Inno3D Home Page Inno3D Home Page


FAQ News Search Archive Forum Articles Tweaks Technology Files Prices SETI
Visit NVIDIA's home page.
Favorite Pics
Click to Enlarge
Articles/Reviews
OCZ Tech Titan 3
Absolute MORPHEUS
1.0GHz Pentium III
eVGA MX Shootout
nForce Preview
AMD AXIA CPU
VisionTek GeForce3
2001 Spring Lineup
GeForce3 Preview
eVGA TwinView Plus
VisionTek GF2 Ultra
OCZ GeForce2 Pro
Winfast GeForce2 MX
GeForce2 vs Quake3
Linksys Cable Router
GF2 FSAA Shootout
GeForce2 MX Preview
Benchmarking Guide
Quake 3 Tune-Up
Power DVD Review
Live! Experiences
Sponsors
Memory from Crucial.com


FastCounter by bCentral

 Visitors Are Online
Powered by Perlonline.com
Drivers/FAQ
NVIDIA
3D Chipset
Gamers Ammo
Reactor Critical
GeForce FAQ
Associates
Dayton's Misc.
G-Force X Sweden
Maximum Reboot
Media Xplosion
NVchips-fr
nV Italia
Riva Station
3D GPU
nV News Home Page

eVGA e-GeForce2 Ultra Review - Direct 3D Performance

By: Jonathan Martini - March 7, 2001

 

Skip To:

 

Test System

The details of my system:

  • Pentium 3 550E @719MHz

  • 256 Mb of Cas 2 RAM

  • Abit BE6-II Motherboard

  • eVGA e-GeForce2 Ultra

  • Hercules 3D Prophet II 32 MB

  • Official Detonator Drivers version 6.50

  • SB Live X-Gamer (High Quality Sound)

  • 20 GB Quantum LM with ATA/66

  • Windows ME

  • DirectX 8.0a

I chose to leave the sound on for the benchmark results so that realistic scores are provided from the barrage of tests. I've also included scores from the Hercules 3D Prophet II 32 MB (reviewed here) to compare the speed between the Ultra and it's older 32 MB GTS sibling under the CPU intensive 640x480 resolution as well the GTS' 1024x768 @32bits sweet spot.

Now let's see what this chip can do:

3DMark2000

3DMark2000 has become the de-facto standard of Direct3D benchmarking programs or games. It sports a wide variety of test to truly test multiple aspects of any system.

3DMark2000 Performance

The e-GeForce2 Ultra performs admirably at low resolution CPU intensive speeds just as its GTS counterpart. The impact of 32 bit colors only becomes a major performance degradation factor at 1280x1024, which is one resolution higher than the GTS' 32 bit color impact occurring at 1024x768.

I've included the all the results from the barrage of test 3DMark2000 performed on my system:

3DMark2000 GeForce2 Ultra Details

 

Just how much faster is the GeForce2 Ultra? I've compared the results of the GTS and the Ultra at the CPU intensive 640x480 resolution, and at my previous gaming resolution of 1024x768, since that was the sweet spot for 3D gaming on my GTS. The Ultra has since upped my gaming res. to 1280x1024.

GeForce2 Ultra vs. GeForce2 GTS @ 32bits

GeForce2 Ultra vs. GeForce2 GTS @ 16bits

The performance difference is barely noticeable at 16 bit color, whereas the Ultra begins to pull ahead at 1024x768 @32 bit color as the 32 meg frame buffer of the GTS begins to fill and the 64 meg buffer of the Ultra is just aching for more. ;)

Once again I've included the results from all the tests comparing both cards.

3DMark2000 Details GeForce2 Ultra vs. GeForce2 GTS

 

Evolva

Evolva's use of DOT3 Bump Mapping along with a reputable benchmarking demo makes it a suitable benchmark utility to test the Bump Mapping limits of the two GPU's NVIDIA Shading Rasterizers (NSR). The pixel-shading capabilities of the GeForce2 architecture is one of the new features yet to be exploited by many game developers.

Here's a quick quote about the NSRs in our GeForce3 preview:

NVIDIA's Shading Rasterizer is a second hardware processor that's was integrated in the GeForce core and brings with it a technology to render real-time per-pixel lighting. The shading rasterizer can perform seven pixel operations in a single pass in each of the GeForce pixel pipelines.

Haven't seen what DOT3 Bump Mapping adds to a game?

Take this:

Evolva - Normal Mode

Add a sprinkle of this:

Evolva - Bump Map Texture Only

And Voila!

Evolva - Bump Mapped Mode

Evolva's rolling benchmark takes three primary stats; the max frame rate attained, as well as the minimum frame rate and the average. The most important factors to consider for smooth gameplay is a playable average frame rate (which is around 30-40 fps for third person gaming) and a high minimum frame rate so the game doesn't get all blocky in those all important battles with several enemies on-screen at once.

Evolva Performance

I only included the average frame rates for the above tests. The performance hit entailed by enabling bump mapping is quite large, but what's the use of getting an Ultra unless you intend to make your games look their best? 1280x1024 @32bits is my selected gaming res as, once again, the frame rates achieved may seem low for an fps gamer, but are well within the extremely fluid range for less dizzying 3rd person games.

How does the GTS fare against the Ultra at 1024x768?

GeForce2 Ultra vs. GeForce2 GTS @ 16bits

Very similar performance, though the Ultra's minimum frame rate is 16 fps higher.

GeForce2 Ultra vs. GeForce2 GTS @ 16bits with DOT3 Bump Mapping

The Ultra's faster memory bus certainly helps keep the average frame rate in the high 80's with bump mapping enabled.

GeForce2 Ultra vs. GeForce2 GTS @ 32bits

Easy cruising for the Ultra here, as the GTS is starting to break a sweat.

GeForce2 Ultra vs. GeForce2 GTS @ 32bits with DOT3 Bump Mapping

The Ultra's additional 32 megs of on-board memory really make it shine through this texture memory intensive benchmark. It almost doubles the minimum frame rate attained by the GTS!

Next Page: Open GL Performance

Skip To:

 

Last Updated on May 14, 2001

All trademarks used are properties of their respective owners.