RE: [stella] Stella 2.0 Alpha 2 available

Subject: RE: [stella] Stella 2.0 Alpha 2 available
From: "Fred Quimby" <c9r@xxxxxxxxxxx>
Date: Sun, 26 Jun 2005 20:49:08 -0400
I've been meaning to bring this up for a while.  I did some experimentation 
a while back with trying to duplicate 2600 frying in Stella.  I think it 
might make sense to put this functionality in the debugger.

I tried dozens of different schemes, but the one that worked best 
(replicated reported results correctly about 80% of the time) was this 
amazingly short chunk of code:

if (frying) {
   for (uInt16 ZPmem=0;ZPmem<255;ZPmem+=(uInt16)rand()%4)

The way it was implemented, it repeated itself once per frame as long as a 
button was held down, and usually a short press was all that was needed (so 
the chunk of code may have run two or three times.)  All the code does is it 
clears some random bits in memory and the TIA.  If it's easy to stick this 
code in somewhere and attach it to a button to click on, that would be cool.

I was going to send Stephen a diff of the entire source code, but I mucked 
it up to the point where it wouldn't compile anymore and didn't have time to 
fix it.  But the above is what did the trick, so this is probably all you 

>Here's the latest win32 binary:
>There are a lot of changes, including label completion in the prompt: you
>type a partial label (like "w") as an argument to a command, press Tab,
>and the prompt attempts to complete it (say, to "WSYNC"). This is almost
>exactly like the tab-completion in bash, if you're familiar with that.
>Also added "define" and "undef" commands to manipulate labels, and a
>"savesym" command to save the current symbol table. This will hopefully
>be useful to people who are disassembling commercial ROMs.
>Warning: there are a few commands that lack error checking ("savesym"
>is one of them). If you forget to give it a filename, it should print an
>error message... but instead it either uses part of your last command
>as a filename, or else crashes Stella. I am in the middle of rewriting
>the command parser, but won't finish until late tonight or tomorrow,
>so we're posting this binary with the old parser.
>I haven't had time to respond to all the discussions about new features,
>but there are some damn good ideas there. When I get back home (am
>currently out of town), I'll go through them in detail...
>Archives (includes files) at
>Unsub & more at

Archives (includes files) at
Unsub & more at

Current Thread