A bit of a change of pace this week, here’s a snippet from a post I made elsewhere on the Ether-web.

Flying through space alone, only to be ambushed by three enemies popping out from behind asteroids within firing distance. Firing off a few shots to distract, and then running like a bitch, dodging like hell through asteroid fields and flying like a crazy man, while screaming like a girl on voice comms for backup. Surviving for a couple of sectors by the skin of my shields while my squad mates are assuring me they’re almost there. Almost intercepted by another bad guy coming in from another direction, but then punching past them and through a gate, to find 3 of my squadmates just turning up on the other side. The green guys come out of the gate to find that it’s turned from a chase into a battle in a matter of seconds, and I get to turn round and join in the most satisfying 4 a side battle. And when they’re all down and we only lost one of ours, I’m jumping up out of my seat and shouting “take that you f&**ers”, and then collapsing back down in fits of laughter, and a smile that didn’t go away for hours.

The game was Jumpgate, a game originally sold to me as “online Elite”; and with a description like that I was instantly sold. Elite has been a favourite of mine, and many others, since before high school. Jumpgate took the same open ended ethos and made it massively multi-player. Even in its heyday, Jumpgate rarely boasted more than 500 players on-line at any one time, but those who did play showed an amazing dedication and passion for the game that turned a small universe into a hive of activity. Combat, deep-space mining and trading all co-existing in space (relatively) well together; however it was the combat that generated the most devotion from the players. From all 3 factions squads and pilots relished the character traits of their factions, and role-played them to the hilt. There was a fair amount of out-of-character banter, but the meat of the game was always the warring between the factions.

Unusually for an MMOG, the complex flight model means that success is based entirely around your skills as a pilot. While you can earn money and experience to gain ready access to better equipment, the only thing that can improve your ability is practice and training. Undoubtedly this puts many people off, but for those with the ability and perseverance to learn, the adrenalin pumping rewards are phenomenal. My own skills in the sky were sadly outclassed by my contemporaries; I had the privilege of flying with the Octavian Vanguard – generally accepted as the finest squad on the server, and certainly filled with some of the most able pilots.

Sadly, a litany of poor decisions on behalf of the developers (Netdevil) sent Jumpgate into a spiral of decline. Arguably it was simply suffering from the natural cycle for MMOs – a big initial following which dwindles as newer and more interesting games appear. However it was clear from the long cycles between patches that there were few people behind the scenes supporting the game, and that the engine was fragile and prone to bugs when any significant changes were introduced.

The combination of stagnating game-play and bitter disputes amongst the player-base with claims of ‘griefing’, mean that today the server population rarely tops 50. A space once filled with new players and the vibrant hum of travellers lies cold and mostly empty; the few who remain live primarily for conflict, and continue to test their combat skills against each other. The much vaunted role-playing aspect has all but disappeared. However there is life in the old game yet, and talk of an upcoming graphical update to bring the visuals forward from their pre-2002 level has a few old heads interested again. And while there is still apparently just a single developer working on the game, no-one seems to want to see Jumpgate retired just yet.

So I get the feeling that the amount of spam comments that come into the blog is actually increasing at an exponential rate. Since I last checked a few weeks ago, more spam has received than Akismet had processed in total prior to this lot. I guess this is an inevitability as the blog is around for longer and appears in more spam-bot lists. Anyway, Akismet is doing its job nicely, and only 9 of the several thousand comments slipped through, and since I have moderation on, none of those made it onto the live site.

Busy week last week – approval meeting for milestone 2 of the current project, and a meeting up in Dundee with 4J as to our next one. One of the things discussed was how much I would say on this blog about it, and for obvious reasons I’ll be keeping shtoom about most of the detail; however its looking pretty likely that we’ll be working with 4J for a few months, on an unannounced title, on unannounced platforms. So that will mean that my blog entries for the next while will be tending towards rants on software and the games industry in general, and commentary on the gaming news.

Well that’s milestone 2 about wrapped up, on schedule and no dropped features. I’d be hopelessly naïve if I thought that two successful sprints in a row was down to fantastic management or process, or even down to Pete and I and our amazing skills; rather more likely is that these first two milestones have consisted of things that we knew we could deliver without fuss. Next sprint is a bit more meaty – two sub-games, with all of the design ambiguity and art bottlenecks associated with content generation. We’ve dealt with the engine functionality needed to deliver the content, now we have to deliver the content itself.

Looking over all of the legacy code though, I am having to resist the temptation to write it all again from scratch. Yes, the code is tested in fire, and exceptionally functional, but in many places it is ugly, unmaintainable, and downright hard to use. Combine that with the fact that it came from a team who were constrained to work in straight C in most places, and the code isn’t exactly clean and easy to read. So with that in mind I’m inclined to use it as a reference point rather than a base to work from, and pare it down to the essentials. But then the businessman in me kicks in, and reminds me that we’re not here to make good code, but rather good games. So the decision is really – what will help us make better games, faster and more reliably. Decisions, decisions…

Lots of positive stuff the last couple of weeks, which is good as it helps to negate some of the awful “grr” moments I’ve been getting, while browsing through some of the worst code in our legacy stock. We’ve been talking to a couple of other companies relating to our next contract, both interesting projects, although we’ll only be able to take on one. Most likely we’ll take the full game project rather than the pitch/demo project, simply down to the likelihood of it actually coming to fruition.

I’ve also ticked off another couple of things in our annoying tax/paperwork burden – quarterly VAT return, and end of year payroll all out of the way thanks to our efficient payroll person. Still have the corporation tax return and annual accounts to sort out, but that should be the last for now.

Milestone 2 is looming (1st of June), and things are coming together for that – a little later than I’d hoped, but such is the way of things. Still no sign of development hardware, but I suppose that’s just the way it goes. I’m sure we’re not hugely high on the list of priority customers!

Not much for posting recently as we’ve had our heads down getting things done. We passed milestone 1 of our build at the start of the month (well actually a few days before – how good are we!), and have been cracking on getting more visible functionality in, less on the under-the-hood work. It was quite a nice moment when we saw the models and environments from our old game going in (the most complicated art I had to hand), after a few false starts.

Most annoying thing so far has been the two days spent on thrashing out a problem in the graphics projection code, only to find out it wasn’t a high level conceptual problem at all, but rather an assumption that our underlying maths library was fine. Turns out we were transforming 3D vectors by a 4D matrix, but ignoring the 4th column. Of course, the 4th column is rarely used for most 3D work, except in projection matrices. D’oh! All fixed, and another story to be marked as done.

It’s right handed. Get over it. It is an arbitrary decision, but so what? Everyone has their own preference for how they visualise things (I for example interpret +z as forwards/into the screen, and +y is up, and don’t really care about x), in the same way that everyone visualises the flow of time differently. The maths is no easier or harder either way, but it’s a real pain to convert between them when people make different assumptions. DirectX doesn’t insist on left handed (no matter what anyone has told you – it provides left and right handed versions of all the view/projection functions that care), OpenGL assumes right handed, all the big modelling tools use right-handed. The majority have spoken already on this, but every single developer that chooses to buck the trend makes the rest of our lives more painful.

And if I ever have to debug another handed-ness related problem, I think I’m going to cut off my right hand and replace it with a sharpened set of axes, then go round visiting all the games studios in the land, slashing and gouging all those who want to argue about it.

Lots of beavering away on our first milestone at the moment, hence the quietness on the posting front. We’re not delivering what we’d like to, due to bits of kit and software turning up late, but we’re still delivering! All nuts and bolts stuff at the moment – nothing you could call a game, or even a full engine.

We’ve been following the Scrum process carefully, and it seems to be working out well. It hasn’t seen any real stress yet, but it has been good in that it gives us visibility of our goals in a very physical sense. It is sometimes very hard to get a grip on a software project, given the lack of tangible results of your work. But breaking down the project into stories and putting them on cards allows us to say “everything that is to go in the end project is on one of these cards. Move all the cards onto the ‘done’ pile, and we are finished.” That is a good morale booster, because each of the cards can be more easily broken down into tasks, and we can say definitively that it’s either done or not. And so we get a good handle on what we have to do each fortnight/month, and whether or not we are on target or not.

I’ve always maintained that the best way to improve team spirits and morale is not having parties, nights out, or team building days – it is to give the team a chance to succeed. Teams are filled with smart people, and they’re aware of whether or not the team as a whole is doing well, and they are likely to be down-hearted if it isn’t. Breaking down the project into milestones gives us a chance to succeed every month, even if the overall goal is harder to see, and hard to hit. Conversely, I think there are few things worse than putting a team through a tough crunch period, then when it’s done saying: “That was great guys. Now straight onto the next one,” without acknowledging the value of that success.

Of course, the end of our second tax year is not nearly so significant a date as the 2 year anniversary of our date of incorporation. But that was yesterday, so I’ve missed it! Gah. In my defence, I was working very hard yesterday, as I was feeling productive. And I still have a wretched viral infection, although I’m not feeling as bad as I was at the weekend, so no celebrating for me. Boo, hiss.

Anyway, here’s a quick link to another new start-up – Power of Two Games, run by Charles Nicolson and Noel Llopis, both of whom post regularly on the game-dev mailing lists and have done some great stuff which we’ve learned from here (not least of which, UnitTest++, a version of which we use at the Company for our own unit testing). All the best to the guys, and it’s great to see that I finally have an office that’s bigger than someone else’s!

Site was down for a little while again this morning, due to this problem. I realise this particular outage wasn’t their fault, but I’ve been annoyed recently at our hosts (Dreamhost), so I think I’m going to take down the link on the side, until I feel confident again recommending them to others. This post by my friend Gossip sums up my take on it really (yes, this is a ‘me-too’ post).

Anyway, didn’t quite appreciate the fact that it’s Easter holiday weekend, and so have a bored girlfriend hanging around, annoyed that I don’t have more time to spend with her. So I’m trying to do bits of work here and there and the rest of the time enjoy a relaxing weekend. I guess this is yet another case of me needing to remember how to work less and enjoy more!

Finally finished Zelda: Twilight Princess last night – very good final sections which made good use of all the game-play mechanics introduced gradually over the game. A properly satisfying final cinematic, and end-bosses that weren’t ridiculously hard like some games insist on. Admittedly, I was crippled and hung over, so had no intention of getting up from the couch and going to the PC to consult the walk-through that I’d been cribbing off the whole game, so it was good I could do the last bit on my own. I think all in all I used the walk through for about 20% of the game, which is a good learn-to-cheat ratio for me.

Also, the end of March saw the end of our financial year, which means that I have to go through all of our accounts for the last year and make sure everything tallies and cross-checks. Oooh, paperwork, yeah. Got to love it. Anyway, preliminary numbers suggest we made a small loss this year – which is great! No, seriously. We made a profit last year, but then Mr I-don’t-like-small-businesses Brown changed the rules so that any profit, no matter how small, gets taxed (previously there was about 10K UKP of profit allowed before tax kicked in). So we made a small loss which ate some of the overall profit we’ve made since incorporation, which means no tax for us this year. Yay! Turn-over’s up around 50% on last year, which is about consistent with taking on Pete part-time.

So, all in all things for 2006/07 went all right. Not great, just all right. I’d have liked to see things kick off last summer when they were supposed to, but the project we were angling for stands or falls based on the government’s whims, and getting money out of them is like bleeding a rock. I have much higher hopes for this year, so tonight I will be raising my glass in toast to the future.