Tom Forsyth, a software engineer working on Intel's Larrabee project, has said that rasterisation is not dead.
An Intel software engineer working in the company's visual computing group has revealed that Larrabee will focus on rasterisation, and not ray tracing like some of the marketing bods would have you believe.
"
I've been trying to keep quiet, but I need to get one thing very clear," said Tom Forsyth on his personal blog. "
Larrabee is going to render DirectX and OpenGL games through rasterisation, not through raytracing."
"
I'm not sure how the message got so muddled. I think in our quest to just keep our heads down and get on with it, we've possibly been a bit too quiet," he continued.
Pat Gelsinger's
now infamous statement about current graphics architectures based on rasterisation being "
no longer scalable and suitable for the demands of the future." at IDF in Shanghai last month is what caused the confusion, Tom, but it's good to know that it was essentially
flat out wrong.
"
Larrabee's tech enables many fascinating possibilities, and we're excited by all of them,"
said Forsyth. "
But this current confusion has got a lot of developers worried about us breaking their games and forcing them to change the way they do things. That's not the case, and I apologise for any panic.
"
There's only one way to render the huge range of DirectX and OpenGL games out there, and that's the way they were designed to run – the conventional rasterisation pipeline. That has been the goal for the Larrabee team from day one, and it continues to be the primary focus of the hardware and software teams. We take triangles, we rasterise them, we do Z tests, we do pixel shading, we write to a framebuffer. There's plenty of room within that pipeline for innovation to last us for many years to come. It's done very nicely for over a quarter of a century, and there's plenty of life in the old thing yet.
"
There's no doubt Larrabee is going to be the world's most awesome raytracer. . . . but it is absolutely not the focus of Larrabee's primary rendering capabilities, and never has been – not even for a moment.
"
We are totally focussed on making the existing (and future) DX and OGL pipelines go fast using far more conventional methods. . . . We would not and could not change the rendering behaviour of the existing APIs."
It's good to hear that Larrabee's engineers have their heads screwed on the right way, because breaking billions and billions of dollars of investment in software based on rasterisation would be a bad thing for the industry – or at least for Intel, a company trying to break into the discrete graphics market. Ray tracing is not the future of graphics, it's a part of that future – it's just one tool in the developers' toolbox and I know for sure that it's not a Swiss Army Knife or Leatherman.
Larrabee continues to interest me immensely and the picture is slowly coming together – one thing is clear from Tom Forsyth's comments though and that's that Larrabee will
not render DirectX and OpenGL applications through an emulator.
Are you excited at the prospects of Intel entering the discrete graphics market? Share your thoughts
in the forums.
They will then begin to produce GPUs that can also act as your computers main CPU via a OS written in CUDA!!!!
haha
Anyways, more competition is good. Nvidia should buy VIA for the x86 license. That way we'll have three CPU/GPU manufacturers.
It's all about branding and developing a solid user/fan base.
I'm sorry, Tim - the promised tenfold increase is just not enough. Unles you mean to compete with Geforce 7-series cards, two years from now!
I think Larrabee will be quite a bit faster than the current generation hardware from Nvidia/ATI (at least theoretically - there's more to it than that though as you know).
Of course it would be reeally good if the gfx card from Intel has the ability to upgrade the GPU/Memory like we do with a normal pc
Personally I think both Nvidia and ATI have terrible driver quality. ATI has much more structure to their releases, but CCC is bloated and unnecessary. Nvidia is sporadic. One driver is great, then one gets released because of a new game and it kills the performance on your other games. It is hit and miss with their drivers.
If Intel puts out a decent set of drivers I will be very surprised. I say it takes them at least a year before their drivers come even close to ATI/Nvidias.
Im not going to take anything they say in untill i see people with the chips running programs people run and benchmarking them for true figures with just a pinch of salt.