RSS Feed

HCW Tech Blog

For the latest info on computer hardware, tech, news, video games, software tips, and Linux, check out our new improved front page: HCW Tech Blog

Reviewed by: Bryan Pizzuti [01.10.01]
Manufactured by: Hercules
MSRP: $59.99 (!)


Unreal Tournament - THUNDER.DEM

This test was run using the Thunder demo by Reverend. Filtering was set to bilinear, and detail was set to high.

Well, in Unreal, the 4000 seems to manage to keep pace with the GeForce2MX, for the most part, and even exceeds it in ultra-high 1280x1024x32 resolution. Of course, by then it's still down to an unplayable level. Also note that while trying to benchmark this card in that ultrahigh resolution, several graphical anomalies appeared, and the system tended to reboot itself. This may have been due to an overheat on the card, or the GPU simply being overloaded and unable to sustain itself doing that much work.

WolfMPTest - ATDEMO8

I'm still using the test version of WolfensteinMP, though this may be the last time I do so. ATDemo8 is a good demo for tilting results in favor of HSR routines, since it mostly takes place indoors in narrow corridors. This was run in high detail mode, with trilinear filtering enabled.

The 4000XT is able to make a decent showing here, performing equally to its faster brother in the lower resolution 32-bit modes. The GeForce2MX card can't even keep up in 32 bit mode, though it does well in 16 bit.

Star Trek Voyager: Elite Force

I've recently started using Star Trek Voyager: Elite Force as a benchmark, and I think it will be replacing WolfMPTest in my FPS mark line-up. The 1.2 patch for the game provides a built in demo (demo0001) which is very long, and provides large segments of both closed-in and open-field play. It's also a good test of video card efficiency, as well as showing whether hidden surface removal actually makes a difference. All settings were set to high here, and filtering was set to trilinear. Ansitropic filtering was turned on for this mark. I was unable to get meaningful and consistent scores from the MX based card (this is the last mark I ran, and I believe the card is starting to fail) so only the Hercules Kyro cards are displayed here.

The 4000XT is unable to compete with the 4500 here; the large difference in clock speed results in a large difference in framerate. When transiting from the closed-in corridors to the large open area where the demo takes place, framerate drops VERY noticeably, nearly to a crawl on the 4000XT at high resolution.

Next Page: (5)