While nothing concrete is known about the 90 nm refresh of the G70, we can speculate that the product replacing the 7800 GTX will at least reach 550 MHz and that the high end could possibly hit 700 MHz at maximum. NVIDIA will not stop at just clock speed. We can expect that it will have a full 8 quads of pixel shaders, 10 vertex shaders, and 16 “Super†ROPS that will be able to handle the output of those shaders. I have heard rumors that the AA unit will be getting a makeover and it will be able to handle HDR anti-aliasing. I have also heard rumors that texture filtering will also be getting a boost and we can expect texture quality to match that of the older FX series. This product could easily hit 380 million transistors, and with the addition of 90 nm Low-K (remember, the regular G70 is 110 nm FSG- it does not get a transistor performance increase by using Low-K) this product will hit some impressive clockspeeds. One thing that does not look to change will be the memory controller. NVIDIA does not feel the need for a programmable memory controller like ATI has as of yet, and will instead rely on faster GDDR-3 memory to make up the difference.
The introduction of the 512 MB 7800 GTX is another success for NVIDIA in a year that has proven them dominant in the market. Do not expect NVIDIA to have such an easy time next year though. While ATI may have floundered with the R520 and RV530, the R580 and its midrange/budget derivatives are well on their way to the market. ATI is apparently very happy with how the R580 is coming along, and it looks as if all production on R520 parts will be halted by December. This means that the final R520 chips off the line will be delivered in a late February timeframe for fulfilling the final orders for the X1800 series. This means that ATI is gearing up for a new product to replace the R520 around that time. It is likely that the 512 MB 7800 GTX will be overshadowed by an ATI product at that time, but for the next four months NVIDIA will keep a hefty premium on this part. The big question that comes to mind is if NVIDIA will be able to supply its 90 nm part in quantity at that time and offer instant availability for the high end to continue to compete with ATI? While NVIDIA has promised instant availability of products upon release, they have their work cut out for them to deliver by late Winter/early Spring of next year.