Today has been a bit more of a success than yesterday! Set off to the car boot sale this morning, and almost immediately found a box of solder suckers and spare tips. For 50p, I got a brand new solder sucker, and for 50p more, I got a selection of spare PTFE tips. Then I found a couple of HP infra-red modules for laser printers -- they let the laser printer accept data from an IRDA-compatible handheld device. Now, I wasn't sure if I could use this with my old LaserJet 4+, but given how hard they are to find, I risked it for 50p each. Turns out, the LJ4+ can't connect to 'em. Finally, I spent another 50p on a large boxed set of electronic projects, originally sold by Maplin's in 1982 (it came with original receipt). The really good thing about this set (apart from the chips and solderless prototyping breadboard) is the manual; it's "Adventures With Microelectronics", by Tom Duncan. I wanted that book anyway, and the Maplin's set is a kit of all the parts you need to go with the book.
Back at home, I set about wiring the SID chip to the Aduino and making it work. First step, generating a 1MHz Phase2 signal, was already done. Next, some fairly tricky in-line assembler to make the Arduino synchronise with the clock and generate simulated 6502 bus cycles. Debugging that was made much easier by using the HP LogicDART -- it's a handheld three-channel logic analyser, and yes, it'll talk to HP printers via IR (I need an HP 82240B). Once the timing looked right, I wired all 17 connections to the Arduino (address and data bus, plus control signals), connected up an amplifier, applied power (the 6581 SID needs 12V), and it worked! First time!
The test program plays just a single note, "A" at 440Hz, repeating, but it has a simple ADSR envelope and it seems to be reliable. I made a recording, which is
here (158kb, MP3). I'm now working on a better demo, with an actual melody, but that'll take a little longer. There's quite a bit more to the SID than the AY-3-8910 which I used in the UK101.