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: Petz who wrote (92295)	1/15/2003 3:23:37 AM
From: fyodor_	Respond to of 275872

Petz: One way to judge the health of UMC's process technology is to compare ATI graphics chips to NVidia's. Yes, as you said, NVIDIA's NV30 will be .13µm (as opposed to ATI's, which is .15µm) - and the pss-poor state of TSMC's .13µm is exactly the reason for the 3 month or so delay of the NV30. Very good call of ATI to go with .15µm! The 0.13u NVidia NV30 series is late, but will almost certainly leapfrog the ATI chips.* The comparison is not quite fair - the NV30 is a huge, huge chip. And it's clocked much higher than ATI's (and don't even get me started on that dust-buster NVIDIA calls a cooler, providing initial tales are true). NVIDIA has chosen a somewhat different approach than ATI by going with an extremely powerful core, but doing relatively little in way of revamping the memory subsystem. This may come back to bite them, IMHO. There's no reason why ATI shouldn't be able to launch a higher-clocked version of its chip, even if they stay at .15µm. Additionally, they've already demonstrated a working board using DDR2 memory, so there is even more bandwidth to go after, should ATI want it. NVIDIA is probably not going to get off that lightly, even: When going to .13µm, ATI will likely double the shader units (and might even refine them a bit too), plus allow for longer and more complex shader programs. That would kill every little advantage NVIDIA has and should outscore the NV30 on virtually all benchmarks. The way I see it, ATI will probably announce a souped up version of their top model right around NVIDIA's launch (hoping to steal some of NVIDIA's thunder, as well as osbourne the NV30 to some extent), which will be very competitive with the NV30 (esp. with decent anti-aliasing enabled). When they feel the need (which could be quite soon, if the competitive situation requires it), they will launch a .13µm shrink, along with doubled shaders and various other improvements. And there go another 2-3 poor quarters for NVIDIA. In case you were wondering, I consider ATYT a huge buy atm. (and, lest we forget, there's also the matter of the chipset business, which ATI can only improve in). -fyo