I noticed this old message when I was cleaning up my mailbox. Levente partially confirmed an old theory on $d011 destroying the RAM contents. Maybe adding the coil would even cure the 6569R1 bug that destroys the RAM contents when switching from hi-res to text (toggling $d011 between $1b and $3b). Marko Date: Sat, 03 Jan 1998 13:29:52 From: Levente Harsfalvi <email@example.com> Reply-To: firstname.lastname@example.org To: email@example.com Subject: Re: Adding a SID, VIC-20 & others (stuff deleted) [C-64 AEC signal] Marko, you may remember, some years ago we discussed how and why could the Plus/4 crash when a routine fiddles with $ff06 (...same as $d011 on C64). I supposed it was failing memory-refresh cycles, but you suspected that TED (and VIC) pulls AEC low too early, and some of the faster ram chips may be able to catch the wrong data flying on the bus. All this happens, of course, if the program forces dma cycles by the videochip's registers. Well, I can confirm, you were right! I got a broken c64c which I fixed yesterday. One of the 41464 drams was burnt, so I had to replace it. But I had no 41464 so I decided to take a 514256 with a socket and to do some 3d-net wiring :-). This was a 70 nsec dram. After assembling the parts it worked on the first try. But as I tested it with several demos, it crashed sometimes. Interestingly, it only happened when it started to play with FLI, dma delay and similar techs. ...I kept on thinking, then tomorrow I seemed to remember this discussion. ...After that, I took a coil and inserted into the Aec line (between the VIC's pin and the board. It seems, the delay I got was enough. No demos crashed since that (even if I tried some of Crest's demos, which I guess contain more than enough d011 effects). BTW the coil is a 2.2uH one (it was from another broken c64 board). ... Maybe, it's too, a good occassion to anyone experiencing such problems to find the solution. Levente
Archive generated by hypermail 2.1.1.