kevin@cornell.UUCP (Kevin Karplus) (03/07/85)
A couple of years ago, I managed to buy two Pratt, Read keyboards directly from the manufacturer. They do not like dealing with single unit orders, and have some of the worst staff I've ever had to do business with. I did manage to get 2 61-note membrane keyboards (no velocity sensing, but a reasonable "organ" feel). They tried to sell me a more conventional J-switch keyboard at twice the price, but I convinced them to let me have the cheaper membrane keyboard. There was no documentation with the keyboard, but the matrix encoding was easy to figure out. You have to build all the electronics (including matrix diodes) yourself. I don't know their current prices, nor whether they will deal with individuals. It might help if you tell them you are developing a new product, and need to prototype it. Write to: Pratt, Read, & Co. Ivoryton, Connecticut 06442 Some old phone numbers (maybe still useful) (203)767-8282 Ivoryton, Conn. (803)639-2461 South Carolina At the Souh Carolina number, I managed to talk to some people who knew what the keyboard was, how it was wired, etc. I talked with no one who knew anything at the Ivoryton number, but I had to place the order through them (after finding out the part numbers from the South Carolina people). If anyone knows of keyboard manufacturers besides Pratt & Read who'll sell in single units, I'd be interested in hearing how to get info about their products. ---------- On another topic entirely, the Digitar algorithms developed by Alex Strong and myself are going to be used in a commercial instrument costing less than $250. I can't say when, or who, or what, but the license agreements have been signed, and royalties paid. Given usual product development cycles, I don't expect anything on the market for at least another year. Meanwhile, the algorithms are simple enough to implement on a Macintosh. Some preliminary studies (not resulting in any working code, but doing some cycle counting) indicate that the Mac is capable of 2 or 3 good string sounds with separate pitch and decay rate for each string. This uses an improved version of the Digitar algorithm that has not been published. I've been planning to write the code, but never seem to have the time. If someone out there who has done some significant Mac hacking (preferably using sumacc) wants to collaborate on developing the guitar sounds for the Mac, let me know. [Note: if this will be a commercial development, a license would have to be negotiated with Stanford. We can do the research for free, but we can't sell anything.] Kevin Karplus