I watched this the other day - on the one hand, I wish he'd gone into a bit more detail... on the other hand, I know it would all have gone straight over my head. Fascinating stuff nonetheless.
Funny thing, toward the end, when he brought up the question of "Well, if the C64 can do all this, why weren't these techniques used in games?" because that's an attitude I've often taken in my absolute ignorance of coding. Nice to have a clear and concise answer.
Yes indeed, the whole point is to use up all of the computer’s resources to create the most fantastic looking demo.
This where I disagree with him when it comes to modern PCs. He said that there was no challenge, because they can easily generate these effects. However, he’s missing the fact that there’s still a limit, only way, way higher and it will also vary according to the PCs spec, so they can still be pushed to the limit for even more spectacular effects.
I guess with the heat output of modern gaming PCs, they’d just catch fire if worked so hard.
The challenge with the PC demo scene these days seems to be how much they can achieve using as little resource as possible - such as that FPS game in only 96K.
Reminds me of an old acquaintance from the SAM/Spectrum scene, who got himself a job at Microsoft on the basis of what he'd been able to achieve in 48K. While there is little challenge to making these things on a PC simply because of the resources available - in terms of high-spec processors, vast amounts of RAM and, of course, stacks of development tools - it's a lot more impressive when something that looks and moves well can fit several times over onto a floppy disk, when games these days tend to occupy a few gigabytes.
Because these resources are available, very few will bother coding for efficiency anymore...
There was a common argument in the home computer press (and particularly from a certain person with a fairly high profile in the early SAM scene back in the day) along the lines of "these demos are all very well, but why don't these coders turn their talents toward something more constructive, like games?"
On the one hand, I can get behind that argument. It really would have been interesting to see what some of these people could have achieved by turning their limit-pushing and special-effect-creating skills toward an interactive experience... but, on the other hand, it rather misses the point of the demo scene, which was, in some ways, like art-house moviemaking - delivering a precise audio-visual experience that hasn't been seen/heard before, and making people wonder how it was achieved.
Several demo groups have gone on to develop games, and I seem to recall that with a handful of notable exceptions, a lot of them were pretty disappointing - either unoriginal or just all-round pedestrian, with the possible exception of the music. Then you have that thing of fairly basic games with delirious demo-style effects thrown in just because (I'm looking at you, Jeff Minter!). Do they make the game any better? Probably not... but it's showing off what the hardware is capable of while also delivering a solid gaming experience, and it certainly makes for one hell of a signature style.
I kind of get the impression that the main reason demo coders didn't/don't make games was that they didn't/don't want to make games...
Exactly - if I remember correctly, he also had a habit of trying to tank a game if its creators didn't want him to publish it, which pretty much says it all.
The other thing that amazes me about demos, generally, is the way they present the appearance of achieving something spectacular while, behind the scenes, doing something fairly simple. The ray-traced juggler animation is a case in point - it's a mostly static image with a section of the screen running a short, pre-defined animation - there's even a version of it on the SAM Coupé:
And then there's that C64 driving game that (I think) came up in the WhatsApp chat some time ago:
If you look at the roadside detail frame-by-frame, it's really not that impressive but, in full motion, the effect is excellent (if a little repetitive). Likewise, the NPC cars all stick to their lanes and drive at a perfectly consistent speed.
Which brings us to the great philosophical question: if a demo does something simple, yet presents the appearance of achieving something spectacular, is it achieving something spectacular nonetheless?