TR: AMD has been talking about Mantle for what seems like forever now, and yes games are coming out with it, but it's been a jolty launch. Where do things stand with Mantle now? And what's the long-term plan for it?
RH: We've got a release candidate driver at the moment, and we'll wrap that up at some point this year. Then we start to look at things like Mantle 2 and the future, and that's a very interesting space.
Mantle 2 - if it takes us about a year to get through a Mantle iteration - then Mantle 2 will come around the same kind of timeframe as DirectX 12. DX 12 brings a lot of the goodness that Mantle brought. We had a lot of conversations with Microsoft about what we were doing with Mantle, and in those conversations, they said, 'OK, if you really can solve this problem of building a better throughput system that runs on Windows, then we'd like to take that to Windows as well and we'll make that the extra software functionality that comes in DX 12.' So that's how DX 12 has come about.
We'll take the leanings from DX 12 and take it to Mantle. I'm sure they can help us do a better job than we've done. We will also take the extra hardware features with DX 12, and there are at least a couple of key features which are coming there.
They are pixel synchronization, which let you do some cool transparency effects and lighting transparent substances which is very, very hard on the current API. There's something called bindless resources which is a major efficiency improvement again in how the GPU is running, making sure it's not stalling waiting for the CPU to tell it about some of the changes that are needed. At that stage we'll have Mantle 2.0 wrapped up which covers the same kind of functionality as full DX 12 and gives all the performance benefits that Mantle currently gives plus anything else we learn.
Then we look to the future because DX 12 is not the end of graphics. Mantle 3, Mantle 4, etcetera give us the opportunity to expose any of the new features that we develop in our hardware. There are some that have speculated that Mantle will die when DX 12 arrives, that we'll just put it down and walk away it. Heck, why would you need it?
Well, the answer is it's perfect for portability. AMD does graphics in a variety of places, and [Senior Vice President and Global Manager, Global Business Units] Lisa Su mentioned that about half our business by the end of next year will be on the traditional PC platform and the rest will be elsewhere. Elsewhere, there will be AMD graphics.
DirectX is a generic APi. It covers Intel hardware, it covers Nvidia hardware and it covers ours. Being generic means that it will never be perfectly optimized for a particular piece of hardware, where with Mantle we think we can do a better job. The difference will dwindle as DX 12 arrives. I'm sure they'll do a very good job of getting the CPU out of the way, but we'll still have at least corner cases where we can deliver better performance, measurably better performance. We think we have a good future with Mantle, and games developers can tell you they don't want us to drop it afterwards.
TR: If I'm consumer looking at AMD with a Gaming Scientist now, what can I expect from you in six months, a year, two years?
RH: Six months you shouldn't really expect a huge change. The transition points that are coming are for things like 4K gaming, which is a gentle transition. You see DX 12 coming end of 2015 and you will see Mantle 1.0 released at some point this year. That will happen in the next six months we're committed to doing that within the year.
I think the long-term benefits that I hope to bring through my work at AMD will show much more in the 2-5 year timeframe.
TR: Just curious, what do you think of AMD's APUs, like Kaveri?
RH: I like the balance that is in there. When you build an APU you have to decide how much of the chip you dedicate to graphics and how much to CPU. I think with our APUs we've put a good deal of the emphasis on the GPU because for many consumer uses, the CPU is naturally fast enough anyway - going to the internet, running your document viewer or something like that.
When it comes to a game all the heavy weight lifting that needs to be done by the game is actually done on the GPU. On an APU that means that something like 70% of your silicon should actually be dedicated to the graphics part of it. It should actually be different from Intel's balance; Intel put more emphasis on the CPU, and you pay your money and you take your choice. If you want heavy emphasis on one side, then you can choose where you go. The balance that we have is definitely focused on giving the best gaming experience that we possibly can for a given amount of silicon.
TR: All about gaming, again.
RH: Yeah. It's a really big deal. It's a really big deal because most of the other uses of PC don't push it that hard.
To read the first part of this interview, click here.
