It will surprise no one who knows me that I was quite into computer games as a youngster. Like most of my contemporaries, I played all the embryonic FPS titles of the time, in particular the ones from id Software; Wolfenstein, Doom and Quake. Let’s not talk about how we got copies or their respective age ratings. But I got really into them, not so much actually just playing but in making add-ons and modifying them.
First it was with Doom, which allowed for the construction of third party maps, but more interesting to me was a third party tool called DeHackEd. This was basically a friendly UI that allowed the executable to be binary edited, adjusting constant values, animation sequences and the like. I made new weapons, turned gravity off, all the fun stuff. It’s safe to say that thanks to these possibilities, I got the game development bug. I had a copy of Quake (arguably the first true 3D FPS), but because my computer couldn’t handle it, I didn’t get properly hooked until Quake 2. Quake 2 had a vibrant modification scene, in no small part thanks to id Software having embraced the mod community. My two favourites were so-called total conversions: Action Quake, which I played extensively with friends in a clan, and Gloom, by Team Reaction. First released in 1998, this was a game played by two teams, Aliens and Humans. This in itself was unusual in that there was a fundamental asymmetry there, but that wasn’t all. The teams were class based and one of the classes was a builder, whose responsibility it was to build and maintain each team’s base, without which they would inevitably lose. It was a truly innovative for its time and most importantly lots and lots of fun, if you were prepared to learn how to play it.
Gloom was held back by Quake 2 though, where strictly speaking only the server side code could be modified. This led to compromises and hacks, and despite my huge enjoyment of the game, I couldn’t help but feel the concept deserved to be expanded upon. When Quake 3 was released in December 1999, I decided I’d have a crack at it. In hindsight this was insanely ambitious, not least because with each successive generation of hardware enabling more possibilities, the bar to entry was raised and the total quantity of work increased. Indeed, I think the Quake 3 era of the early 2000s is probably the last generation where amateurs could feasibly pull off a total conversion. Nevertheless in my spare time as a first year undergraduate (which lets face it, is in ample supply), I started experimenting. I don’t remember the exact timings but probably by towards the end of 2000 I had a viable working prototype; a long way short of an actual playable game but proof that it might be within my capabilities, to myself if nobody else.
Now obviously the coding is only one part of making a game, there is no way anyone could do a total conversion alone, so I started putting out feelers in the Gloom community to see if anyone wanted to help out. I got a couple of bites from interested parties and there was a lot discussed, but nothing concrete materialised. It was all a bit of a tumultuous time and there was a fair bit of mine and others’ teenage angst flying about, some of which I now regret. As such I was about ready to throw in the towel since nothing was progressing. Despite this, the thing now had a name, Tremulous. I’m not sure how we arrived at that, I suspect it was just from flicking through a thesaurus. The game design also now had a somewhat of a shape, but it was very ambitious. Originally the intention was that instead of aliens, the team opposing humans would be some kind of mechanical droid. Big tech trees were drawn up and I had a class system implemented that could have handled it, but given the lack of art progress I was getting pretty dispirited.
Around this juncture I got talking to Mike ‘Veda’ Mcinnerney about my woes. Mike had made a map on the original Gloom, and was undoubtedly talented. He was also a handful of years older than me so had a maturity that I lacked, something I probably didn’t appreciate aged 18. Eventually he agreed to help out, making a map true to theme that eclipsed my feeble efforts. Between us we also hashed out the game design, hugely simplifying it and returning to the aliens theme. Simply having some viable art direction and progress was a huge shot in my arm and my enthusiasm was reignited. The secondary effect of having Mike on board was that him trusting me brought a modicum of credibility to the project. I was no longer just a guy with no track record, promising the moon. Subsequently more talent was attracted and it was starting to look like it might actually be possible to make this thing happen. Ultimately Mike didn’t spend that long on Tremulous as he was just starting to get his own professional games career underway. As far as I know he’s still working at Blizzard (of Warcraft and Starcraft fame). Anyway I’m very grateful to him for giving me the impetus to keep going, and indeed everyone that preceded his involvement; maybe nothing came of it but it gave me confidence, if nothing else.
Describing Tremulous’ development this way probably makes it sound more intense than it actually was. The reality was that in between my studies (and later work) things progressed in fits and starts, there were certainly periods of time where nothing got done. On the other hand, it now at least had momentum, and in the distance there was a faint glow at the end of the tunnel. Importantly, in Nick ‘Jex’ Jansens and Robin ‘OverFlow’ Marshall, we now had an art direction and artistic ability, things I lack. Equally importantly, I think we all share a slightly silly sense of humour, which is a good catalyst toward working well together. It’s safe to say that without Nick and Robin’s help, Tremulous would never had been completed. There isn’t really much to say about the development itself at this point, it progressed at a steady pace, albeit sporadically, if indeed you can be steadily sporadic. The point is that work was being done, indeed I don’t really remember much impediment to anything at this point besides just the willingness to put in the hours. Some game design inevitably changed; the one thing that springs to mind is a fairly significant reordering of the alien classes, yielding a more cohesive progression. Several others also chipped in to development, mappers Jan, Gordon, Tristan and Nina, also Paul for saving the day with the human player model late in the day. And I can’t forget Chris for doing much of the sound.
After I graduated things slowed down a bit as I had to get a job, which obviously cuts down significantly on one’s free time, especially in that my first job post-university was in the video games industry; when you’re doing game development during the day that leaves little motivation to do it in the evening too. Nevertheless following a patch of redundancy after the demise of said employer, I had a bit more time to spend. It was now a full five years after I had first put fingers to keyboard and between us we now had an actual playable game. For a game as complex as Tremulous it obviously wasn’t born perfect, so we scheduled regular play tests in order to work out the kinks and more importantly make sure there was some balance between the asymmetric teams. It’s here that I must make a special mention to Asa ‘Norfenstein’ Kravets. He had been a somewhat prolific player of Gloom and had become interested in Tremulous. Initially he joined us for play testing, but it became clear in short order that he had an innate ability to know when things were too weak or too strong and what to do to fix it. I’ve lost touch now, but if Asa didn’t end up in some kind of QA and/or game design role it’s a crying shame. It’s a miracle that Tremulous has any sort of balance at all; there are so many variables that it’s impossible to shake them all into position. Certain game sizes and maps always favoured one side or the other, but in the end it was “good enough”.
In August of 2005, it was finally done and Tremulous was released. What followed was, honestly… underwhelming… and a little disappointing. Don’t get me wrong, people did play it, but it failed to gain much traction, and as far as I remember it was only really at weekends that it got much play time. In many ways this was not surprising, development had taken a very long time and Quake 3 was nearly 6 years old – very much previous generation technology, and as such less enticing to the discerning game player than it had been. In the months that followed we released a couple of patches to address bugs and balance, but it made little difference. Had it not been for a coincident event, I think that would have been that, and it would have likely been only a footnote in the annals of mod development.
That event was the release, literally the week after Tremulous, of the complete source code to Quake 3. We then had a decision to make, do we declare Tremulous a modest successful defeat, or do we shake off the shackles of being a mere mod of Quake 3, and make it a stand alone game in its own right! I think we were all exhausted from a long and at times arduous development process, so it wasn’t necessarily a given that we would commit our noses back to the metaphorical grind stone so soon. I don’t specifically remember, but I don’t recall much dissent toward the idea. Perhaps the slight disappointment of Tremulous’ initial reception spurred us on, I’m not sure. For me I can only say that if we hadn’t got back to work, it would have forever been a personal regret and huge What If? in my life.
Long story short, we replaced all the remaining copyrighted assets from Quake 3 (of which there were mercifully relatively few) and bent the engine into a suitable shape, and on the 31st of March 2006, Tremulous the stand alone game was released. Following the previous year’s disappointment, I had tempered my expectations, but in no time our IRC channel had over a hundred people in it and there were dozens of servers running, all with active games. We were all a bit shocked by this development – I was at any rate. In a matter of a few hours the previous few years’ work had all been validated and in a way it was a huge relief. It’s hard to know why it took off to such an extent when in essence it was identical to the game released as a mod a few months previously. Perhaps it was because we were among the first out the gate with respect to the products of the Quake 3 source code release, or maybe it was just the convenience of having a single (and free) download. I suspect the latter.
In terms of numbers it’s surprisingly hard to know exactly what impact it had, but our file host SourceForge (remember them?) recorded a little over 3 million downloads. Now many of those will be crawlers, aggregators, or just folk downloading more than once, so it probably doesn’t paint a true picture. My own (albeit, started late) tracking of the master server suggests something like 1.4 million unique player names. Again people using multiple monikers or AFK names mean that’s probably not actually indicative. Nevertheless if you look at it conservatively, the likely “real” figure is probably somewhere in the hundreds of thousands of total players, and that’s way, way more than I ever expected would play our little game. Moreover, our web forum (remember them?) has about 10000 members and over 200000 posts. For me this was an unexpected consequence; indirectly creating a community in which friendships were formed and fun was had. Incredibly people still post to it, although rarely, but even more incredibly a very small number of people are still playing Tremulous, almost 20 years later. In early 2007 Tremulous was voted the moddb.com Player’s Choice Standalone Game of the Year for 2006.
This is all to say that of all the things I have made in my time, Tremulous is easily the one I’m most proud of. I also made a bunch of friends in doing so. In particular Nick, Robin and I have kept in touch over the years and still talk from time to time. We finally got around to meeting up in person in 2015, when Robin was visiting Europe.
