Phew. If you haven’t seen or heard from me in the past few weeks, it’s cause I was flat out preparing for linux.conf.au, aka LCA.

I foolishly chose a talk topic which is entirely open-ended, it can never be finished, leading to me spending endless days and nights working on it, always thinking of “just one more thing” to investigate.

In the end the talk went down fairly well, I had more material than I needed – preferable than the opposite situation. No one walked out, and everyone I’ve talked to said they enjoyed it, so that’ll do me.

If you weren’t one of the blessed few there, there’s a movie available here (or here), and my slides are here – though they’re not that useful on their own.

I’ve also uploaded the demo movie, if anyone wants to scrutinise it.

The rest of the conference was a bit of a blur personally. I had a fair bit of real work on leading up to it, and during, and also spent a lot of time working on my talk. Lesson for future LCAs, finish the talk before going!

I still managed to have a good time though, blur or not. The Penguin Dinner at the Queen Vic markets was good fun, followed by beers on top of some office block, and the PDNS was good fun as usual. The Google party was a pale imitation of its former self, owing to there being no beer after 8:30pm, I guess they ran out of cash?

The hackfest (programming competition) was a success, with quite a few entrants and some pretty solid efforts by both the winners and some of the non-winners – winning a Playstation 3 is obviously worth an all-nighter for a few people – good to see.

Next year should be awesome, Tasmania is amazing, who’s on for a “Riding up Mt Wellington” BOF? 😉

It may be a new year, and we may have a new government in Canberra, but there’s still plenty of causes that require a bit of your hard yakka. And by hard yakka I mean signing internet petitions! (Yes you can have a VB after every petition you sign).

So for a start get yourself over to the Wilderness Society and sign their petition urging the QLD government to protect thirteen of Cape York’s wild river systems. They said they’d do it, but they haven’t – keep the bastards honest.

Some of you might notice that the petition is being run by none other than the original bassist from Petuey. Please don’t bother him with fan mail, he’s very busy.

This video is damn funny, I wish I knew who it was doing the voice-over, anyone know?

Meg and I made roast vegie risotto and grilled fish, it was yum.

Sometime a while back it was my birthday, I almost blinked and missed it. Meg made me the best ice cream cake ever!

It seems there are still people out there who I haven’t regaled with my splinter story, don’t dispair you won’t miss out!

When I had to move out of Dickson I ended up with three cupboards I didn’t really need or have space for. I managed to give one and a half away but the other one had to be dumped. In order to get it in the trailer I ripped it apart and in the process whacked my hand against the half destroyed cupboard.

I was pretty sure I’d got a splinter, it hurt a fair bit and a bit of blood and air came out of my hand, but there was no splinter visible and within a few hours it was healed over.

Once the swelling went down it was clear there was a splinter in there. To cut a long story short, I had to have my hand cut open twice to get the damn thing out – and this is what we found!

I hope I never do myself a serious injury, that was enough for me!

Just a quick plug for Johnny Butler and band, who serenaded us on Monday night.

As usual it was an excellent gig, and sort of a suprise given that it was postponed.

I’ve had a few people over the years say that they’ve seen JBT “too many times”, but it hasn’t happened to me yet, and I’ve lost count of how many times I’ve seen them.

Most people can guess roughly where he stands politically, but I was impressed that he didn’t make any direct references to any party or policies, given that we’re in the heat of an election campaign. What he did say was that people should get out and inform themselves, and think of others before they cast their vote – all things that are hard to disagree with.

In a comment to one of my previous posts, “TimC” asks if the performance of my simple raytracer is “anywhere near realtime”.

At the moment it’s certainly not, although there is definitely the potential for it.

I haven’t spent much effort on optimising the ray tracing algorithm, because I’m more interested in optimising what I’ve got, and the issues I come across that are interesting and different because I’m using the Cell. So at the moment I have no spatial partitioning at all, I just do a brute force raytrace per pixel. Even a simple bounding box would significantly speed up the current implementation, especially for the test scenes I’m using where there’s a lot of empty space.

IBM’s Interactive Ray Tracer

Some IBM guys in the US have been working on a serious raytracer, they call the “Interactive Ray Tracer”, it is seriously impressive. They have a video demo, as well as a cluster version running on three PS3s. From memory their test scene has something like 300,000 triangles, compared to mine which is about 30 or less.

If you have a playstation 3 running Linux you can even download and run it yourself. There’s also a forum if you have trouble getting it going.

Getting somewhere ..

Running live on the PS3, rendering straight to the frame buffer.

With next year’s Linux conference fast approaching I realised that I didn’t really have anything to do a talk about. I’ve done plenty of work in the last year, but it’s all been pretty dull kernel internals, nothing with any wow factor. Mostly just a zzzz factor.

Ever since I started working on Linux on Cell, I’ve been curious as to how you actually program those pesky SPEs. And so this is where I kill two birds with one raytracer.

The raytracer so far, click to watch them spin!

The plan is to write a raytracer that runs on the Cell. Specifically we’re gonna get it running on the Playstation 3, as that’s the most common Cell system. Hopefully in the process there’ll be some interesting lessons learnt and so on, and perhaps we’ll even get some good performance out of it. If not, at least we’ll have pretty pictures.

So far I’ve written a basic raytracer (screenshot above!), starting from nothing, which has been a great learning experience to start with. That gives us a nice simple base on which to begin porting to the Cell. My plan for the talk, if it gets accepted, is to try and take people from pseudo-code right through to an optimised Cell version.