Ok, first thing's first: I tried; I really did, to get monkeys to run the benchmarks and analyze them for me. It would have saved me a heck of a lot of work if I had been able to get them to do it, but they were more interested in tormenting Dakota's (fellow 3DGN writer) cats than running benchmarks. That wasn't being very productive, so I fitted the cats with little stun-gun taser-like things on their right front paws to shock the crap out of the monkeys. That looked like it was going to work (as well as being EXTREMELY entertaining for me), and I had one of the monkeys understanding how the console works, then they all started throwing pens at the cats from the bookshelves. So I lured all the monkeys into my truck with fried bananas and dumped them at the local Blockbuster Video store. They seem pretty happy there, even though they're confused about the title of "Seven Monkeys", as the movie has nothing to do with monkeys. I also really hope they don't find that one movie with Matthew Broderick about the government experiments where they irradiate chimps in flight simulators. That would probably piss them off, and you don't want to be anywhere near pissed off monkeys in a video store who are out of fried bananas.

Second, a word of warning: This article isn't going to be your standard 3DGN "laugh about the review filled with toilet humor and references to boobies and evil monkeys cleverly inserted" sort of thing. Ok, so I lied about the monkeys, but the clever part is still lacking. Let's face it - benchmarks and statistics just aren't that fun. They should, however, be very, very useful to you if you want to get the best balance of visual quality (eye candy) and performance out of Q3Test. There are a lot of game options and graphics settings and stuff in Q3Test that nobody outside of id Software has really known exactly how they affect game performance. The last version of Q3Test (1.05) had timedemo broken, so that's part of the reason why. The other reason is that it's really, really boring to run the sort of tests necessary to accurately determine this stuff, and it takes a few hours per system to do so. The rest of this article (until you get to the conclusion) isn't funny. If you're not interested in statistics, it can be downright dull, but the numbers provided are there to tell you why I make the recommendations, and just exactly what sort of differences they make.

Finally, before we get into the lightly-braised-with-breast-milk-sauce meat of the article, here is something of a disclaimer: While I have included comparative benchmarks between my system and Dakota's system as something of an indication of relative performance of TNT cards versus Voodoo2 cards, this article is NOT intended as any sort of authoritative benchmark. The two systems are NOT exactly equal, and though they normally perform pretty similarly, I KNOW that results will probably vary a bit when so and so happens or under such and such test system. Numerous hardware and gaming sites provide benchmarks of the latest and greatest stuff on uncluttered testbed system. The machines are as identical as feasibly possible, usually running absolutely high-end hardware that very few people have, and rarely, if ever, have more than a fresh install of the operating system, the game, and the program being used as a benchmark. In addition, the latest video drivers are the only drivers other than the standard 16 color VGA drivers the system has ever seen. While those benchmarks are cool, and hardware companies usually love those things, they're unrealistic. They are best case scenarios with high-end hardware, and for strictly comparative reasons, "this is the highest possible sort of performance with this kind of configuration, WHICH YOU'LL NEVER GET IN THE REAL WORLD" tests, are fine.

Nobody I know plays their games with only Windows98, their hardware drivers, and a single game on the system. Most gamers I know have 3 or 4, if not 10, games installed, as well as office applications, internet programs, MP3s, and other junk on their system. The two test systems for this review are the sole working and gaming computers Dakota and I have currently. We play games on these, we surf the web on them, we play MP3s, we write for this site in MS Word on them, etc. Chances are, you'll benefit more from tests on systems like ours than on the latest nifty wonder test systems of Bob's Hardware Brothel.