[an error occurred while processing this directive]


 Home

News

Forums

Hardware

Guides

Articles

Buyer's Guide

CPU Prices

Nostalgia

Games

Links

Pricing



Sharky Extreme :



[an error occurred while processing this directive]

Regular Sections

- Private Eye Editorials
- The Buyer's Guide
- Weekly Downloads
- Site Info
- About Us
- Sharkbait Game


After having tested Diamond's Voodoo Banshee product (the Monster Fusion) last week, I figured it'd be rather handy to see just how it stacked up against a reference design from 3Dfx. Hey what better way to spend a weekend in LA? To be frank though, I've always felt it necessary to use a 3Dfx reference design when comparing other OEM versions harboring their chipset. At least you know who's on the right track and who isn't. Then again you can also find out who has gone and added that little bit extra and who hasn't- in other words who has just carbon copied a reference design. Why am I getting my nickers in a twist over this? I don't know… but it's funny when OEMs slap eachother senseless by claiming that their boards are better than their competitors ones because they've not just gone and manufactured a reference design.

The boards themselves are in absolute terms identical. They both carry some 16MB of SGRAM on board and were AGP versions. The only real difference was the addition of an S-Video output and a TV-Out port, which weren't yet working parts. Well to be honest I didn't fancy testing the TV-Out in any case. I could always rub the dust off of that pesky N64 if I want low-res 'free' anti aliased graphics in any case. Other than that the boards were basically carbon copies of each other- right up to the memory vendor, which in both cases was EtronTech you'd expect the same results right? Wrong...

It's all in the drivers- and the differences were indeed marginally significant. Where as Diamond's Monster Fusion scored in the region of 28 frames per second in Quake 2 at 1024x768, the 3Dfx reference board turned out scores that were usually a frame or two higher. Where 3Dfx reference board totally out performed Diamond's was in D3D under the Incoming Lux Et Robur benchmark. The differences in terms of frame rates were really quite high. 3Dfx's reference design was some 30 frames per second faster than Diamond's Monster Fusion in 640x480, which was really odd- but there you have it (I'll check this out with Rage Software). In fact 3Dfx's reference design totally kicked ass in Incoming, not only was it faster than a Voodoo2 but it almost matched two Voodoo2s in SLI. Go figure...

next page




[an error occurred while processing this directive]


Copyright © 1998-1999 Akula Internet Publishing Inc. All rights reserved. Terms, Conditions and privacy information. Site design by Anders Hammervald