512 disk drives later, Floppotron computer hardware orchestra hits v3.0 The Floppotron computer hardware orchestra has reached version 3.0. The question is, where do you even find 512 floppy disk drives? Its creator, Paweł Zadrożniak, tells all. The Floppotron is a marvellous bit of engineering. Its tones frequent many a YouTube video (this writer was rather taken by the rendition of "Take On Me" …
> "The floppy disk drives as well as the scanners are getting more difficult to find as hardly anyone uses them today and they are no longer manufactured. If nobody needs them today, most of them may up in the junkyards."
Both floppy drives and scanners are still manufactured and scanners are still very much in demand. Not sure where he gets that idea from.
They might not be as modifiable as the older generations, but that statement is complete nonsense.
I'll concede a bit on the scanners, as they're definitely still in use. They now have a lot more firmware so controlling their pitch is harder (mostly changing the requested resolution), but changing when they start and stop the noise-making part is probably prohibitively difficult for most modern models. As for floppy drives, I'm guessing someone who has gotten 512 of them knows a bit more than either of us do about how easy it is to find them. If anyone is in fact manufacturing them today, I'm guessing they're very expensive, not generally available, or both. I've found some USB ones for sale, but nothing clearly indicates that these weren't made in 2010 and they probably don't have the pitch control abilities that the ones in use have.
Not sure why you think this. I have been a fan of these sorts of thing for a very long time, this being my favourite and I commend it you:
https://www.youtube.com/watch?v=hkZbAJHeu9w
I merely take issue with that specific assertion.
As for availability of floppy drives, they are poor quality but absolutely still being made although they almost exclusively have USB interfaces. Not having the old-style interface makes it harder to interface with directly in the typical way, sure, but not impossible.
Scanners have declined in utility by quite a significant margin but many manufacturers are still making them. They have rarely had common interfaces that could be manipulated in a musical way like the older style floppy drive. Instead their hardware is interfaced more directly.
"I merely take issue with that specific assertion."
The issue is that your comment only took issue with that specific assertion - rather than recognising the excellence of the achievement in general and having your issue-taking as a by-the-by. Thus it comes across as miserable and pernickety.
It's like commenting on the 1969 Moon landing by saying Armstrong & co "...might not have been quite as prone to blowing up as previous missions, but that waste of fuel searching for a landing spot was nonsense".
It's missing the point.
I love this project and have been following it since the old days. The Floppotron 3.0 definitely sounds better but I wish he hadn't added a synthesizer - it seems impure.
It reminds me of the really old, six decades old, days when we used to play music on an IBM 1403 line printer (and I still have my 1403 ruler, because of course I do). To listen to one playing Raindrops Keep Falling on my Head go to:
https://web.archive.org/web/20060514113719/http://www.computerhistory.org/exhibits/highlights/1403_raindrops.mp3
He does mention in the write up that instead of every floppy drive in a bank playing a single tone, he changes how many are playing over the course of the note to approximate a guitar, or piano etc.
I guess that sort of makes it a synthesizer, but as you say, that's the whole point!
The drives are grouped in columns by the software and the whole column can only play one tone at a time, but with varying number of drives playing. By changing the number of active drives I can make the tone volume change over time and mimic piano keystrokes or plucking guitar string which volume decays exponentially. The envelope can also be used to get other nice sounding effects, like vibration (sinusoidal envelope), etc.
Like Lego, but for audio ;), a surprising and beautiful application of the idea of signal approximation by (finite) expansion in terms of a chosen set of basis functions (Lego - bricks, Floppotron - drives etc.).
One could consider following through formally - perhaps Zadrożniak has done this - and writing an algorithm which would, given a piece music, compute the coefficients of the basis expansion and output the connexion matrix for the basis elements.
Perhaps also one could combine Lego and Floppotron by connecting the basis elements with mechanical linkages so that when the music began to play, the assemblage of Floppotron elements would configure themselves à la Transformers into a robot that would dance to the music it was playing.
https://youtu.be/6CTR3d2Ly80
We've got a (fairly worn out) Utlimaker 2 at work which has pretty loud servos. You can hear what shape the head is drawing from the different sounds because the x and y axis have slightly different tones. A curve will produce a rising and then falling tone (or vice versa), diagonals at different angles are different chords etc.
If someone with more electronics ability than me wired straight into the servos (or motor controllers?) they could certainly coax some tunes out of it.
Or now I think about it, it should be possible to design a 3D shape, which also played a tune as it printed. Although if you wanted it to actually be a printable shape then your tune would have to have a repeating structure with new sounds being introduced slowly.
Fun idea now I think about it.
Now he's had his fun, it's time to put those 512 floppy drives to serious work... create a RAID array of them.
It'd be even more impressive than the one that other person did a while back.
With 512 drives, it'll be able to hold as much as a CD-ROM, and I'm sure that Big Data and many other businesses will be interested in a giga-scale solution like that.
(Disclaimer: I'm not just old enough to remember when 700MB per CD *was* a lot of data. I'm old enough to remember when the 1.44 MB you could store on a *single* 3.5" HD floppy- or even 720 KB on a DD one- was a lot compared to the 120 KB per side my Atari 1050 5.25" drive handled. And I'm old enough to remember when even *that* was almost a luxury and infinitely preferable to the horrendous tedium of loading from cassette.)
You think the Spectrum tape loading was bad? My Atari 800XL did it at 600 baud and I had quite a few tape-only games that took c. 15 minutes (if not longer) to load(!)
Fantastic when you had 5 blocks out of 170 (or whatever) remaining and you got the dreaded "LOAD ERROR - TRY OTHER SIDE". :-(
(Which is why I appreciated the disk drive all the more and wished I'd bought that tape-to-disk transfer package years previously!)
As for type-in listings... I know some people enjoyed correcting the bugs introduced by their own mistakes, but I always found the process gratingly tedious, even for the ones that had checksums to make sure each line was correct.
"Now he's had his fun, it's time to put those 512 floppy drives to serious work... create a RAID array of them."
In 1985 I developed a floppy disk controller that supported four floppy drives in any combination (mixed 5.25" or 3.5", single or double sided, 40 or 80 track and single or double density) and co-authored a DOS that treated the entire stack as a single virtual drive. It was designed and sold for the Acorn Atom.
Those were the days!
That sounds pretty impressive (and something that might have been genuinely useful back then rather than just some proof of concept fun).
Did you manage to sell many? I'd have thought that the market for Acorn Atom peripherals was already pretty much dead commercially by 1985, even though it was only a few years old then. (Wikipedia says the computer itself was discontinued in 1982 and I get the impression it was quickly overshadowed by the likes of the BBC Micro and other shiny new machines).
@Michael Strorm
You've summed it up pretty well, although WP got it wrong - the Atom was still on sale at least until late 1984 and I negotiated with Acorn in 1983 to provide extended support. We only made one batch of the disk controller, and it went almost entirely to university-based research projects. The most interesting one was control and data logging of crack propagation in the monocrystalline turbine blades destined for the HOTOL stratospheric aircraft (which sadly never got off the ground).
The chief advantage of the Atom was the complete processor-native system bus off the back, so you could extend effortlessly into a rack without being tied to an Acorn specification to much. My development machine had a 3U 19" rack attached with half a dozen peripheral devices.
Impressive creativity, but all that work is a fleeting piece of art that feels wasteful.
I wish that someone so talented were practicing restoration and preservation, rather than silly exploitation.
A high-quality VCR, refrigerator, washing machine, etc., is no longer manufactured, today. Worse, when one of yesteryear breaks because one part wore out, the lack of economical restoration/preservation expertise compels us to throw out the device.
The VCR has no new replacement, and today's expensive, overly complex, replacement washer and fridge -- each built to fail in a few years -- are crap and not worth fixing. To the landfill!
Meanwhile, snowflakes are denouncing plastic straws.
These days it is far cheaper to toss a broken something than to try and fix it.
Most consumer items are also assembled without any means of disassembling it.
The older items of kitchen equipment can still be disassembled, fixed, and returned to working operation, but the latest and newest... not quite, since it now mostly have lots of electronics in, and not everybody have the expertise or means to troubleshoot electronical issues - easier just to toss and replace.
Mentality need to change, and it will not be easy.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
