Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : Advanced Micro Devices - Moderated (AMD) -- Ignore unavailable to you. Want to Upgrade?

To: pgerassi who wrote (215120)	10/26/2006 8:02:01 PM
From: eracer	Read Replies (2) \| Respond to of 275872

Re: Why don't you understand thermal limits? That's what killed the 4GHz Tejas. And the 10GHz P4... I understand them well. You conveniently ignore GPU thermal limits in your 2GHz example, which is what I was highlighting. A 1x1 Fusion die would be good for OLPCs, PICs and palmtops. At 45nm it would have a $20 packaged cost. Getting $50 for it means 60% GMs. $30 packaged gets you a 2x2 Fusion. Selling it for $75 still yields 60% GMs. Higher speed ones would cover the mainstream market. 4x4 Fusions would be used for the high end HPC markets enjoying the higher margins. 2x6 Fusion pairs are likely to be desired by the extreme gaming crowds. Some of the GPU cores would likely be used for physics to allow that blow it away FPS crowd to get its kicks. Yes, and in an ideal world AMD would have already captured more than 50% market share and had a net income > $1 billion last quarter. Making money in practice is not quite as easy as on paper. You fail to note that less pipelines would be needed at those higher speeds to get the same performance. That is where a lot of the die area reductions would come from... And you missed my point entirely. All I was saying is that without pipeline reductions, which would be an absolute worse case scenario in terms of die size, it would still be possible to produce CPU-GPUs for the low-end market at 45-nm. But they can and state that they will use the GPU as an option for replacing a core in a multicore die. Which flavors would sell into which markets is unknown at this point. The point is that AMD has the flexability to find out which percentage mixes yield the best values and profits in each segment. Also as many have proven before, there is value in the "have it your way" approach. I haven't said anything different. I'm just concerned as to the timeline when AMD delivers the "have it your way" approach. Intel is behind now on the graphics side of the business, but are ahead on transitions to smaller nodes. Intel can potentially deliver many more cores, or more powerful cores, when they move to 32-nm while AMD is still stuck at 45-nm. You are missing the point. Its what their customers decide, not what AMD thinks. Just remember customers can't decide to change the laws of physics. That is the point! Because AMD will allow a OS, OEM or user to decide what the GPU multiplier should be for their given situation. AMD would limit the multiplier to the bin it was packaged for. It doesn't stop the silent PC crowd from lowering the multiplier to go fanless. As for the high speed GPU crowd. That is one of the problems. They don't have the option of a high figure of merit process (process that gets high performance per watt expended). So the power requirements require lots of slow pipes. If they could get their hands on a process where they could double speed at the same power, wouldn't they opt for fewer double speed pipelines? Especially, if they would get them for less money? NVIDIA had the SOI option with IBM. It didn't happen. There was no doubling of frequency and yields were poor on what GPUs IBM did produce. There is no evidence that AMD's SOI manufacturing could deliver anywhere near a 100% frequency increase over bulk at TSMC. Then there is the issue of how AMD plans to come up with the 100+ GB/s of memory bandwidth that high-end gamers will already be accustomed to by the time Fusion arrives. By then the fastest video cards will probably have memory bandwidth in excess of 200+GB/s.