Well, I finally got my system running stable enough to do my comparison between my old 16MB Voodoo3 2000 PCI and the 32MN TNT2 M64 AGP that I picked up a week or so ago. WOW. I'm sure there are a few people who will read this and say "I told you so!", because the Voodoo pretty much smoked the TNT2. I would think that the newer card, with twice the memory and running at 4X AGP would have at least -some- advantage, but NOPE, not at -all-. The Voodoo, at regular clock speed, gets around 80 points better in 3DMark2000 than the TNT2 at the highest speed I could push it to and run the test. And the Voodoo outperformed the TNT2 in UT as well, beating it by around 10fps at times. Here are the actual results, from my 3DMark2K runs and my UT testing...
TNT2:
3DMark2K, Celeron 600@750:
CPU/Memory 150/175: 2415 3Dmarks
150/185: 2538 3Dmarks
UT: CPU/Memory 150/185: (ran through 4 successive levels, and then played "Pressure" twice through- results are averages after each level)
52fps, 46fps, 46fps, 43fps, Pressure: 52fps, 49fps
Voodoo3 2000:
3DMark2K, Celeron 600@750:
Stock speed: 2615 3DMarks
150Mhz: 2701 3DMarks
160Mhz: 2793 3DMarks
170Mhz: 2833 3DMarks
180Mhz: 2897 3DMarks
UT: Voodoo3 at 165Mhz (highest I can get it to behave at in UT)
52,56,48,47 Pressure: 55,52
Needless to say, the Voodoo gets to stay- the TNT2 will be making a trip to eBay. Sometimes, newer isn't always better.... ;D
TNT2:
3DMark2K, Celeron 600@750:
CPU/Memory 150/175: 2415 3Dmarks
150/185: 2538 3Dmarks
UT: CPU/Memory 150/185: (ran through 4 successive levels, and then played "Pressure" twice through- results are averages after each level)
52fps, 46fps, 46fps, 43fps, Pressure: 52fps, 49fps
Voodoo3 2000:
3DMark2K, Celeron 600@750:
Stock speed: 2615 3DMarks
150Mhz: 2701 3DMarks
160Mhz: 2793 3DMarks
170Mhz: 2833 3DMarks
180Mhz: 2897 3DMarks
UT: Voodoo3 at 165Mhz (highest I can get it to behave at in UT)
52,56,48,47 Pressure: 55,52
Needless to say, the Voodoo gets to stay- the TNT2 will be making a trip to eBay. Sometimes, newer isn't always better.... ;D