tashtari
PIC Whisperer
MacRecorders are still fetching $100+ on ebay, it seems like a market ripe for puncturing...
There seems to be a schematic out there from BMUG that claims to be similar, but looking at it, it doesn't quite match what I understand about the MacRecorder. Mainly, the sample rate doesn't match (9321 Hz), and on top of that, it seems to be resorting to some strange trickery to get the ADC0831 chip's output to match what the SCC expects...
I don't own an actual MacRecorder, but by the sound of it, it operates the serial port in asynchronous mode with a fast clock (16x the baud rate) being supplied over the HSKi pin. Various sources give the sample rate as about 22 kHz, which is about 1/160th the NTSC color burst clock (~3.58 MHz) - this makes sense because 160 is 16 (16 times the serial baud rate) times 10 (start bit + 8 data bits + stop bit).
If this is the case, writing firmware for the PIC12F1840 (or even the PIC12F1822, its little brother) would be pretty straightforward. Just clock the PIC from a commonly-available 14.31818 MHz oscillator, divided by four to get the instruction clock, output that to the Mac's HSKi pin, clock the UART at 1/16th the instruction clock, and repeatedly read the ADC and spit its sample out over the RxD pins. I might need some help with the analog side to get an audio input wired up to the ADC, but... yeah, seems doable.
Am I missing anything? Have there been other failed attempts to clone the MacRecorder that I ought to know about?
There seems to be a schematic out there from BMUG that claims to be similar, but looking at it, it doesn't quite match what I understand about the MacRecorder. Mainly, the sample rate doesn't match (9321 Hz), and on top of that, it seems to be resorting to some strange trickery to get the ADC0831 chip's output to match what the SCC expects...
I don't own an actual MacRecorder, but by the sound of it, it operates the serial port in asynchronous mode with a fast clock (16x the baud rate) being supplied over the HSKi pin. Various sources give the sample rate as about 22 kHz, which is about 1/160th the NTSC color burst clock (~3.58 MHz) - this makes sense because 160 is 16 (16 times the serial baud rate) times 10 (start bit + 8 data bits + stop bit).
If this is the case, writing firmware for the PIC12F1840 (or even the PIC12F1822, its little brother) would be pretty straightforward. Just clock the PIC from a commonly-available 14.31818 MHz oscillator, divided by four to get the instruction clock, output that to the Mac's HSKi pin, clock the UART at 1/16th the instruction clock, and repeatedly read the ADC and spit its sample out over the RxD pins. I might need some help with the analog side to get an audio input wired up to the ADC, but... yeah, seems doable.
Am I missing anything? Have there been other failed attempts to clone the MacRecorder that I ought to know about?
