Final Words

While we usually don't tackle these types of articles, with all of the confusion surrounding NV30 and its future we thought it was necessary. To summarize what is publicly known about NV30 at this point:

  • NV30 will be a 0.13-micron chip produced at TSMC.
  • The NV30 architecture will be similar to R300 but with more features (e.g. more shader instructions supported), higher core clock speed and faster memory.
  • Although support for a multichip solution is built into NV30, the NV30 platform will be a single chip solution just like every other NVIDIA card.
  • NV30 taped-out just recently and is 3 - 4 months away from final production silicon.
  • NVIDIA will most likely launch NV30 before their November/December shipping dates much like ATI launched R300 and Matrox launched Parhelia.
  • Despite very low yields on the parts, NVIDIA will make sure that enough NV30 parts are available to meet demands for this year.
  • Realistically speaking, December will probably be the month that NV30 will be readily available.

With NVIDIA's relatively spotless execution history since the Riva TNT, it's no wonder that NV30's delay is garnering so much press. It is entirely possible and within NVIDIA's reach to ensure an "on-time" launch of NV30 before the end of this year, but the GPU's already delayed schedule will put ATI in a leadership position that they haven't been in for a while. Should ATI take advantage of this opportunity, it may take a lot more than NV30 for NVIDIA to regain lost ground.

Debunking Myths - A Multichip NV30?
Comments Locked

0 Comments

View All Comments

Log in

Don't have an account? Sign up now