like running it that hot) can run two 580's in SLI just fine. SC2 is about the only game that ends up being CPU bound and only because it's got retarded core usage in single player (someone should do a multiplayer benchmark to see if it's like BF3).
 
....then Haswell is supposed to be Intel's sojourn into APU GPU performance realms.....HD 5000 will probably still look second grade to a then year old Trinity....
I wouldn't judge what Haswell's graphics could do yet. However, I will say that ~30% from 3K to 4K was probably not a great sign for Intel. 300,000 extra transistors from SB to IB, most of it was for IGP, and they get 30%. I doubt they will get away with making the chip size too big, so I don't see them improving much more than 25-30% on Haswell either, if they spend more than a few minutes on the cpu improvements.

Advantage: AMD
GCN is already out, in full force. Drivers will be ready to take advantage of the GCN stream processors more than Intel's for their equivalent (IE's, I think. Probably not.) GCN is a fantastic arch, and it almost seems like it was designed for use in APU's.

Overall, based on what I've seen from Intel out of Ivy, I think AMD will be ahead in graphics performance for quite a while going forward.
 
Recording puts strain on my hard disks and chops fps by around 10fps or so.

Single 480.


[flash=560,315]http://www.youtube.com/v/WCrWg5nc3xU?version=3&hl=en_GB[/flash]
odd. I have a 2.5" HDD i record to and I can record 1080p@60fps no problems. I'm fairly sure you have more graphics power than I do too. Single 7850.
 
odd. I have a 2.5" HDD i record to and I can record 1080p@60fps no problems. I'm fairly sure you have more graphics power than I do too. Single 7850.

Don't trust anything from a youtube post. That opens up a whole new world of tweaking for one side or the other. I could show a I7 stuttering like mad, but that would involve be playing with certain settings to force it.

I just look at my own screen, no stuttering and stable 60+ FPS. End of story.

Although I can easily imagine a few things one could do to force stuttering. Having had to clear up that issue before myself, driver settings are incredibly important.
 
WTH??? makes me scratch my head and ask what they were trying to accomplish. APU sockets are about low power computing, multi-GPU power gaming setups make about as much sense as a 4.2L V8 in a Civic.


A V8 FF Civic?

Darn... Do want! lol

Anyway... Just for kicks... I never got frame drops while recording in the phenom, but I did in the Athlon X2 =o

Cheers!
 
Point [missed] completely????

Since when did 0.00000001% of the market ever matter a damn? :ange:

Saying this you must also say that Amd's APU's are great for 80% of the people who do give a damn.
I was addressing the laughable claim that the clockspeeds a CPU attains when overclocked under L2, is a meaningful positive in how well that CPU is going to sell to the general public.

That's it, that's all I was saying, there is no need to bring in matters completely unrelated.
 
I was addressing the laughable claim that the clockspeeds a CPU attains when overclocked under L2, is a meaningful positive in how well that CPU is going to sell to the general public.

That's it, that's all I was saying, there is no need to bring in matters completely unrelated.

But... Just like a V8 Civic is 'because race car', that record is 'because high OC' and gives you braggin rights, lol

And danm you Paladin... Now I can't stop imagining that in a drag race, rofl!

Cheers!
 
Wonder where palladin ran off too...
Seems like some folk are narked...

Truth must hurt when a single 480 is getting too close to 2 watered down 580's....

Why can i get 90+fps in SLi on my 2500k on BF3?
And yet palladin's 580's are getting 60+...

Huh?

Its 12:47PM here, aka lunch time. Actually I take lunch at 1PM cause we stagger our lunch schedule. I'll let you figure the rest out.

Getting high frame rates is easy, did I mention that I do 3D vision and it was the driving force behind the decision for two 580s.
 
High frame rates is easy?
On a single 480? vs 2x 580's?
I want some of that food too, i will think alot more like you!


I'm really trying to understand what your saying here? Are you comparing a 480 to two 580s? Or are you trying to say an I5 + 480 will beat a PII + two 580s?

In either case, I think everyone here knows the absurdity of those statements.
 
Status
Not open for further replies.