Ninesquirrels, an IndieGamer poster made a great post on a recent thread about an indie dev going full time that I think every indie should read, it makes perfect sense to me.
I just followed a link to this final speech by Randy Pausch, who is a Carnegie Mellon University professor of computer science who is dying of pancreatic cancer. It’s an incredibly uplifting and inspiring speech on achieving your childhood dreams, and I recommend it to anyone who wants some inspiration.
Raph Koster’s company Areaa(sp?) announced what they’ve been doing all this time today, and it sounds pretty mindblowing.
Essentially it’s a web 2.0 inspired, entirely open-platform MMO software thingy. It’s second life, the metaverse (or place) done right. Good job Raph, now give me a job, I want to be that excited when I go to work.
Even if Raph doesn’t give me a job, I can see potentially building some cool shit (and tools for other people to build cool shit) on top of this platform. If you’re reading Raph, hook me up and I’ll make some cool stuff for your launch.
I’m sure all 3 of you read this blog already, but I found this one tonight and am adding it to my Google Reader as well as the blogroll here.
Lennard is the maker of Real Estate Empire and Battle Castles. I think it’s very cool that he made a game about buying and flipping houses when that’s what he does in his real life. 
Totally unrelated to this blog, but I almost missed this opportunity and I don’t want anyone else to do the same.
They’re back for one show in London at a respectable 125 pounds a ticket. However, to get a ticket you basically have to enter a raffle for the privilege of buying one, limited to two per household. The prices on these are going to be absolutely insane on ebay.
I’ve entered, if you love them like I do, you should too
Sorry for the prolonged absence - I’ve been working like a dog at my new job.
That’s starting to calm down now for various reasons, and so starting this weekend I am back at my game. I’ve built a lot of subsystems working in Ogre for my current company that I’m going to just port straight over to the game.
For instance, I’ve had to build a GUI library (I know, everyone’s got their own).. It’s not quite as data driven as CEGUI is, but it’s a hell of a lot more elegant to use. And it performs well, which was the primary reason I built it since I was getting horrible performance with just about every other GUI library. It’ll be perfect for my in-game track editor. I also designed it with the idea that I’ll be writing tools for it.. I tried to do so for CEGUI and it didn’t seem possible to make a nice comprehensive tool with the documentation available, since there was magical things in the XML.
I’ve also redone my app framework a bit, so I’ll be incorporating that as well.. Nothing dramatic there, just a well engineered skeleton framework to build games on.
I really need to get back on this game.. My last few months have all been about infrastructure, and the little bit of games development I have been doing has all been low tech FFP stuff.. I’ve almost lost what little chops I had with shaders, but that won’t take long to get back, I’m itching to get back into it.
Well, not much to report. As expected, the only time I can even consider doing my own work due to burnout is now on the weekends. I’m doing interesting stuff (technically if not artistically) at work, and still learning things - hanging out with great artists is giving me lots of insight into how things should be working.
This weekend I drove 6 hours each way on saturday to pick up a Great Dane puppy. She’s cute as hell, and we’ve wanted one for a while, but that completely shot the weekend of course. I am going to try to finish off the track this next weekend, hopefully it will go smoother.
So this is what it’s like to be an indie with a day job. Damn it sucks.
I’m sure the other game developers reading this out there know the feeling I am having now. My motivation to finish this game (the racer) is nearly toast. In fact, the only thing keeping me going working on it now that it’s served its purpose (getting me a good job) is that I’m not going to let myself put it in the trash after all the time and effort I’ve put into it. I WILL get a working game out of this. And that means that I won’t let myself start on any of the other, more exciting projects I’d like to be working on in my copious spare time either.
That said, progress this weekend has been light but significant.
I had to refactor my entire track system, as the previous incarnation was a hack left over from the very first implementations of the track. So now it’s one unified track object.. That track object owns the meshes, the vertices and the physics object. The track is split up into an arbitrary amount of segments (currently 10) and in the generation process (which I had to do some major surgery on) the segments are now all more or less uniform length. Since this involved moving control points around, this refactor was necessary from that angle.
The meshes generated are individual objects for culling, and when I get lightposts in finally (after this refactor is completed) they will be objects owned by the track as well. The physics object is just one giant mesh of the entire track - this is another necessary change, as seams between objects can cause my car to catch its bottom cube on one and flip into the air.
I also spent some time looking into writing a Python exporter for FXComposer 2, which is fantastic. It has a full (almost completely undocumented) Python interface in there, and I was poking around trying to figure out if I could write an exporter that would export to the Ogre .material format. It looks like I can, but there’s still quite a bit of poking around to do, plus a goodly amount of work. I’ll probably just manually convert them for now, as I’ll definitely be trying to use this new tool.
That’s right, I got hired at a games company here in Vancouver, so things are going much slower than they were before. I’m now down to evenings (if I don’t feel completely fried) and weekends (if my wife lets me.)
That said, I got a good job - I’m the lead 3d guy at a small games company, and I expect to be hiring a bunch more people to work under me, both here and in China.
Now I’ve got to start my Mandarin lessons again.
Things have been going slowly, but I’ve been feeling better the last couple of days and actually have been getting work done.
I’ve got the editor GUI in place and the camera code working again. One of the problems I tackled today (and haven’t finished yet) is making sure that each one of the subsegments of the track is the same size (roughly) - this is needed so that the track textures can be put on in a consistent fashion and look non-distorted.
I figured it would be really simple, but it turns out it’s not. For instance, calling my Catmull-Rom spline code to interpolate the points at 0.1 and 0.2 should give you two points the same distance apart as 0.2 and 0.3, when the spline is straight.. But it doesn’t. The distances between the points get successively SMALLER as you approach 1.0. It’s crazy. Using another CR spline implementation (the one built into Ogre) gives me different results, but still not good ones. I am assuming at this point it’s a limitation of the spline.
Screenshot - current Editor using CEGUI.
So now I’m going to have do some goofy successive approximation stuff with a binary search. It’s really hacky, but I am so tired of trying to find elegant solutions for stupid problems like this. I just want the damn game to work.
Anyway, tomorrow morning when I’m awake and alive again I’ll do it, it should work fine.
I got another simple camera feature in that I never bothered to do in my XNA game - when you select a camera angle other than the rail camera or the free camera in the editor, it frames the current track segment in the frustum and the camera keys let you zoom in and out.. Currently it uses the bounding sphere of the segment, which isn’t perfect, but tomorrow if I get a chance I’ll change it to use the AABB of the object instead, that will work much better for end-on views of the track segments.