keithe@teklabs.UUCP (06/24/83)
I was all ready to report my success at using Alejandro's interface circuit with my Commodore 64. But unfortunately success has so far eluded me and I wonder if others are having the same problems... First of all, unlike Alejandro's circuit, it turns out that the motor control line from the C64 *MUST* be used to turn the drive on and off during program loading. I have discovered, using a borrowed Datasette*, that the C64 loads BASIC programs differently than does the VIC: assuming that the play button has already been pushed, when you type in "LOAD", the screen goes blank, and the cassette starts to go; then the cassette stops and the screen re-appears with the message "FOUND <program-name>". After a few seconds - 4 to 10, approximately - the screen again goes blank and the cassette starts moving again. Finally the screen reappears with the message "READY". At this point the program is ready to run. I seldom have such good luck, however. The problems are usually one of two kinds. Either I get an error message like "? OUT OF MEMORY ERROR", which occurs after the first part of the loading but before the second part - in the 4 to 10 second time period mentioned above. Or, during the second part of the load, I get an error message like "LOAD ERROR" (I've forgotten the exact wording) and the thing prompts with "READY". Anyway, it appears that the interface must include the ability to turn the drive motor on and off; the only problem here is that this is probably best done with a relay (as opposed to a solid- state device), as you may not be able to guarantee the sense of the current flow through the remote on/off connector. This connector may be floating, so don't ground either side of it. Also, the waveform arriving at the input to the C64 is apparently different than that expected by the VIC. Putting a 'scope on the incoming data line from the Datasette* shows an almost square wave, which is then (phase/position) modulated to encode the data bits. This is in contrast to the note Alejandro included about his wave-shape circuitry generating the "pulses that the VIC likes to see". So I modified the circuit to generate a more square-wave like signal. That still results in only very spotty luck loading tapes. It's been suggested that the problem I'm having is that the recorder I'm using is (a) plugged into the wall and, (b) fairly old; so I may be getting 60/120 hertz hum in the output which is screwing up the waveforms to the C64. I'm going to try it with batteries - to eliminate the possible source of hum - and try again. If that works I may try using a smaller input coupling capacitor to reduce the low-frequency input from the recorder. I'll work on it more this weekend and let you know what I find out. On the whole I'd rather have a disk drive... keith ericson - teklabs ------------------- *Datasette is a trademark of Commodore Business Machines, Inc.