On Mon, Apr 4, 2011 at 3:18 PM, Michał Pleban <lists@michau.name> wrote: > Hello! > > I have been quiet for some time, but I spent that time productively ;-) > > In this blog post, I have coined a probable theory explaining why the > 8088 board does not work in Commdore 610: > http://www.cbm-ii.com/commodore-610-and-720-chip-inventory-comparison/ Well sleuthed! > Basically, according to the schematics, the chip U20 is supposed to be > 74S86 quad XOR gate (with Schottky diodes). But on my 610 board, it is a > plain 7864, which is a slower version. Coincidentally, this very chip is > used to feed the memory RAS signal to the 8088 board! Interesting. I recall a similar issue with a 74S74 vs 74LS74 on a 68000 design I worked on in the 1980s (an embedded intelligent serial card, so nobody here would have seen one unless they were working on DEC hardware doing IBM Mainframe serial datacomms). The important point was that our design engineer made a *big* deal about not subbing a 74LS74 for that particular part. It could have been speed, and it also could have been fan-out draw affecting signal slew. Either way, S logic is "beefier" (and hotter) than LS logic and sometimes, you just can't use LS. A quick check shows me that (sweeping generalizations) S is 3X faster than LS, but uses 10X the power. > I will order a 74S86 and try to make a replacement. I look forward to hearing about your experimentation. It sounds like you have good odds for success. -ethan P.S. - if the 74S86 doesn't help, consider trying a 74F86. F logic has a reputation for being noisier, but it's approximately the speed of S logic at half the power consumption (and then there's AS logic - 50% faster than S and 1/2 the power). Message was sent through the cbm-hackers mailing listReceived on 2011-04-04 21:00:07
Archive generated by hypermail 2.2.0.