   |
|||
|
|||
|
:
News
:
Archives
|
|
NVIDIA SLI Interview |
|
|
A Few Words from Chris Daniel |
|||
|
|
|||
|
by Josh Walrath |
|||
|
The latest drivers allow the use of a Coolbits option to create or edit custom SLI profiles. Will this functionality ever be exposed without going into the registry? What are you doing to simplify this process for users? Currently, the user only needs to go into the registry once after installing the driver. Thereafter, the user has full control to create new applications profiles in the control panel. That said, NVIDIA is constantly working to provide a better user experiences for its customers with each driver upgrade. The simplest way for NVIDIA to handle this is by adding profiles faster (which is our goal) and by educating developers on the intricacies of programming for SLI, thus eliminating the need for profiles in the future. The enthusiasm for making sure that games scale has been virtually unanimous. NVIDIA is actively working with all developers and educating them so that their titles can benefit from SLI as much as possible. Education was one of our goals at GDC this year and we held several sessions on SLI. When you combine this with the developer events and road shows, we have made big strides with developers in a short amount of time. By developing future titles on SLI technology, game developers are able to optimize their titles for SLI configurations. Additionally, the performance of SLI today gives developers a glimpse of how tomorrow's single GPU systems will perform. This allows them to tune for all elements of the system including CPU utilization. That, combined with the fact we are the only GPU family that supports Microsoft DirectX 9.0 Shader Model 3.0 makes NVIDIA SLI technology a very popular development platform. Developers are generally making sure that any new development systems they bring in house use SLI technology. NVIDIA has stated that it will utilize its Turbo Cache technology in cards other than the 6200 TC. If the high end cards start to utilize TC, how will this affect SLI performance in texture data intensive applications when both cards are attempting to address a portion of main memory through the PCI-E bus? TurboCache is designed to lower on-board memory requirements, enabling system builders to deliver a more feature-rich GPU and achieve graphics performance previously unattainable at entry-level prices. It has been a big hit so far. That said, you know we can not discuss unannounced products.
If you have found this article interesting or a great help, please donate to this site.
Copyright 1999-2005 PenStar Systems, LLC. |
|||
