pouët.net

Virta

I think it's in a way the most personal and most succesful piece I've done. My then-relationship was disintegrating and a lot of that emotion went to it. Technically it's nothing special, which is often the case with my work, but it succeeds in carrying the kind of distant and bittersweet emotion. Gargaj's soundtrack does most of the hard work with that, but I think the flowing particle field, the falling particles and the ending spiraly thing are among the most beautiful things I've done.

I'll be break the rules and also mention two more. Fairytale is something I have really fond memories of, because that's the big blockbuster that was done as a big group effort and had a really great bunch of people and mood when working on it. I really miss that. And isolated, in sunlight was made really fast and is a very good representation of the state of my mind at the time, as well as a blueprint for things I'd really like to make if I still had the state of mind to make demos. I think it succeeds in pulling you into a world of its own.

I wish I had something more recent to choose here, but I'm old and lazy and complacent so it's still going to be Puroresu No Seishin.

Why do you think the demo succeeded?

I chose a theme that me and maybe five other people on the whole demoscene care about - Japanese pro wrestling - but I managed to communicate my love for the art of puroresu in a way that's entertaining to people who have no idea what Bullet Club is. I don't think you learn anything from it, just because you see a picture of Jushin Thunder Liger for 2 seconds doesn't mean you even remember what you saw a minute later, but I think it does convey the feeling I get from watching an exciting match. If you're the third party to a love letter, you don't need to have all the context to appreciate the vibe.

What could others learn from the demo?

Technically? Probably nothing. I do claim that when it came out in 2020, I was the PICO-8 top dog, but I don't make that claim anymore, not after 4 years of relative inactivity and with The Mind firmly establishing that Haujobb are now the top dogs. Even when it came out, my code was optimized for PICO-8's quirks to the point that a lot of it is borderline obfuscated even if you read the original non-minified source code, not to mention that it was optimized specifically for the latest PICO-8 version at that time. I don't think it will run well on most other versions of the fantasy console, older or newer.

Design-wise? I think I have a flair for sync and pacing, and I think Puroresu No Seishin is where that shines the strongest. Watch it, look at the transitions like you're listening to a jazz drum fill, and see what you can pick up.

What were your biggest hurdles?

I had two core goals - it has to run at 60 FPS with no obvious framerate drops (there are a few times where I use flashes to mask a couple of frames worth of precalc), and it has to be the absolute pinnacle of what I can fit on a single PICO-8 cart. What you see is the result of a lot of tweaking, and ultimately, compromises. It's never fun to concede that something you want to include is just not going to be feasible, but it's part of the deal when you're working on a limited platform.

Toolset, Engine, Style, Directing?

GIMP and Aseprite used for graphics. Picotool used for minifying the code. And of course, PICO-8. All other tools are self made, including a poorly documented asset packer that I'm fairly sure no one else than me has ever used for anything.

My direction style is basically that I start with a very rough idea of the flow of the demo, with a few key scenes in mind, and then I fill the blanks between those scenes based on a combination of vibes and whatever unused effect experiments I happen to have laying around.

The most positive comment you got?

The pair of these: "As a huge wrstling fan I marked out for this. Superb stuff!" and "I have no interest in the topic, but the design is great, and the final text motion blur effect is a wow." I'm glad the demo can be enjoyed both by smarks and people reading this who are now wondering what the hell "smark" means.

The money shot?

That's easy, the Mandelbrot twister. I don't usually do effects for the sake of wow, but when I do, I try to make it count.

My favourite demo from the ones I got involved with has to be Metamorphosis ( https://www.pouet.net/prod.php?which=50127). It is the most complete and closest to what we could achieve as a team for a long time. It has a starting point, and ending and a short story. It still looks ok on big resolutions. Also killer soundtrack.

Both Sisyphus demos are also quite good. Easiest and best "value for time invested" has to be "Rupture". Last noise demos we did (Entropy and Epitaph) will grow on me. I watched them again after a year and thought "damn... that's some good noise there".


Lifeforce and other big Asm demos are ok, they were fun to make but broke me for a few months each time :-) . Something to do with coding all of it as a long spaggheti crapola.

Biggest hardle is ALWAYS the idea, never the implementation. And trying not to copy self and others too much.

I mean really it's Eon but for the purpose of this let's take Way Too Rude as a case study.



While the visuals are pretty basic, the whole thing lives and breathes to the soundtrack. It's like synaesthesia. Every last bit of sync is hand crafted.



Either put your musician in charge of sync or at least ask them for suggestions.



Building an Amiga synth engine and the demo crashing 5 minutes before the deadline.

my favorite is let go, for the atari st. can't say it necessarily succeeded in any meaningful sense (4th place in the compo, britelite had a bad time coding it, and it didn't particularly set the world on fire afterwards either) but i'm still really proud of it. the graphics are wonky, but i still think the presentation is like, almost exactly what i was going for. more importantly for me, it was probably the first time where i tried to write an actual _song_ (with lyrics!) and pretty much succeeded!

like, i know i've worked on demos that have both fared and been better in practically every way, but "let go" holds a special place in my heart.

It's Konsum.


For me, it's raw and honest, on a profound level.


I can only share what I learned from it, and what it reminds me about:

To have fun making shit with friends. To let things fall into place, without forcing them. To embrace every contribution without judgement. To say 'yes' instead of 'no'. To refine and iterate one step at a time. To be open-minded and open-ended. For the sake of it.

We were all equally responsible. And there was no fuzz about it.


To stop caring about arbitrary definitions of success or beauty.


For me, it stood the test of time. It's self-explanatory, done and perfect.
I never felt such a strong sense of something for a few lines of text.
Because it's not about me, but about us.

Seems as if this thread can still be saved so i do my share:


SaboTagE


It didn't. People complained about the video capture for covering too much of the intro but we did receive positive feedback as well.


It has a few effects where reducing a primarily complex effect into one specific use case an outdated 16-Bit computer can handle worked out well.


Revising the memory management over and over again to make it run on a 2MB machine from a DD floppy disk. Making it run on 1MB was out of reach though.


I used TurboAsm on the Falcon, mostly, with a set of hand-written tools to convert input data from Dan and Zweckform. The "demo engine" is based on Evil's pre-release version, by now heavily modified.
Directing was handled by Dan, writing dozends of Post-Its in different colours for each screen and effect and plastering his appartment's floor with these to get the screens into proper order to support the story. Then, 505, naturally, defined the exact timing.

The best comment i remember was Chuck of Dune asking me on STNICCC2015, "The z-buffer, it's a fake, right?" and when i answered "No, it's a true z-buffer", he shook his head and walked away - he apparently couldn't believe that this was possible.

Best thing about it was the collaboration which took us 5 years in total. I think 505 gave up a couple of times but Dan always returned to the project, so in the end, we, as a team, managed with everybody, including raZen and Zweckform, contributing.

I love this thread, keep 'em coming!

Undesigned: https://www.pouet.net/prod.php?which=64506



If I remember correctly, this is first demo, where I bringed a non-XE scener to XE production.



Do not hesitate to ask guys from other platforms to support your prods. Break your barriers.



Seems, people like this kind of design :)

50 shades of k.



Trying to have somewhat a cohesive style.



I personally don't think much could learn from it. To me looking back it was a very by the numbers prod. Which makes it somewhat stale. The fact some technical glitches in there like improper texturing doesn't help matters. And the fact to me its somewhat incomplete.



Trying to make something cohesive thats not boring. I was at the time trying to follow what people *might* like.

Big mistake.

I should have followed what I wanted creatively and not just follow Pouet votes, which is why I been so against glop farming. If trying to win when creatively coding, then I learnt over time that's probably a not so good way to do art. and actually grow from it. Anyone can make a demo people instantly like: just follow trends, but then again is that part of the creative process if you have somewhat a different vision of what you truly intend to be in a piece of art?



* Custom code and custom framework
* Tried to do things with mandelboxes
* Excessive use of alpha blended FBOs
* Rote standard GNU Rocket, 4klang, squishy. I was at the time and still am working on my own LZMA based EXE packer but it didn't get finished at the time.

At the moment for future prods my code morphed to: GL4.6 only (all bindless), a built in sync editor based off Win32 GNU Rocket but with extra interpolation modes (hoping to rewrite in dear-imgui), dear-imgui based shader editor, almost complete C runtime replacement with some C++ classes for "vectors" and "strings", a fairly complete LZMA1 based executable compressor that can compete with UPX for normal executable use, custom image compression, MOD/XM/S3M playback, dear-imgui based UI to the shader editor and shader preview.





"50 shades of ok". Thats about as much as I can hope for.



Not sure, a lot of things can be done a lot better. But it was a stepping stone, I'd love to remaster it someday with the things I originally wanted to, with the new codebase I now have. I'd love to revisit all my old prods and redo them with the new code, so I can die happy with what I've done.



Reflections, with alpha blended FBOs for text, logo, etc. Someone in person remarked how I always do prods with extensive layers, and they loved that.

Great, simple question because for every demo I've coded, I've always felt I've rushed it out. Wish I had a patron to give me time for the art, but since worklife, this has been the case. So I feel everything I've release since is shit.

But that's actually not true. And we should be proud of the demos we make!

The moment or a specific purpose is my only motivation, it seems. As soon as I leave inspiration out, it takes ages and becomes worse (to the point that I have a ton of demos still unreleased, and will likely remain so unless I see a spark somewhere.)

So despite my firsts, I feel like every prod is shit on release.

Maybe a better measure is the one where you had fun and really got into it and enjoyed it, every day for weeks. <3

So by that measure, it's Blenkvectors, Corkscrew, Stunner Dentro, and Blitter Sweet.

Everything else was quicker (tho I enjoyed some of them and especially the collabs with people that made them rather more and better!)

...As a tally though that's rather sad, actually. But there is no way to find weeks of enjoyment like that ever since worklife. (and I'm not being negative, just working it out in my mind how something I'll be proud of will happen again. Because I want to goto there. Please... ';( )

It has a proper flow. Superb loader tech made it possible to have constant stream of great parts. Some of the effects like bumpmapping was just breaking on PC and we pulled it off with C64. The chills still return when I remember Assembly audience roaring as the bumpmap part was shown live.

Do not store your last, greatest, part of the demo on a single floppy and place it on top of power supply with magnets over night.. thus 75% :D

Time and exhaustion as always is the biggest hurdle. Nowadays we try to avoid crunch, but were not very successful with Superselection. For this year, on both of my productions, we agreed that music comes first and rest of the content follows. Made it much easier for me personally. Wode and derpiipo still had their hands full.

In the beginning, we had just a bunch of effects and a great loader tech by Mr.Sex and DrDick. It transformed into nicely flowing demo during two weeks of crunch. We got the awesome tunes from AMJ and graphics from Mike.
Did some supporting graphics with both c64&joystick and amiga&mouse. Poetry using Collins Cobuild, not AI :) .. The whole flow of the demo was planned with pencil on A3 paper, trying to figure out what parts could follow each other from tech point of view - next part is loaded and unpacked while previous effect is running. Every cpu cycle and bit of memory is in use.

I personally enjoy deep bass ooompfs on second part. Unfortunately blocky parts and pallette looks much better on CRTs. Emulator users also easily miss a cool trick when turning disk - the correct disk side is recognized blindingly fast and chaos zoomer begins.

C64 demoscene is still very much alive. We keep getting more high quality prods every year and FTS3 starts to show it's age. Which I think is the most positive thing about it - c64 scene survived over those critical years of Amiga&PC onslaught.

In 2026, FTS3 turns 30 and Byterapers 40. 100% version is still missing...

Unsound Minds - Follow The Sign III is amazing yeah. Music by the Audiomaster J. Smashing sounds.

IMO this thread has been a success so far, lots of good reminders for myself.

Love reading these, moar pls!