Re: memory corruption on $d011 effects (was: Using large DRAM modules inside the C64)

From: Marko Mäkelä (marko.makela_at_hut.fi)
Date: 2005-10-14 12:06:04

On Fri, Oct 14, 2005 at 10:30:26AM +0200, Hársfalvi Levente wrote:
> I also noticed that not all computers are subject to this problem... I 
> wonder how the "not subject" =-) ones manage to keep the memory content, 
> if the digital logic is similarly buggy in all VIC-II and TED revisions 
> (I seriously doubt that this has ever been changed). ...I have a late, 
> "new" case C64 with the old (but late) MB. ...Once I had to swap its 
> original VIC-II (a 6569R5) to a 6569R3. Whilst it never hanged before on 
> $d011 effects, now it almost always did... Nothing but the VIC-II was 
> swapped... I doubt that any digital logic should be differ in the the 
> two VIC-IIs, and everything else, including ttls, rams, weren't 
> changed... Would someone please solve the mistery?... =-)

I'm by no means an expert on analog electronics, but could it be that
the speed of the transistor(s) that participate in the bus glitch
varies?

As part of my studies, I took some courses on electronics.  I heard that
when chips are produced, the trace width can differ between batches,
because of variance in the chemical process of lighting or etching.

The lecturer said that you should design chips so that they are tolerant
to this variance.  I guess the designers of the SID (remember, they
do not sound alike?) or the VIC didn't take similar lessons. :-)

	Marko


       Message was sent through the cbm-hackers mailing list

Archive generated by hypermail pre-2.1.8.