Last Christmas saw the release of Quake 3 Arena, the latest in a long line of first person shooters from id Software. Almost a year on and yet another new patch has been released for the game, adding some new features and removing some old bugs. Normally we would be applauding a company for supporting their product for so long, but the new patch doesn't stop there; it also changes some vital parts of the game, resulting in an outcry from the small but highly vocal online community surrounding the game.
White Men Can't Jump
As with most other Quake 3 patches released in the last year, the new version (v1.25 if you're keeping count) is incompatible with all previous versions, meaning that you can't play online unless you can find a server running the same version as you. But as well as the usual changes to network protocols and encryption that render Quake 3 completely useless out-of-the-box, v1.25 also tinkers with the very heart of the game - its physics and weapons balance.
id Software described the changes to the game's physics as the "correction of newly discovered errors", but these errors have been known about almost since the game's initial release. Why have they chosen to wait this long to track down the causes of the problems and fix them? Surely, a whole year after the game was first released, the company shouldn't still be making these kinds of fundamental changes? Especially given that Quake 3 is no longer "just a game", it is also now an established sport, with hundreds of participants around the world, and millions of dollars in prize money up for grabs each year.
The problem is that hardcore gamers discovered soon after Quake 3's release that the distance you could jump depended on your frame rate. This kind of bizarre problem is nothing new - some trick jumps in Quake 2 could only be accomplished if you had a high enough frame rate, and the new famous rocket jump was only made possible by a quirk in the physics code of the original Quake. What is new is that this summer somebody with far too much time on their hands actually worked out what was causing these odd glitches in the physics code...
Coriolis worked out that the variations in jump height and distance are caused by simple rounding errors in the physics code. How often the computer works out your position depends on your frame rate, and inaccuracies creep in due to the limited number of digits stored. For example, if your vertical velocity is 1.2156 and the computer can only store three numbers after the decimal point, then it gets rounded up to 1.216, and you have just gained 0.0004 units of upwards speed.
That might not sound like much, but at certain frame rates these rounding errors tend to be positive much more often than they are negative, and as they can be happening dozens of times each second, the effect is small but very noticeable. The result is that jumps which shouldn't be possible, such as the hop up to the megahealth on the popular map Q3DM13, can be achieved .. if you have the right frame rate.
After this was discovered, Orange Smoothie Productions added a new feature to their OSP Quake 3 mod, allowing everyone, regardless of their frame rate, to have their location sampled at (ironically enough) 125Hz, a rate which had been proven to allow longer jumping distances. It seemed like the perfect answer to the problem; now everybody could jump like a pro. Within a week though, id Software had released their own solution to the problem in Quake 3 v1.25.
Unfortunately id Software's solution was to make the rounding errors always be negative, making you move slower the higher your frame rate. This also makes all of the trick jumps which relied on having a particular sampling rate impossible. Needless to say, this has a drastic effect on the way that an experienced gamer can move around some of the levels, including maps such as Q3DM13 and Q3Tourney2 which are used in many professional tournaments.
And that's not the only major change that was made to Quake 3 for the new version. The default amount of ammunition which you get when you pick up a lightning gun has been reduced from 100 to 50, making the weapon far less useful in maps such as Q3Tourney2, especially where there isn't a ready supply of ammunition available.
The spawn time for the megahealth has been locked to a constant two minutes, whichever map you are playing on, whereas previously on Q3DM13 the power-up reappeared every 35 seconds. As the megahealth is possibly the most important item on Q3DM13, changing the timing of it completely changes the tactics of the map, and Q3DM7 (a popular teamplay map used in many tournaments) is also seriously effected by this change.
And another widely used tournament map, Q3Tourney4, has also been radically changed by the patch. Not only does it have a megahealth which now respawns at a different rate, and some trick jumps which are now impossible, but it also has thin walkways. Previously if you fired rockets or plasma up at the under-side of those walkways, anybody standing above would be effected by the splash damage. Originally this was considered a feature of the map, but now id are claiming it is a bug, and have removed it.
This really doesn't make sense though when you think about it, as the "bug" is very obvious, has been used to great effect by pro-gamers for a number of months now, and even the game's AI bots use this feature. How can id now turn around and claim that it is a bug that they've only just spotted? Removing this feature changes the nature of the map, as without the ability to cause splash damage through the walkway floors, a player standing on the top level has a huge advantage over a player on the ground below.
Bring It Back
The confusion continues though, as the v1.25 patch was labelled as a public beta, and id initially promised to take account of feedback from players. Perhaps unsurprisingly, much of the feedback went something along the lines of "It's a nice idea, but if you were going to change things so radically, why didn't you do it a year ago?"
Unfortunately it soon became clear that id wasn't really interested in hearing what the players had to say, with Robert Duffy of id declaring in a response to GameSpy.com that "splash damage through floors was a bug; it is fixed, live with it". Hardly conciliatory. In fact, just a few days ago he told fan site PlanetQuake that "we are not dropping any [of the new] features at all" in the next patch, although the controversial new physics code will be tweaked somewhat to "give that same feel back, or at least something very similar". Whether this will please the hardcore gamers remains to be seen.
So where do we stand now? The community is split between two versions of the same game which play in radically different ways, with some servers running the new beta 1.25 and some running 1.17, the last official release. Most of the major Quake 3 tournaments and leagues are sticking with v1.17, for the moment at least, because that is what everybody is familiar with. And worst of all, it's all going to change again with the next patch; the new physics code won't be the same as the v1.25 behaviour, but it won't be a full return to the original game's either.
The ridiculous thing is that this isn't the first time id Software has done something like this, but they don't seem to have learnt their lesson. Back in 1998 they released a patch for Quake 2 which radically changed that game's physics. There was an outcry from the fans, as a result of which the next patch returned the game to its original physics.
Sadly this time it looks like id are going to continue to fiddle with the game's basic physics and balance instead of going back to how it was before. If this had happened nine months ago when the game was still fairly new, the chances are that there wouldn't have been much of a fuss. But now that the game has been out for almost a year and everybody has got used to its little quirks, the idea of continuing to mess around with the fundamental rules of the game is rather absurd. Surely it's time id stopped tinkering with Quake 3 and got on with their next game?
Disclaimer - The opinions expressed in this editorial are those of the author, and do not necessarily reflect those of this site, its staff, or the gimp.