[xbit] AMD Talks Fusion: Vision, Solutions, Software.

[Gloomy]

http://www.xbitlabs.com/articles/cpu/display/amdfusion-2011-neal.html

Interesting read. Been out for two weeks now but I haven't seen it here and I just started reading it myself just this night. I believe this is a follow-up piece to this article: http://www.xbitlabs.com/articles/cpu/display/amd-fusion-interview-2010.html I saw posted in the CPU section.

Enjoy.

 

[Gigantopithecus]

"X-bit labs: But do you think there is a place for standalone CPUs for consumers in the next five years?

Neal Robison: Definitely! In the short-term future there will be standalone CPUs without graphics cores. [...] [It is in our interest] that the transition from [CPUs to APUs] happens as quickly as possible because of the obvious benefits of combining a CPU and GPU together."

...You know he was thinking "and because we do it better than Intel!"

Thanks for the links - I had missed those and I've been trying to stay as updated on Fusion's development as possible.

 

[ArchAngel777]

"AMD Talks fusion" that says it all. That is all they ever do is talk. Fusion is like 5 years late. Well maybe that is hyperbole, but still...

 

[OCGuy]

"AMD Talks fusion" that says it all. That is all they ever do is talk. Fusion is like 5 years late. Well maybe that is hyperbole, but still...

Pretty much. Bulldozer better be a hit with consumers or it could be another 3-4 years of buy-out and bankruptsy rumors for AMD.

 

[Gloomy]

Definitely a lot to process there with the article, it probably won't be discussed as much simply because it covers so many topics. I thought this was interesting though:

X-bit labs: What about Bullet Physics engine in general? What developers use it?
Neal Robison: Right now a very small number of PC titles [designed by PlayStation developers] use it. But I know that since Bullet is open-source, people like DICE has used portions of that library to complement their own physics implementations.
X-bit labs: Should we expect games with OpenCL GPU hardware-accelerated physics in 2011 or 2012?
Neal Robison: I don't think there will be a large number of such games this year. Hardware-based physics does not seem to be a huge priority for software developers. We want to make it available as we have that technology, but it seems like a lot of developers are still choosing to use their own physics implementations simply because they want to make sure that performance and gameplay is consistent for all of their customers; a large number [of game developers] are just really concerned about making sure that the experience the consumer has is consistent no matter what graphics card they have in their system or whether they have a very powerful discrete GPU or not. The technology is there some developers will take advantage of it...

I'm still a little confused about the whole physics war thing. I mean who wins if Bullet becomes the go-to physics thing? It'll run on Nvidia cards same as AMD right?

X-bit labs: A general stupid question that you have received for a million of times, but a little bit paraphrased. Do you think that APUs will challenge - eventually - standalone CPUs and GPUs?
Neal Robison: I think APUs will definitely challenge standalone CPUs. I believe that the future of consumer as well as commercial computing environments are characterized by the ability to present a compelling visual experience. Taking a GPU core and a CPU core and using them together on one chip will definitely challenge standalone CPUs.
I do not think that APUs will challenge discrete GPUs on anything, but on the lowest-end systems. When you look at adding a discrete GPU that enhances performance of the graphics side, it makes a huge amount of sense as it scales [performance] on a wide amount of applications because of the rich visual experience that everybody expects now when they are actually using their computing device.
X-bit labs: But do you think there is a place for standalone CPUs for consumers in the next five years?
Neal Robison: Definitely! In the short-term future there will be standalone CPUs without graphics cores. [...] [It is in our interest] that the transition from [CPUs to APUs] happens as quickly as possible because of the obvious benefits of combining a CPU and GPU together.

Hybrid Crossfire? That would be a reason to go AMD-AMD on a new system. Maybe that's why Nvidia's changing her tune concerning SLI on AMD systems?

 

[cotak13]

"AMD Talks fusion" that says it all. That is all they ever do is talk. Fusion is like 5 years late. Well maybe that is hyperbole, but still...

You can buy Brazos fusion products now and Llano's not far off.

As for it being late. Did you expected it to be quick? There's two ways to do it quick and right. AMD could have stuck a GPU onto CPU die without any thoughts to architectural designs and sold it as fusion from the get go. But that would be engineering by marketing.

And of course there were delays due to 32nm process issues.

And they have shown what fusion can enable vs a fast traditional CPU in their Llano vs i7 SB video. So no. I don't think it's just talk.

 

[jimbo75]

"AMD Talks fusion" that says it all. That is all they ever do is talk. Fusion is like 5 years late. Well maybe that is hyperbole, but still...

Yeah at least it got here, unlike Larrabee.

 

[Idontcare]

As for it being late. Did you expected it to be quick? There's two ways to do it quick and right.

In all walks of business we say "For any new product you can only have two of the following three be true: (1) have the product come to market quickly, (2) have the product be low-cost to develop and manufacture, (3) have the product be high-quality and high-performing".

Given AMD's financial health this past decade I don't think they really had a choice about selection (2)...they couldn't throw tons of money at the project...leaving them no choice but to choose between prioritizing the delivery timeline versus the performance target of the product.

On their shoestring budget, AMD surely could have rolled out Fusion sooner, but it prolly would not have been a product that performed well enough that consumers would have bought it.

 

[Morg.]

As for it being late. Did you expected it to be quick? There's two ways to do it quick and right. AMD could have stuck a GPU onto CPU die without any thoughts to architectural designs and sold it as fusion from the get go. But that would be engineering by marketing.

Well, afaik, the company that does that seems to fare quite well xD

But yeah, AMD are well positionned for the future since they bought ATI... They may not have the most powerful single die GPU, but they've got CPU and GPU in the same company.

Although I'd never use one of those combined chips in a gaming PC right now, I can see fusion-type products replacing every old-style CPU before 10 years.

 

[Dark Shroud]

I'm still a little confused about the whole physics war thing. I mean who wins if Bullet becomes the go-to physics thing? It'll run on Nvidia cards same as AMD right?

Yes Bullet uses OpenCL so it's card neutral.

We probably won't see a big push for GPU Physics in general until PCI-EX 3.0 is released or the next gen of consoles.

Hybrid Crossfire? That would be a reason to go AMD-AMD on a new system. Maybe that's why Nvidia's changing her tune concerning SLI on AMD systems?

That's what I'm thinking.

 

[SlowSpyder]

In all walks of business we say "For any new product you can only have two of the following three be true: (1) have the product come to market quickly, (2) have the product be low-cost to develop and manufacture, (3) have the product be high-quality and high-performing".

Given AMD's financial health this past decade I don't think they really had a choice about selection (2)...they couldn't throw tons of money at the project...leaving them no choice but to choose between prioritizing the delivery timeline versus the performance target of the product.

On their shoestring budget, AMD surely could have rolled out Fusion sooner, but it prolly would not have been a product that performed well enough that consumers would have bought it.

Somewhat similar to this, in the car world we say, "Cheap, fast, reliable. Pick two."

 

[Dark Shroud]

Although I'd never use one of those combined chips in a gaming PC right now, I can see fusion-type products replacing every old-style CPU before 10 years.

Well those are what's in the new slimer 360. Sony had also tried to make to do it with their CPU cell stack and couldn't. So they just slapped that beefed up Nvidia RSX into the PS3.

So we'll probably see PowerPC style APUs in both the Xbox 3 & PS4.