Yep, I know they aint Sony hehe. The main issue with LRB was that it would have had to be compared against software which was targetted at different architectures and run that inefficiently whereas if it was the target for said software the efficiencies of the hardware could be exposed. It was mainly an issue with it being a little too early in terms of process technology and early in terms of software efficiency. What they essentially did was can LRB 1 and kept it as a test bed, canned LRB 1.0 refresh and then went straight to LRB 2.0 whilst trying to get the software back end up to speed.

Where LRB 2.0 re-emerges is with Sandy Bridge a next generation architecture built on Intels 32nm process. With the ring bus Intel can stitch the LRB 2.0 cores onto Sandy Bridge circa 2011 or 2012 and with some basic improvements in hardware and software efficiency Intel then has their Fusion competitor. LRB 1.0 didn't need to really get released, the main competitor was Fusion which as typical of AMD got delayed anyway so Intel was not in a particular hurry. If Sandy Bridge is year end 2011 then LRB 2.0 integration will probably be the refresh part of Sandy Bridge in 2012.

Slagging realtime raytracing has become popular in the last few years because A) there are a lot of people invested in the current way of doing things and B) for the last few years rasterisation has been able to keep improving via more and more convoluted hacks -- but at the cost of a massive increase in development costs. The article for example points out that realtime raytracing is bad at subsurface scattering. Ok, there are good approximations of the full monte-carlo approach, but even if there weren't rasterization doesn't do subspace scattering at all.

Given infinite hardware resources raytracing will always beat rasterization, it's just a fact. Now the arcticle implicitly argues that given current realistic hardware constraints, realtime raytracing has to be superior in every way (not just with shinny balls, glass, and water) for rasterization to be supplanted. I would argue that it just has to be good enough because of how much development costs would be reduced.

This wouldn't be "extra" power as much as "different" power and the motive would be significantly reducing development costs. Not that I think nintendo would ever be the first company to do it as in my opinion they are a very risk-averse company when it comes to hardware.

That is what I would have thought, however Charlie D at SA says that even Intel's 2013 processor (Haswell, that's two after Sandy Bridge) does not use LRB graphics on die. He does know all about Intel's roadmaps as he gave us the Haswell+1 codename: Rockwell. Intel is conservative, they will want a seperate chip for a while before integrating it.

I respect Charlie D, hes been very insightful and correct on a lot of his predictions. However theres three points of uncertainty here. The first is that the roadmaps are conservative, so they represent what Intel knows it can do which leads onto the second which is that they are always subject to change which leads on to the third which is that they don't always represent the full product lineup in that they keep some commercially important secrets off their roadmap, I.E. If theres 3 SKUs of Sandybridge 22nm refresh, they may only list two on their roadmap.

OK, I agree there's some room for it to happen anyway.

Since Larrabee is so new, Intel will first have it in an expensive, high-end product and then gradually lower the level. It'll also take a few generations to catch up to AMD and Nvidia, they're not going to be perfect on the first commercial go.

So it will be a while before we see a competitive desktop Larrabee in a mobile power envelope, and probably another generation after that before we see on-die integration.

They also have to get the Fusion-style memory controller right; that's the magic sauce that will make Llano work. Just wiring them together will get integrated graphics level performance, you need GDDRx-like access speeds in order to make a real GPU work. Another issue is drivers - none of the Intel integrated parts have had good drivers despite years to perfect them and a stationary target. The 965s never worked, G3x and G4x still sucked and Clarkdale has major shortcomings (though bearable). Look at the difference between the open source drivers and the binary drivers on Linux for AMD cards to see how much drivers matter (open is 1/10 to 1/3 of binary performance depending on game).

Charlie D did say, with regards to consoles, that Larrabee was a dead cert provided it hit performance and timeline milestones. A few months later he said it had missed both of those and was now out of the running. So it comes down to how reliable he is as a source on the subject of Intel, and my opinion is still in the 'unproven' range. (With Nvidia news it's now 'credible').

Charlie has probably heard it from at least two people, he always verifies his sources independantly before he posts an article which is why his hit rate was so high especially in relation to Fermi. This means his article represents the truth as his sources see it however its not infallible. Im actually tempted to ask him what its all about.

The reason why I believe it will show up in integrated form is because Intel is aiming to integrate GPGPU and GPU functionality into their core CPU die and because the future of the desktop market is limited relative to the mobile platforms. If they released an expensive add-on board the developers could safely ignore it due to the fact it would represent a very low fraction of the overall GPU architecture market share. However if they released it as an intergrated GPU first the developers would be forced to adopt it due to the sheer massive market share they have for their integrated GPUs.

If a LRB core is ~15mm^2 on 45nm, they could likely fit a bank of 8 onto any single CPU die they wanted, and possibly 16 of them and still fit within a reasonable die size on their 32nm or 22nm processes. I wouldn't call it a space issue.

As for the direct memory access issue, well its something they have to work out whether its for their standard IGPs or LRB. I don't doubt that its an issue they could solve, especially as they have a cross patent agreement with AMD so given enough time they can simply borrow AMDs implementation if they so chose.

In regards to the consoles, the extension of the deadline for new consoles has likely given them a reprieve. If they do get a design win they already have their samples for the LRB 1.0 processor and they could always make some more if required. The longer the generation lasts the more powerful their position becomes at least relative to Nvidia and IBM as they are one of two companies with unified solutions. If they were to throw in cheaper flash memory it'd probably seal the deal and at this point its a question of where there priorities are and how far they are willing to go to ensure industry adoption of the architecture.