Church organs may be mechanically complicated and super old-school, but they share something in common with the earliest computer sound chips. In theory, and largely in practice, they produce very simple waveforms. The primary reason that church organs seem so full and rich compared to your old Commodore 64 is that they have the benefit of a whole church’s worth of reverb to fatten out the sound. [Linus] demonstrates this with the Sixtyforgan.
The Sixtyforgan is a Commodore 64 hooked up to a spring reverb tank. By running the relatively basic waveforms from the Commodore’s SID chip through this reverb, it’s possible to generate sounds that are eerily similar to those you might hear at your local Sunday service. While we won’t expect chiptune luminaries like [chipzel] to start busting out songs of praise at events like Square Sounds, it’s kind of awesome to think of the composers of antiquity rocking out to some mad Game Boy jams way back when.
It’s a great demonstration of the Commodore’s musical abilities, and we particularly like the application of the chromatic button layout borrowed from the accordion. We’d love to see this setup combined with an orchestra of the retro computers, like this demonstration playing The Sugar Plum Fairy. Alternatively, Billy Corgan on the Sixtyforgan playing Tiberius would be pretty great, too. Pretty sounding video after the break.
Wonderful! What a clever idea! :-)
I like this one too!
Interesting experiment, clear explanation, cool project!
Super nice, there’s even “clack” of keyboard, which you can sometimes hear near church organs (made by valves).
It is the sound of the tracker not the pipe valves
Pipes move the air whereas electric sounds need enough speakers to move the air within a given space.
Nice work. That’s some serious talent. I wonder what I could have done if I had talent.
Even the most talented individuals have to work at it, and daft fun idea’s like this can come from anywhere. Find something you want to do and give it a go, keep trying, read/watch anything to educate yourself more widely, talk to others about your project and get the useful feedback. You can look stupendously talented if you do.
Sure some folks find things easier to pick up, I’m one of those sort with anything mechanical, mathematics based – need never have seen that area before but I can jump right into the deep end after some research. However anything language based I’m lost at, being barely able to spell my own language and foreign languages just don’t sink in very well (yes I’m dyslexic). So I find many things trickier than you might expect, never did as well in exams as my understanding of the subject would suggest (still did well, but dropping ‘easy’ marks alot – as questions are always written “use x process on y” – with no context I don’t remember what x is as the name given to it isn’t meaningful to me. Nor am I good at remembering formula so in many maths and physics exams I’d have to recreate the formula they expected you to just use from the more basic assumptions/given information). Programming too is problematic – the read along somebody else’s program is usually really easy, as that is for most languages just a bit of looking up commands that are not obvious by name and formal logic, but actually writing my own means being able to remember the foreign wordy parts of the language…
” Programming too is problematic – the read along somebody else’s program is usually really easy, as that is for most languages just a bit of looking up commands that are not obvious by name and formal logic, but actually writing my own means being able to remember the foreign wordy parts of the language…”
I excel at English, and am good at picking up words in foreign languages and what not. In the right context (with the right need) I am sure I could pick up 4 or 5 languages without too much trouble. And I am RIGHT THERE WITH YOU on programming. I can read others work and modify it but writing my own is pretty hard.
Your words about talent and skill are spot on.
Qué?
J S Bach would be delighted.
This must be a joke…
How do you figure?
Suddenly I remembered a certain basic program on the Amiga Extras disk..
I’ll never hear a church organ the same way again.
I really like the graphical Amiga Basic reference when he plays J S Bach…
I suspect that polyphony consisting of more than 3 notes might have something to do with it too.
A relatively small part of it, I would say. He didn’t demonstrate what six-part polyphony sounds like without the reverb, but I assure you, it still sounds more like a 1970s video game than a church organ.
None of this should be a surprise. The human singing voice starts out with a “blatt” sound and filters it through multiple adjustable resonances to (potentially) produce a far more pleasant timbre, so why wouldn’t this be a significant part of the church organ sound?
Now I have to go look up accordion keyboard. It does put a wide range of notes within reach of one hand, and is easily transposed.
Then: an organ that sounds like a computer.
Now: a computer that sounds like an organ.
Yes. Like he said in video, the circle is now complete.
I enjoyed that video a lot, for both the tech and the music. Well done!
Great project, really enjoyed watching the video. If only he had an MSX, it’s matrix scanning allows to press any key combination and it will reproduce the keys pressed. I used to play with this, holding down lots of keys in basic until patterns emerge and started scrolling :)
Playing games on MSX emulators was hard sometimes as PC keyboards suffer from the same limitations like the C64.
And what just makes it worse, is that the USB treatment for keyboards only allows for tracking the states of no more than five simultaneous keys, which is unfortunate for those of us who have ten fingers.
(For those who are interested, three half-steps up per column as you move east, and two half-steps up per row as you move southeast, so an octave is only four buttons away to the east, or three buttons away north-northeast.)
Whoops – responded to the wrong comment of mine – this was about the arrangement of notes on a chromatic button keyboard on an accordion.
USB keyboards can send the data in different ways. That limit exists with the basic way, or compat way so to speak, used by default. The advanced one can send many keys, but requires a smarter host.
And not only that, but USB keyboards and mice can send data at 1000 Hz, if they support it, just in case someone wants to point another limitation. Basic mode, yes, advanced mode, no.
Maybe HaD writers could do a small article about bInterval and what it means in polling freq, bNumInterfaces bigger than 1, etc. If there was one, write a refresh.
* Six keys, and that’s only when using USB “boot protocol”: https://www.devever.net/~hl/usbnkro
It used to be hard to find a USB keyboard that went beyond the minimum requirements, but there are plenty of NKRO options now.
Maybe I’m reading this wrong, then, but it looks like the USB standard will allow a keyboard to report the states of any arbitrary number of keys. Which is great. Now can you name an operating system that will accept those arbitrary codes?
(Another post vanished or still waiting for moderation…)
Linux does. Checked it when I got the keyboard, and rechecked again, nothing changed so far in handful of years. Gamer keyboard, crappy finish with fadding paint and noisy illumination but the mechanical switches go great (and just set the LEDs to zero, then no buzzing). lsusb -v must show the keyboard has at least two in bNumInterfaces, then one with bInterfaceSubClass “boot” and wMaxPacketSize 8 bytes, and other without and bigger packet. And confirmed with xev, no ghost keys (it has hardware NKRO, and USB transfers all keys fine).
It also lists a tiny bInterval, meaning it reports events more frequently than original USB keyboards (another typical complain: “too much lag, PS/2 is better”… it was, not anymore). Advantage of gamers’ junk.
HaD could write an small article about USB nitpicks: better DIY keyboards, minimum lag gamepads, etc.
Just to be sure we’re not comparing apples and oranges: you have used the term NKRO. This refers to rollover, meaning that if you press multiple keys, they will all be recognized, in the order they were pressed. This has nothing to do with key states, which are significant when you are playing music. That is, not only reporting when each key is pressed but also when it is released. Does the keyboard you mention do this, and does Linux recognize this? If so, then I would have to admit that I was wrong about this.
Yes it does, xev reports everything, I can mash the keyboard and all keep on appearing, both for press and release. The USB mode uses packets of 64 bytes, enough for all keys.
NKRO with diodes is the hardware part. USB non-boot mode is the software part. IIRC some keyboards appeared as multiple keyboards all being “boot” as hacky way, instead of the proper mode, just like others use padding trickery. Read what poiuyt linked. Or the full specs. There is a proper way, and it is used by some.
“Nothing in USB HID limits a keyboard to only supporting 6-key rollover. The specification contains accommodations to allow a host to use a keyboard, limited to 6-key rollover, if it doesn’t contain a full HID stack and can’t understand report descriptors, but this relates only to BIOS setup screens, etc. All modern OSes have a full HID stack and don’t suffer this limitation.” from that page.
Okay, then. Good to know.
This keyboard predates USB by twenty years…
Also, even a USB keyboard can have the ghost key issue; if the keyboard doesn’t have a diode per key, then pressing keys at any three corners of a rectangle in the matrix will produce a ghost press of the other corner.
Cheap keyboards that don’t want to spend the money on diodes arrange the key matrix to prevent common combinations from doing this, but it’s by no means perfect.
Nevermind, I’ve just figured out your comment is supposed to be a reply to BrightBlueJim’s further up. Silly HaD comment system!
As an organist and an early experimenter on the C-64’s SID chip, not to mention a pre-MIDI electronic music studio student who banged into one too many spring reverb units, I am delighted at what you’ve put together here. Bravo!
Goes to show what processing can do to dry sound. The sound of the spring reverb is 100% wet, no dry sound. That’s why Jim James goes into a silo to rerecord a dry guitar track into a 3D soundspace.
For those that want to experiment with spring reverbs, search Radio Electronics Hot Springs Reverb. Much better results than simply driving the input with an amp. The wet output of one of those is better than a lot of digital types, more musical and acoustic.
Back during the C64 era I wanted to use the Bayon layout on it to get more notes out of that little keyboard. A kit and software laid a western keyboard with 2 octaves on top of the C64 and had terrible touch with castanets sound.
Speaking of dry sound: I recently heard a recording of a gunshot made in a very good anechoic chamber. It kind of went “tick”.
Wow, what a beautiful process and result. Not a [typical] hack.
Indeed. I remember playing with a spring reverb many years ago. It didn’t sound anywhere near this good – it mostly sounded like a spring twanging around.
Spring reverbs can sound terribly boingy when you put an impulse through them, caused by dispersion. That is, different frequencies propagate at different rates through the spring. Better sound was through a long pipe, plate, or a reverb chamber room. Nowadays, digital delays do it all.
An old spring delay can give a vintage sound to a guitar amp. Not a fan, I play in a church, where I have PLENTY of reverb. I rehearse with a digital reverb.
Love this interesting experiment, clear explanation, cool project!
I did not know that simply putting a reverb tank in line creates such a wonderful rich sound.
This really is an amazing hack
Like just 3 mintues ago this dude released a video about his accordian make with 2 64s!!! https://www.youtube.com/watch?v=EBCYvoC4muc Coming up – a new HaD article!!!