AMD Piledriver rumours ... and expert conjecture

Page 41 - 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://vr-zone.com/articles/arctic-announces-first-products-with-next-gen-amd-trinity-a8-and-a10-apus/14712.html


New HTPC is the first one that is solely tailored for the role of a living room device. MC101 carries the tag line "Entertainment Always" and the company is following suit with a very sleek design. The interesting bit is that this compact HTPC is powered by the upcoming AMD Fusion "Trinity" processors carrying a familiar name "A8", and not so familiar "A10". Both processor lineups will be powered by the Trinity silicon, i.e. next-generation CPU core codenamed "Piledriver" with the GPU core codenamed "Northern Islands".




According to the associated slide, AMD "Trinity A8" comes with Radeon HD 7640G, while "Trinity A10" comes with Radeon HD 7660G, implying this will be a higher performance part (current Fusion APUs top out with X500 series). According to Arctic, the products will be released when AMD decides to launch Trinity-based Fusion APUs, which could be CeBIT 2012 in March, or Computex Taipei 2012. Given the timing when Llano was introduced to market, we would assume the second option is a more realistic one.



Do note that Arctic is deliberately not mentioning AMD's "Fusion" brand in any part of the press release. This comes from an interesting legal case taking place: few days ago, a report came out that Arctic actually plans to sue AMD over the Fusion brand, claiming it has filed for a trademark before AMD. The problem with that could be that AMD was saying Fusion for their future products before that power supply trademark was filed - both came out in 2006. However, that is a legal matter and knowing Arctic's previous love for threatening individuals or companies with litigation, it might be that MC101 line of products is a way of making a peace offer.

Then again, the calling of not growing, but exploding market of advanced set-top boxes. For example, Intel is currently shipping over 30,000 set-top-boxes per day... in EMEA region alone alone! Yes, that's 900,000 devices per month, all powered by a variant of a 45nm Core processor, named Pentium for the sake of argument.
 
Sadly, I'm limited to what I can put into this Slimline (one big mistake). I have reinstalled Windows several times and I do maintenance on my computer daily. I don't have any games or any other programs. I use this for school work. The only programs I have are Microsoft Office, FireFox, Skype, Smart Defrag, CCleaner, and Java. I have Anti-Virus system installed (Avast) and I don't go onto any suspicious websites.
I already know that my PC is slow, just because of dated hardware. Planning to build a new one when IB and Keplar are out.


my niece's athlon x2 4200 with a very slight overclock breezes along on the web just fine. I think your Windows installation has an issue.
 
we all know it's better to have a dedicated GPU when possible.

I would just throw out a disclaimer, that it really depends on what you use the PC for. Several of my coworkers, have had me build them PC's, so they have the needed power to work from home, using our VPN. One of the questions I always ask, is are they going to play any games. Most are using them for business needs, and maybe a few videos. I've been able to save them a lot of $$ by using the 2500k and IGP.
 
Well...[taken from anandtech]:

DSC_5228_575px.JPG


and

Screen%20Shot%202012-02-02%20at%209.21.08%20AM_575px.png


So...not worried about desktop, and moving away from lots of cores back to power-performance. Indications AMD realizes BD was a mistake?
 
Well...[taken from anandtech]:

http://images.anandtech.com/doci/5498/DSC_5228_575px.JPG

and

http://images.anandtech.com/doci/5497/Screen Shot 2012-02-02 at 9.21.08 AM_575px.png

So...not worried about desktop, and moving away from lots of cores back to power-performance. Indications AMD realizes BD was a mistake?
Looks rather promising to me, however, can they pull it off?

Also i think people underestimate the number of desktops that are still sold. The enthusiast market is small sure, but the average market is still quite large.
 
Well...[taken from anandtech]:

http://images.anandtech.com/doci/5498/DSC_5228_575px.JPG

and

http://images.anandtech.com/doci/5497/Screen Shot 2012-02-02 at 9.21.08 AM_575px.png

So...not worried about desktop, and moving away from lots of cores back to power-performance. Indications AMD realizes BD was a mistake?

My take on those slides is that there is nothing new in them. They say AMD is going to continue to go down the road of "good enough," power-efficient CPUs and "heterogeneous computing." That is exactly what their APUs are- okay but not stellar CPU performance, inexpensive to make, fairly low-power (especially Zacate), and use of OpenCL and GPGPU with the on-die GPU component. I take the part about not having "an emphasis on cores" meaning that they are going to do away with the emphasis on per-core IPC and the resultant "fat" cores as in Stars/K10 and move towards simpler cores such as in Zacate/Bobcat and Bulldozer. Either that or AMD is going to emphasize the power of the GPGPU portion rather than the CPU cores in their APUs. I have a hunch it's more the latter as "disruptive technology" from AMD pretty much means "GPGPU" from how I see it.

Bulldozer seems to fit in very well with what the slides appear to say. Bulldozer is a very modular architecture and I would absolutely imagine it is designed to be even more integrated with the GPU than the Stars-based Llano ever was. The best example I can think of is that the FPU is no longer very tightly associated with the cores. I would predict that before too long, AMD will use the IGP as the FPU and end up with absolutely massive FPU power and overcome some of the current restrictions and downsides of using a discrete GPU to do calculations. That would certainly be a disruptive technology as current GPGPU implementations are very fast but also very limited in what they can do well.

I don't think AMD is giving up the desktop computer market. The desktop computer market continues to shrink, and much of what remains is corporate desktops. Most of those are small-form-factor units with slower, low-powered CPUs and IGPs. That's exactly the kind of machine an APU will do well in, and AMD will certainly continue to sell chips to that market. AMD also realizes that their high-end APUs such as the A8 and their repackaged server CPUs (FX series) are also "good enough" for enthusiast desktop usage. They get playable framerates in games when paired with an appropriate GPU and are at least in the same ballpark as Intel's CPUs with doing most real high-end desktop work like video encoding and the like. Sure, they won't win many benchmarks against Intel's top-end chips, but they will get the job done in most cases. AMD has really only ceded two markets to Intel. The first is the >4-socket server market, which is a small fraction of a percent of the server market and continues to decline. AMD gave that up when they decided to arrange the HT link topology in G34 Opterons to a maximum of quad-socket operation as compared to the previous Opteron 800/8000-series' 8-socket operation. The advantage of that choice is that it made for a better-connected and better-scaling 4P setup. The other is the competitive benchmark market. To paraphrase one of AMD's marketing directors, "yes, we could be in those markets, but we would definitely spend billions in R&D to earn maybe millions in sales. You do the math."
 
the thing with AMD taking is approach is semi enthusiast are just going to always recommend Intel and the intel brand name would shut AMD out of the market like its been doing until the APUs launched. Even now intel laptops sell much more than the llano even tho llano will generally offer better general performance for a less price.
 
but see I know what I need to play what game.
I'll take the better CPU and then upgrade the GPU at first chance.
sounds like all the others need to get on Tom's and get the correct word before they make such a purchase.
we all know it's better to have a dedicated GPU when possible.



Lol i think you missed my point man i said if you can't change and just upgrade the GPU. A lot of HP/Dell buyers can't just do that. Without dropping a lot of money. Not to mention The bad cases they give consumers for air flow. Its the same thing with laptops(which i'm buying a Llano or maybe a trinity soon) If some one told me i can get a I7 laptop or a A8 i'm going with the A8 unless their is a video card inside that I7. I like to play some games and i like to watch 1080P video with No dropped FPS.
 
From http://www.anandtech.com/show/5492/amds-rory-read-outlines-amds-future-strategy

From a product standpoint, AMD is really focusing on its mainstream and entry level APUs. Rory didn't come out and say it here but no where in AMD's future direction is a focus on the high-end x86 CPU space.

Also note that AMD isn't going to be as focused on delivering high performance products on the absolute latest process node. It views Brazos as one of its biggest successes to date and that architecture was built on a 40nm process with an easily synthesizable architecture. It's likely that the future of AMD is built around more of these easy to manufacture SoCs rather than highly custom, bleeding edge CPUs.

So it looks like AMD's margins will continue to be on the low side, and that AMD will become just a commodity designer instead of high-end.

As for not being on the 'latest process node', that indicates Intel will continue to have a substantial advantage in performance per watt at the high end, and able to bump up performance at the low end as well. It is ARM that Intel is aiming for, but IMO AMD is the one most likely to get injured since the low end mobile market looks like it will be heating up in the next few years..
 
The extremist gamer market is so small it is really of little consequence.
The average person cares about cost not cpu power.
Amd 's sale of 30 million lano chips will motivate them to concentrate on apu as they offer more than Intel chips similarly priced.
Piledriver will sell very very well as trinity but alone it's sales will be negligible
Through the lenses of extremists Intel offers more but unfortunately during a double dip
recession the average person wants more than a fast cpu for the money .
Piledrivers will sell very well just not in extremist guise

BD was so badly made frankly I can't see Amd not squeezing at least 15-20% more power per watt.
 
If amd is to be believe about piledriver in trinity, they will be getting amazing performance per watt compared to bulldozer. 35w chip with 400 gpu cores + 2 modules compared to the 125w chip with only 4 modules. I mean trinity is suppose to get a 100% increase in performance per watt compared to llano from what amd says, maybe just the gpu part but the cpu part would still need to draw less power too. The 17w chips are suppose to be comparable to the 35w llano chips.

Effectively I'd guess piledriver to drive down the power usage a lot, probably not to intel level but at least better than phenom II. Bulldozer's high power usage are mostly due to the crappy 32nm process anyways.
 
the thing with AMD taking is approach is semi enthusiast are just going to always recommend Intel and the intel brand name would shut AMD out of the market like its been doing until the APUs launched. Even now intel laptops sell much more than the llano even tho llano will generally offer better general performance for a less price.

Recommend Intel... To the high end enthusiasts? Sure. Value/mainstream people? Time will tell.

Intel can pretty much destroy AMD if they want; we all know that, but AMD doesn't have to make things that easy for them 😛

Point is: BIG VALUE is what AMD is looking after here. Intel might destroy AMD in a lot of CPU-wise things, but AMD can leverage that in the "enough to play a flash without choppyness" category, hahaha. That's a very big market indeed.

What I'm afraid of though, is that with the process advantage Intel has, they will pretty much destroy AMD in the enterprise market (not servers exactly). Companies tend to have hundreds of PCs on almost 24/7, so every W saved matters (remember UPS and diesel generators). And those are a lot of pennies for Intel.

Would be better if AMD just sank their entire budget to market their APUs cause consumers are blind and the only thing guiding them is advertising.

I agree, but they should never ever sacrifice R&D for marketing budget. That's the road to being a copyright troll, lol. I mean, with the constraint in budget AMD always face, I don't think they can get better or "bigger" marketing. It's a tough game Intel makes them play, lol.

Anyway, I kinda like AMDs approach. Mr Read is doing/making interesting gambles here. If Trinity delivers, then that road map is going to be a PITA for Intel.

Cheers!
 
it seems that Intel and AMD whether on purpose or not
are splitting up the market
Intel for desktops and performance
AMD for mobile and basic systems
I do think that Bulldozer/Piledriver especially with Windows improvements will still be viable in desktop market
AMD has been good about giving "bang for buck"
just scary how they are EOLing so many viable CPUs (denebs,propus etc)
 
There's a reason, Nvidia Kepler GTX680 lenzfire.com/2012/02/entire-nvidia-kepler-series-specifications-price-release

that's enough for many to abandon it the APUs momentum fan club.
#1 that has nothing to do with this thread
#2 that article is complete bullshit, nvidia can't physically launch the real 680 any time soon, they are going to launch the 660 and brand it as a 680 to get something on the market
#3 All the info on the article are fake leaks thats been floating around the internet for months
 
Recommend Intel... To the high end enthusiasts? Sure. Value/mainstream people? Time will tell.

Intel can pretty much destroy AMD if they want; we all know that, but AMD doesn't have to make things that easy for them 😛

Point is: BIG VALUE is what AMD is looking after here. Intel might destroy AMD in a lot of CPU-wise things, but AMD can leverage that in the "enough to play a flash without choppyness" category, hahaha. That's a very big market indeed.

What I'm afraid of though, is that with the process advantage Intel has, they will pretty much destroy AMD in the enterprise market (not servers exactly). Companies tend to have hundreds of PCs on almost 24/7, so every W saved matters (remember UPS and diesel generators). And those are a lot of pennies for Intel.


Cheers!
Generally the trickle down effect of information will make intel sell even to people who would be better off with AMD products. A person might be told intel is best for gaming, he goes to tell everyone he knows to get intel laptops with integrated graphics. People on this forum often recommend intel to just about any cpu thread because it is the best, people are going to see the name intel and buy it.

People will recommend the i3 over the trinity APU simply because it has better gaming performance with the 7970 put in without looking at the need or use of the system. Many people buy i7s just because they are the "best" when i7s are complete waste of money except for people who do rendering/video transcoding ect. With the money saved from the i7 people can buy an ssd for much better general performance. Theres the marketing at work because realistically people will do what they thing is good whether they know what they are doing or not.

Intel is greedy which is what is keeping AMD alive in the desktop market at all. If intel launched their products better, they could drive AMD out of the market because they just have the technology. If intel just launched a 3 core unlocked cpu at about $150 and a 2 core unlocked cpu at $100 AMD would not be able to survive in the desktop market. It would just mean intel doesn't make as much money since they will sell a lot more low end CPUs and intel doesn't want that.
 
I wonder who they will be firing next when the performance doesn't get any better during the next release whether it be stepping revision or Piledriver.?
like a revolving door there now with employment ads on monster and such..

Hmm, I wonder who Intel will be firing if Piledriver does well? You, I suppose.
 
so, it's the consumers choice.
and didn't you just say something to this effect earlier.?
OK, we know how you feel now.
let's move on.
Choices based on misinformation. In the great world of "democracy" and "capitalism".

Didn't you also already of that effect? we also know how you feel now. Don't worry, we don't need to move on so fast.
 
Generally the trickle down effect of information will make intel sell even to people who would be better off with AMD products. A person might be told intel is best for gaming, he goes to tell everyone he knows to get intel laptops with integrated graphics. People on this forum often recommend intel to just about any cpu thread because it is the best, people are going to see the name intel and buy it.

People will recommend the i3 over the trinity APU simply because it has better gaming performance with the 7970 put in without looking at the need or use of the system. Many people buy i7s just because they are the "best" when i7s are complete waste of money except for people who do rendering/video transcoding ect. With the money saved from the i7 people can buy an ssd for much better general performance. Theres the marketing at work because realistically people will do what they thing is good whether they know what they are doing or not.

Intel is greedy which is what is keeping AMD alive in the desktop market at all. If intel launched their products better, they could drive AMD out of the market because they just have the technology. If intel just launched a 3 core unlocked cpu at about $150 and a 2 core unlocked cpu at $100 AMD would not be able to survive in the desktop market. It would just mean intel doesn't make as much money since they will sell a lot more low end CPUs and intel doesn't want that.

I can understand the worry behind your words about the "zealot troop", but both sides have people that know better what suit some people's needs (we all here, basically). Besides, is there's a notebook (for example) in display, playing a choppy movie and an A8 playing a movie flawlessly... What do you think it will happen? But if that person wants to see how it takes to render something? If that person sees a 10-20s difference between the A8 and the i7 over a 10 min process, I'm willing to bet the i7 will cost 2x the A8's price. What would you do in that scenario? And we could go on and on about it with situations. That's what it's up to us to identify and tell people. Marketing can go so far IMO, since hard cold truth always arises.

And the bold statement... Well, I deeply believe that AMD would do the same if it were in Intel's shoes, so no need to use the "greedy" card on Intel.

believe me if I worked for Intel, I wouldn't be getting fired.
and if I worked for AMD the minute they decided to go away from the Phenom II, not go Phenom III
and go FX-Bulldozer, I would have quit.... :kaola:
remind you I have an AMD rig as well..

Well, what if you were an engy for both parties? XD!

Hint: synopsys.

Cheers!
 
I can understand the worry behind your words about the "zealot troop", but both sides have people that know better what suit some people's needs (we all here, basically). Besides, is there's a notebook (for example) in display, playing a choppy movie and an A8 playing a movie flawlessly... What do you think it will happen? But if that person wants to see how it takes to render something? If that person sees a 10-20s difference between the A8 and the i7 over a 10 min process, I'm willing to bet the i7 will cost 2x the A8's price. What would you do in that scenario? And we could go on and on about it with situations. That's what it's up to us to identify and tell people. Marketing can go so far IMO, since hard cold truth always arises.

And the bold statement... Well, I deeply believe that AMD would do the same if it were in Intel's shoes, so no need to use the "greedy" card on Intel.

Cheers!
I have nothing against intel or any corporation with a basis on making money being greedy. Only thing I don't like is from the consumer stand point its hard to get a cheap cpu for exactly what you need it for, I'd have totally bought a sandy bridge unlocked 3 core cpu if they sold one.

Given that, I do believe that AMD can do well with the APUs but I think what lingers of their desktop and server market will most likely end up nowhere. In server space, intel is so dominate theres hardly any way AMD can compete even if they do perform competitively simply due to servers after reliability and low power consumption, both of which intel has done well. In the desktop market, it seems AMD can't even afford R&D for as they are just putting server chips and marketing them as desktop machines and the performance just doesn't scale well due to software limitations.

Focusing on the mobile is probably AMD's best choice and they are taking some steps towards it (launching piledriver on trinity first) but I think they can do more to market it and do more to support things like opencl. Would be interesting how things like gpgpu computing ends up, it could make the APU more powerful than the i7 in some applications.
 
What we need in this thread is some sort of leak about the Piledriver architecture.

I have already committed to an I7 build as soon as I can get myself to leaving behind my beloved X6. I'm going from 6 cores to 4. I will never again look at benchmarks the same as I have, they lead a person to make choices. I need to frame my X6 and hang it on my wall.
 
Status
Not open for further replies.