On May 17, 19:55, Louis Schulman wrote: My original one was, too. However, I sold it in 1982 and the one I have now is a Mark 1. Does yours have 48K? It's easy to add extra memory to bring it up to full spec, but there is a timing problem that affects some boards when you fit the third bank. I was unlucky and spent days trying to figure out what was wrong when I bought the extra 8 chips, so eventually I sent it to a Sorcerer dealer/repeairer and he not only modified it but gave me a diagram showing the fix. Not very easily. I don't have access to a scanner at the moment. However, David Williams has quite a lot on his Trailing Edge website. That's not too hard. The Pac consists of a PCB with 4 x 2316 ROMs and a 2-chip decoder circuit, all mounted inside an 8-track tape cartridge shell. It's perfectly possible to build a replacement using a single 2764 and a 74LS10. With a little care, the bare PCB can be inserted into the cartridge slot, so you don't even need to cannibalise a "rare" 8-track cartridge. I have a drawing for such a PCB which I'll put on my website for you. Look for it at http://www.dunnington.u-net.com/public/ExidyROMs/ROMPacPCB.ps Minor caveat: although the date on the file tells me I drew this years ago, the need to use it went away and I never built a prototype. It should work, though. There are 17 places to put track pins to connect one side to the other, two places to put capacitors (a 6V 10mfd tantalum and a 100nF ceramic), at the end of the 74LS10 location, and both IC locations indicate pin 1 by a square (instead of oval) pad. The ROM images are already there, just concatenate them (in order) to make a 2764. -- Pete Peter Turnbull Network Manager University of York