RE: [stella] RIOT RAM test

Subject: RE: [stella] RIOT RAM test
From: Glenn Saunders <cybpunks2@xxxxxxxxxxxxx>
Date: Thu, 28 Nov 2002 03:49:44 -0800
If you attempt to "fry" the 2600 (6 switcher or 4 swithcer) then we can see if the RAM contents come through completely random or if there is a fixed number of possible patterns.

I suspect the latter, otherwise Frying effects would not be as easy to generate on games like Space Invaders.

It would be nice to catalog which bytes of RAM frying alters and to which values to create different effects in particular games. Then we can understand why these values have these effects (like the double missiles in Space Invaders).

At 02:34 AM 11/27/2002 -0600, you wrote:
Okay, I just ran this puppy and here are my results:

1) 2600 heavy sixer (with channel selection switch):

Cycling power would sometimes cause the numbers to change, although the
first two rows would always be:


Removing the power adapter was a more surefire way to get different results
on power-up.

2) 2600 Jr. (long rainbow):

All 00's!  Power cycled (including removing power completely) 20 times, no

3) 7800 (modified with Dev. OS):

Always the same results, power cycling (and removing power completely) made
no difference.

first two lines:


last two lines:


A 17 line display is also what I was witness to.  I have several unmolested
7800s handy that I can also test on, but I don't seem to have a 7800 power
supply in sight.  If anyone feels that will be useful, I'll dig one up.


> -----Original Message-----
> From: owner-stella@xxxxxxxxxxx [mailto:owner-stella@xxxxxxxxxxx]On
> Behalf Of Matthias Domin
> Sent: Tuesday, November 26, 2002 3:19 PM
> To: stella@xxxxxxxxxxx
> Subject: Re: [stella] RIOT RAM test
> Hello!
> Dennis Debro schrieb:
> >
> > Eckhard Stolberg wrote:
> > >When you do this it would be nice, if you could run a little
> > >RAM test program too. I'm curious to know how random the
> > >RIOT RAM really is at power-up.
> >
> > This is a simple program that will output the RAM values to the
> screen. Of course I had to use RAM to get this done. But it does
> at least give an idea of the suspected ramdomness of RAM at power
> up. Nothing fancy here just your typical 6 digit display. I was
> playing around with PAL/NTSC colors and scan lines so the right
> difficulty switch will change from PAL (A) and NTSC (B).
> >
> > I've run this on 3 emulators and they seem to work fine. I'm
> not doing anything illegal so it should work on the VCS as well.
> The idea is to burn this to an EPROM to see what Stella outputs
> at power up.
> On my Junior modell it shows always zeroes (except the $EE area were
> the variables are stored).
> 20 lines running in PAL:
> The last two lines are (as far as i can decipher them):
> 00 00 03 EC 05 00
> F7 00 F7 18 F7 60
> Only 17 lines are visible when running in NTSC on my PAL-TV.
> Regards
> --
> Matthias Domin
> Atari Lynx page:
> Atari Jaguar page:
> Atari VCS 2600 page:
> ------------------------------------------------------------------
> ----------------------------
> Archives (includes files) at
> Unsub & more at

Archives (includes files) at
Unsub & more at

Archives (includes files) at
Unsub & more at

Current Thread