anakin@gpu.utcs.toronto.edu (Anakin Research) (01/19/89)
Well folks I've got a fix for the LUCAS with the Microbotics STARBOARD problem. Before you start cheering, Its working on my machine but that's no guartantee that it will work on all others. The problem is the wide variance of AMIGA 1000 timing. Let me first describe the fix. I talked to Jerry at Microbotics and he has had great success solving Amiga 1000 bus problems by terminating the bus. This is done by using a small network on each of the data lines and the four control signals *AS, *UDS, *LDS, and R/*W. Each component of the Network consists of a 4.7K pullup, and a 1K in series with a 1000 pf. capacitor to ground. Knowing Jerry's work I immediately made myself one this afternoon. I tried the STARBOARD with LUCAS but unfortuneatly, NO GO, but this termination network did allow me to put more peripherals on the bus than I've ever been able to before. Looking at the logics analyser it seemed that the edge of *AS was alittle slow. I burned a D type Pal for U5 and up she came. Its been running a memory diagnotic for about two whole hours and so far it seems O.K. I've had so many questions about this bug that I decided to post a message about this small success. I need to do some more work, but what I'd like is to have a couple of you who are having the problem, test it out on your systems. E-mail me if your interested. I still have no idea why I have this problem with the STARBOARD when LUCAS works so well with other memory boards. Before I get in trouble with the moderator, I know this belongs in comp.sys.amiga.tech but I was asked in this section so I replied in this section anyone who wishes to respond please do so in .tech. Brad