It's fun to speculate...and the reason people are so curious about the NV30, and *if* and how much faster it may or *may not* be are very different...
Had the Geforce4Ti been designed the same time as the R8500, and meant to be released the same time as the R8500, and then run into manufacturing delays to put it behind scheule by 5 to 6 months (yes, according to the
president of Nvidia the NV30 was originally slated to arrive in AUGUST), then I'm sure people would have been curious about it as well...
Especially if the delayed model only had a memory bus 1/2 the size of the one released on time, despite faster core and memory speeds...
The 128bit bus, along with what is clock for clock inferior memory, is not helping either.
Couple that with clock for clock weaker geometry processing...and you're looking at some pretty big holes in the design.
No one can definatley say yes or no, it's fun to speculate..
I would not assume the NV30 to truly be a "generation" ahead of the R9700...their featuresets are nearly identical....
.. Although Nvidia is using .13 micron technology, the absence of low-k essentially makes it a downside to the NV30 because it requires more external components...and doesn't deliver really what one wants from 0.13
Tech sites agree they are of the same generation as well....
http://www.insidemacgames.com/news/story.php?ArticleID=6038
http://tech-report.com/etc/2002q3/nextgen-gpus/index.x?pg=1
Suddenly, you see why we are really comparing two cards designed during the same time period, and why there are obvious questions as to which one will be faster