Ever seen a Sinclair ZX music video?
June 21, 2014 1:52 AM   Subscribe

Camouflage! In 1983, Chris Sievey, the man who would be Frank Sidebottom released a single whose b-side contained three programmes for the ZX81, one of which was a music video to be played whilst listening to the record. This was an incredibly difficult thing to do. Now we can all watch it.
posted by feelinglistless (18 comments total) 23 users marked this as a favorite
When I saw Camouflage I thought of "The great commandment" but this is even better.

Frank you were too good for this world.
posted by cirhosis at 2:14 AM on June 21, 2014 [2 favorites]

I met Chris Sievey many times when I was a child. Going to school in Altrincham (only a couple of miles from the glorious Timperley where his statue now stands) I loved his crazy antics. He used to occasionaly turn up at play time with a megaphone and hand out badges. When the teachers made him leave he would stand behind the fence shouting jokes and telling us to pester our parents to buy something with his picture on.

His tour of Timperley makes me very emotional, something the mights of Hollywood have never managed ;)

I think I was exactly the right age to enjoy Frank. His appearances in Oink magazine are forever imprinted in my memory. His music was precisely as daft as I like it.

I tried to buy a copy of "Camouflage!" a few years ago but couldn't find one. Now I don't own a record player and my ZX Spectrum is behind glass on the wall. All good things become history eventually. But as long as I live Chris/Frank will live on. I miss him far more frequently than I ever expected.

posted by samworm at 2:24 AM on June 21, 2014 [6 favorites]

For more ZX Spectrum video on vinyl, see "Rome By Night" by the Italian synthpop group dhuo.
posted by mykescipark at 4:53 AM on June 21, 2014 [1 favorite]

The screen rolls several times during the video, as though vertical sync was lost for a frame. Anyone know whether that was an intentional effect, or is the video recorded from a real device and it's just a glitch in the capture?
posted by jepler at 7:48 AM on June 21, 2014

Can,t quite put my finger on it, but something,s driving me crazy about the video,s text.
posted by Flunkie at 9:59 AM on June 21, 2014 [1 favorite]

Anyone know whether that was an intentional effect, or is the video recorded from a real device and it's just a glitch in the capture?

I'm not aware of v-sync loss being something that you could control from BASIC, but I didn't have the exact same model of computer so no guarantees. :-/
posted by anonymisc at 10:09 AM on June 21, 2014

The ZX-80 & 81 couldn't refresh the screen and run code at the same time, so you either got flicker (fast mode) or ran only during vertical sync (slow mode).
posted by morganw at 11:08 AM on June 21, 2014

…the (unselectable) text on the youtube video does say it was recorded in an emulator, so it's not merely an analog artifact.
posted by jepler at 11:12 AM on June 21, 2014

Pretty sure the zx81 could run code and refresh the screen at the same time. It was just slow at doing it but I programmed many a game on that little wedge and the screen refreshed happily. I think that some of that video might be running in the more exotic machine code rather than basic which was capable of impressively speedy work when coded right, See Monster Maze and Defender.
posted by merocet at 12:02 PM on June 21, 2014 [1 favorite]

Can,t quite put my finger on it, but something,s driving me crazy about the video,s text.

The ZX8x machines had a somewhat limited character set, so you could choose between , and " for your 's.

The screenrolls are probably just repeated SCROLLs, not sync issues.

(also, the video author links to a CAMO2.P file in the comments, in case someone has an emulator nearby and can figure out what keys to press to load it...)
posted by effbot at 12:20 PM on June 21, 2014

(so not sure why I did that, but I mapped the CAMO2.P file through the character table, and the result looks like a fairly simple, pure BASIC program to me. Including a big block of SCROLL statements...)
posted by effbot at 1:28 PM on June 21, 2014

The two modes were fast and "compute&display" (they didn't like the word 'slow')l, and in fast mode you just had unmodulated carrier coming out of the TV modulator. How your TV reacted to that depended rather on your TV

Don't think that's happening here. and moreover there's nothing going on in that program that couldn't be done in BASIC. You could do quite a lot of clever screen things with PRINT, string slicing and boolean operators. including half-decent animation even in slow mode. ZX Basic was a pretty decent dialect.
posted by Devonian at 2:07 PM on June 21, 2014

In fast mode, any time you stopped static displaying to do some computing the ZX would lose the video timing. This made most TV sets of the day vertical-roll back to sync when the ZX started outputting video again. Very short interruptions such as responding to keystrokes in program entry mode would just cause a little glitch, but anything that took more than a few milliseconds would cause a pronounced sync loss. If it was a good fraction of a 16 millisecond frame it might take several seconds for the TV to re-sync with the computer.

It was of course all CRT's back then and scan signals were generated by analog oscillators which were sync-corrected by feedback. I'm genuinely curious what a modern LCD panel would do since they're all digital and could theoretically generate half a vertical frame and instantly re-sync without jittering the display.
posted by localroger at 2:53 PM on June 21, 2014

I remember that BBC Radio One reviewed Camouflage and its video by playing the music track and describing what was happening on the screen of the TV in their studio—using audio to describe a video which was a piece of digital media that had been distributed in an analogue media format. Even in 1983, aged 16, it occurred to me that this was meta as fuck.
posted by Hogshead at 3:24 PM on June 21, 2014 [5 favorites]

I've posted the decoded source code here.

(I was a bit lazy, so all block graphics are mapped to ▪ for now, and all inverted characters have a ~ before them, the rest is as is. The ZX81 had a 22×32 character display, so those very long strings show up wrapped. PAUSE ≥ 32768 means wait for keypress, while lower numbers count 50 Hz frames, so PAUSE 415 means "wait just over 8 seconds". The rest is straightforward.)
posted by effbot at 5:02 PM on June 21, 2014 [2 favorites]

posted by ardgedee at 5:18 PM on June 21, 2014

Didn't Information Society do something like this in the 80s? Every word of this story is true"?

MetaFilter: I flashed ninja stars all about me
posted by ostranenie at 10:22 PM on June 21, 2014

Also: incredibly difficult? Hmm. Difficult, yes, by all means, and totally gimmicky and totally awesome, and groundbreaking, but.. Wait, what was my point again?
posted by ostranenie at 10:31 PM on June 21, 2014

« Older A Triumph of the Wet-Plate, Among Other Things   |   Service with a Swipe Newer »

This thread has been archived and is closed to new comments