-Yosemite Sam,_Roger Rabbit_) (02/14/89)
Here's some more info on the coco III mod.I built a satellite board so I could easily convert the coco III back to the original configuration.I used a 16 pin dip to dip ribbon cable from RS to connect the board to the original 74ls138 socket on the coco.All the pins on the cable run to the same pins on the 74ls138 except pin 4.Run a wire from pin 3 of the 74ls138 to pin 3 of a 7402.Run a wire from pin 1 of the 7402 back to pin 4 of the 74ls138.Connect pin 2 of the 7402 to E clock(I used pin 34 of the 6809 for this).Tie pin 14 of the 7402 to +5v and pin 7 to ground. The NOR gate is absolutly necessary for this to work.The system will not run at all using just e clock as the enable. I'm not sure wether this will work with the older version of the GIME.The original GIME carries a part # of TCC1014.The new one has an "A" suffix. "Sparklies" are linear flashes of light that appear on the monitor screen especially while floppy access is going on under OS9.The "A" GIME was supposed to cure this,but it didn't make much difference (in my case at least).I've been running my system with this for 3 weeks now,and I havn't seen any problems.I'm running a coco III,old multipak(upgraded),dual mode controller, original coco-xt(without the gating modification),2 5 1/4" floppys,and a 10meg hard disk.Prior to the mod,I have was having a lot of problem doing floppy formats after booting from the hard disk,esp when using an interleave of 1.Sometimes it would work, sometimes I'd get an error 253.I havn't seen any reliability problems at all since I made the modification. Vern
flounder@westfort.UUCP (The Flounder) (03/11/89)
It looks to me like you are curing the SCS timing problem on the CoCo 3. There ave been wide-spread reports of devices working properly on the 1 and 2 and behaving intermittently flakey on the 3, due to differences in the SCS timing. Tandy's official response was that peripheral cards should gate SCS with the E-clock --- exactly what you are doing. I ran into this problem (and "sparklies") when I attempted the fix given in Rainbow in Martin Goodman's column March '89. My computer refused to boot OS-9, although it wouldrun Disk BASIC. Alteration of one of the timing resistors allowed OS-9 to boot, but screwed-up the bus timing, so that the voice pak, already hacked to run at 2 mhz, wouldn't run at all. I ended up removing the mod entirely. Apparently, the techie who came up with the fix (Roger Krupski) was working on the earlier (1986) version of the GIME. I'd like to hear from anyone else who tried the fix. Richard Scranton GEnie address R.SCRANTON uucp ...!osu-cis!westfort!Flounder