pouët.net

unless there's no specific compo for these productions there will be a 486 demo partly coded by me i guess... dunno, some finnish party or some party near finland... i cannot afford long distance trips....

@Marq: 160x100x16 textmode hack is very slow to code for since every other byte must be skipped.

Well, some lowres textmode in 8-bit style is probably the way to go anyway, if aiming for fullscreen rt stuff.

cga, ega and vga all support raster effects. ;). I've love to see the real possibilities of the cga raster palette switching, for example.

Maybe some interlaced palette switching every second line to create more colors. From what I've seen nobody has really pushed the CGA in terms of image viewing and so on, esp. compared to other oldskool gear.

What's a realistic "cycles" setting for dosbox to emulate a 286?
Just to get a feeling...

People say cycles=1000 should be equivalent to an 8MHz 286

Not so sure how realistic it is, but to get a feeling it should be ok...

It's not very realistic, since all instructions are '1 cycle' in dosbox. So slow instructions like div and mul will be unrealistically fast. Use PCEm to get a good feel of speed.

Do you mean this one? http://www.tommowalker.co.uk/pcem.html

New to me but it looks very interesting I'll have to check it out...

I've used PCE ( http://www.hampa.ch/pce/ ) which seems pretty good for what it is, but it has some limitations like having to use floppy images and emulating 8086/speaker only.

floppy emulation :/
i guess i'll just stick to 386 for now.

Use TOPBENCH (dosbenchmark.wordpress.com) to tune dosbox to whatever target you want. DOSBOX is still faster than the real thing but at least with the realtime benchmark you can get close to something.

Scali is right; even with cycles=231 (approx. 4.77Mhz 8088) you get impossibly fast CGA RAM.

That's actually a pretty wonderful emulator.
And it even plays some of the demos which crash dosbox right on the spot.
For serious development it lacks some convenient way to exchange files with the host system, though.
Maybe I should have a look at the source code...

Yes, that's the one.
I tend to develop with Dosbox, because it's more convenient with exchanging files and such (just mount an existing folder). Having impossibly fast compile-times is not bad either :)
And when I want to test performance, I import the binaries into PCEm.

our royal elite ninjas inc. ms-dos team is working on a demo for a summer party.
i wish assembly oldschool demo compo would once again allow 486s...

Well, a decent 486 demo shouldn't have a hard time making it in the Real Wild at Assembly

So CGA dots my first approximation to full demo in i8088 and CGA graphics. Realtime graphics is so hard, but i beleave in LUT, and hand-optimized assembler. I needed CGA artist and PC Speaker musicant. Good i88 asm xor shr shl and other brainfuck bitmask hacker also needed...

I hope that more people will focus on the early PC platforms, like Tronix.
There have been virtually no demos at all for CGA/8088, and only a handful for EGA/286.
There's a few demos that will run okay on a 386 with VGA.
486/(S)VGA has been done to death though, with some of the most legendary demos of all time (not just PC/MS-DOS, but in general) being released for that platform. The 486 era is one of the golden eras in the demoscene, with lots of pioneering of effects. Do we really want more 486 productions? I'd say no, everything has been done to perfection already.

Exploring the limits of the earlier PCs is far more interesting. These have been ignored by the scene so far, so we have yet to find out how far we can push those machines, and what kind of tricks we can do on them.
Because that's what oldskool demos should be about imho: doing clever hardware-specific tricks.
On 486 this was not really the case anymore. Most 486 demos are quite straightforward software renderers (very similar to accelerated AGA Amigas for example). They were interesting at the time, because the CPUs were used to push the boundaries of realtime graphics in general. But in 2013 there is little more to push in that area.
It may be a personal challenge for coders, to see if they can measure up to the 486 demos of the early 90s, but for other viewers I don't think it will be very entertaining.

Sidetracking a bit, on the Amiga 500 it's slightly different. When I watch something like Superoriginal, I see effects that were not really explored on Amiga 500 before, only on more powerful Amigas. And that makes for interesting demos, and interesting competition between different coders (bringing effects to A500 for the first time, or improving upon the first try of your competitors).

So yea, I'd like to see people getting creative within the limits of 8088, 286, possibly lower-end 386, and Hercules/CGA/EGA/VGA.

Oh, and the same goes for the C64 scene obviously. Although it is arguably the oldest demo platform around, there are still a few C64 releases every year where you wonder: "How the heck did they do that!?"
That's what it's all about. That's why you're using an oldskool platform. People have a good idea what the platform is capable of (because the platform specs are fixed, and we know what demos on that platform normally look like), so you want to come up with things that are seemingly impossible to do on that platform.

It's true that, especially during the last few years, quite a lot of C64 demos have bordered on the impossible/sorcery. But I wouldn't say that's what it's all about. After all for every one of those demos there's two demos that are just a scroller with a crappy logo on top. And that's great. Because what it's all about is having fun with a computer you love!

A lot of people grew up with the 486 and have affection for it. It makes perfect sense that they would want to play with it now. And I urge anyone who feels that way to do so, fuck "the audience" and whether or not they'll be impressed. Not that I wouldn't want to see low-end PC's being pushed to the limit too...

Trixter and I recently entered a PCjr demo/intro at @party (the first demo on the platform, actually, not counting Tantro for the similar Tandy 1000). It wasn't big on design, but for a machine which in many ways was even worse than a plain 8088 IBM PC, has some neat 2D effects. He also shared some oldskool PC tips and tricks, including how to do 1024 colors with CGA! It seems there are still some new tricks to be found.

BTW, please don't add the Hornet demo to Pouet yet. The music sounds bad in DOSBox or on a real PCjr (my fault, I only tested it in MOD2PSG2 and XMPlay with a VGM plugin). Also, DOSBox is missing some effects and doesn't quite get the video mode right. The final version should be out soon, with fixed music and various other tweaks, packaged into a 64k intro. And, of course, there will be a proper video capture. :)

Wondering if any of you guys is developing Dos demo or would like to make one?

Well, Scali and myself and a few other people are tossing some ideas around. The PCjr demo was a framework to see if things are possible, and it turns out -- yes, 60Hz effects are possible, some simultaneously even, as long as you understand your limits (CPU and video memory advantages/disadvantages).

I have always hoped that 8088+CGA (and similar, like Tandy 1000 or PCjr) would be explored in at least 1% as much detail as C64, because they are all fixed targets just like C64. And I share the same sentiment as Scali, that knowing what your limitations are makes it all the more fun to try to code right up alongside them.

BTW, I hope to get final version of the PCjr demo + capture done in two weeks... source code has been traveling snail mail :-)

UPS to the rescue. And I don't mean universal power supply, either. (;

The problem may be, 8088+CGA does not have such a legendary affectionate place in the hearts of many scene people, compared to the C64. Ok, you made an 8088+CGA machine do something whacky, but so what. Who cares what the official demoscene critic board say "we want". What those "we" want is irrelevant. What matters is that it's fun to make, and the creators like what they do. There are people who make stuff that they think an average scene person wants, instead of what they like themselves, and that's just stupid IMHO. (Offtopic rant)