Let the beating begin!

trumphsiao said:
I heared the rumor about 7900GTXs were limited to 5983 Pcs. Maybe Someone can pinpoint me why Nvidia dedicated to limiting volume of 7900GTXs
Click to expand...

Well, it must be true with an exact number and everything. They had 5984, but some klutz dropped one on cement, pin side down.

Edit: And the answer is, there is no reasonable explanation for why Nv would choose to limit volume on GTX. So they aren't choosing to.
 
Last edited by a moderator:
Maintank said:
Jesus, they should spin that business off.
Click to expand...


I dunno how can ATI manage to place order in a way consistented with more than 6 GPUs ASIC then meet demand and saturate the inventory at due time with ease.
 
Last edited by a moderator:
1. AFAIK, the process of ATI used is different with NVIDIA.
2. The standpoint of ATI on R520/R580 is different with NVIDIA on G70/G71.

Just like the L2 cache size of AMD K8 is much bigger(almost 2 times) than Intel P4/PM/Core Duo, although their SRAM cell size is almost same in same level process( like 90nm vs 90nm, 65nm vs 65nm).
 
cho said:
1. AFAIK, the process of ATI used is different with NVIDIA.
2. The standpoint of ATI on R520/R580 is different with NVIDIA on G70/G71.

Just like the L2 cache size of AMD K8 is much bigger(almost 2 times) than Intel P4/PM/Core Duo, although their SRAM cell size is almost same in same level process( like 90nm vs 90nm, 65nm vs 65nm).
Click to expand...

ATI and NVIDIA both use a standard cell (though they do have quite a bit of custom work in each GPU) from TSMC. ATI also second sources product from UMC, but they have a different design library for that process (RV515 vs. RV516). Basically it is the same 90 nm Low-K process from TSMC at this time for both companys. As such they use the same design libraries and standard cells. Yes, there will be differences between the two (obviously), but a large portion of each die will be based on standard cell designs from TSMC.
 
You must log in or register to reply here.
Back
Top