From: Andy Wood To: Robert Hart Subject: Casio music barcodes Date: Friday, August 15, 2014 17:17:13 +1000 Attachment: CASBAR.pdf (35.9 KB) Greetings from Downunder, So, there I was, looking for some historical information on Casio keyboard instruments, and I happened to turn up your page on the Casio barcodes. For a short time I had a Casiotone MT-70, and I spent some time trying to figure out the structure of the barcodes. I even produced a document describing what I found. I would not have remembered exactly when that was, but I see the date is on that document of mine - 1983. Wow! That is over 30 years ago. I was interested to see that you had a slightly different take on understanding the codes, but I'm sure it gets down to the same thing in the end. I was able to create my own bar codes, and also wrote a program that could scan the printed ones. You have probably moved on, but FWIW, I have attached a copy of that document I created all those years ago. If nothing else it would allow you to complete your page with the chord values that you did not have: 0 / 8 = No chord (that you called tacet chord). 5 / D = Major seventh. 7 / F = Augmented. You were correct that the high order bit of that nybble is the indicator of "one step or two step", or as I put it, "full or half bar (measure)". Regards, Andy Wood ============================================================================= From: Robert Hart To: Andy Wood Subject: Re: Casio music barcodes Date: Friday, August 15, 2014 19:03:21 -0500 Thank you for sending your document. The Casio music bar code format still intrigues me. I'll probably update my document eventually. I see your document also: * Says a "return 2" code is possible in the Pitch data. * Gives an alternative explanation for the checksum. * Points out the relative sizes of the thin bars to the thick bars (1:2.5) are more important than the absolute sizes. * Says you can send infrared light into the pen to simulate a read. (Previously, I had wondered if holding the pen up to a computer screen and flashing a black and white pattern would work.) Thanks again. ============================================================================= From: Andy Wood To: Robert Hart Subject: Re: Casio music barcodes Date: Saturday, August 16, 2014 18:22:53 +1000 G'day Robert, At 10:03 AM 16/08/2014, you wrote: >Thank you for sending your document. The Casio music bar code format >still intrigues me. I'll probably update my document eventually. > >I see your document also: > >* Says a "return 2" code is possible in the Pitch data. > >* Gives an alternative explanation for the checksum. Yes, I noticed you explained the checksum differently. The way I was doing it certainly seemed to work, in that it corresponded to what I saw in the Casio-supplied barcodes, but also in that the MT-70 successfully loaded codes that I created myself. I did not really look at the two methods to see why they produce the same result, perhaps there is a simple explanation. >* Points out the relative sizes of the thin bars to the thick bars > (1:2.5) are more important than the absolute sizes. Of course, if you make the printed bars too small, or cut the timing of the virtual bars down too much the load will fail. The 1:2.5 value was the best figure I could come up with from looking at the printed codes provided by Casio, and from my experiments generating real and virtual barcodes. By virtual barcode I mean skip actually printing the barcode and point the wand to the IR LED controlled by the computer. >* Says you can send infrared light into the pen to simulate a read. > (Previously, I had wondered if holding the pen up to a computer screen > and flashing a black and white pattern would work.) Pointing the wand to a LED was similar to what they did in that "Music on Spec" article that you linked to (which I had not seen until now, and I note it was published in 1984, by which time I think I had sold my MT-70). I had thought of trying to do some sort of direct connection like they did there, but I was too scared of damaging the MT-70. I never thought of trying to get it to read from my computer screen. That would have been difficult because that was a sort of "dumb terminal" that I could only communicate with at the blistering speed of 600 bits per second, so getting it to flash at the correct rate would have been something of a challenge. I would also question if the wand would react to light from a screen. The wand I had seemed to respond best to IR. For my first virtual barcode attempt, I tried using a bog standard red LED, and I got no reaction at all from the MT-70. It was only when I tried an IR LED that I got it to work (I left the red LED in the circuit so I could still see when it was transmitting). I also ran into a strange and interesting thing when I tried to create a real printed barcode. At the time I had two dot-matrix printers, one of which could not print graphics at all, and the other of which was also not suitable because it could only do very chunky blocks. I also had a daisywheel printer, and using that I did manage to print some barcodes that the MT-70 could read, by repeatedly printing a vertical bar character, but that was really slow and not very practical. However, I also had an A3 sized pen plotter, and I thought that was my best bet. It did not take me long to create what looked like a wonderful barcode. There was just one problem - the MT-70 refused to read it. To cut a long story short, what I found was that the dark bars created by the plotter, which to the eye looked very black indeed, reflected a lot of IR light, so basically the wand could not see them at all! It took a bit of messing around to find pen and ink that could produce bars that the wand could see. Ah, the good old daze. Regards, Andy