Source for Atari AVG Chip or Suitable Replacement?

Maxstang

Active member
Joined
Apr 12, 2008
Messages
1,236
Reaction score
10
Location
California
Up till recently I could get these from an online source, but now I don't see them any more. I know Clay used to make a replacement that used TTL and I'm pretty sure Mark did too. Does anyone know if there are are any more sources for replacement? Thanks!
 
Up till recently I could get these from an online source, but now I don't see them any more. I know Clay used to make a replacement that used TTL and I'm pretty sure Mark did too. Does anyone know if there are are any more sources for replacement? Thanks!

Is Mark not selling them anymore?
 
Hmmm... not Clay's but one I'd seen elsewhere and forgotten about. Anyone have one of these?
 
Hmmm... not Clay's but one I'd seen elsewhere and forgotten about. Anyone have one of these?

Jrok was going to do one, I just don't know that he ever did. He has this on his site.

Next up is the Atari AVG in a Xilinx XC95108, this is a freeware project that I'll put all the details and source/schematics. Expect that soon.... ish..

Edit: I was set straight on the boards below.

Arcadeshop's is Omar's.

avg.jpg


And this one is Clay's.

AVG_new_t.JPG


Edit Again: Here is a prototype of Mark's

AVG.jpg
 
Last edited:
I got mine from Arcadeshop a few months ago. Warning that there's doesn't work on Space Duel. I waited an extra month for a Space Duel fix and it still didn't work. I ended up putting it in my Star Wars and using it's AVG for my Space Duel.
 
The Xilinx one is Omar's, not Jrok's....

2nd one is Clay's.

3rd picture was one the prototype from the ones i built...

Alex Yeckley makes them now too.

I've got a bunch of them built up, but haven't had time to test, so they're not currently FS.

The schematics for the AVG can be inferred by comparing tempest to the later color vector games. Jed Margolin did a writeup confirming that and included schematics of what exactly needs to go inside, which is why so many people made them... nothing too special about it.
 
Last edited:
The Xilinx one is Omar's, not Jrok's....

2nd one is Clay's.

3rd picture was one the prototype from the ones i built...

Alex Yeckley makes them now too.

I've got a bunch of them built up, but haven't had time to test, so they're not currently FS.

Thanks for clarifying. I will go back and edit my post. It looked like it might have been Jrok's work.
 
I got mine from Arcadeshop a few months ago. Warning that there's doesn't work on Space Duel. I waited an extra month for a Space Duel fix and it still didn't work. I ended up putting it in my Star Wars and using it's AVG for my Space Duel.


Now why would this not work on a space duel?

Dave
 
Now why would this not work on a space duel?

Dave

Not sure but mine was displaying random vectors and resetting. I was told before I ordered that they were having issues with SD and it should be fixed in the new version which is what I got.
 
Not sure but mine was displaying random vectors and resetting. I was told before I ordered that they were having issues with SD and it should be fixed in the new version which is what I got.

And it still does not work? Just wondering because I have a SD board and was going to order one...


Dave
 
And it still does not work? Just wondering because I have a SD board and was going to order one...


Dave

Yea, I got the version with the SD fix and it didn't work. You can always do what I did if you have another AVG to swap. Arcadeshop's AVG actually fixed some small graphic issues in my Star Wars. SD plays fine with the SW AVG.
 
Now why would this not work on a space duel?

The logic in the CPLD is a lot faster than the TTL used on my boards and the original tempest design, so it's more susceptible to glitching on the inputs... seems like some strategically placed caps would clean things up.
 
Avg:

It turns out the RAMs on Space Duel are all over the place timing wise. The most common is a 600 ns part but there are also some 450 ns parts. At one point I had asked Arcadeshop to place a notice on his page concerning Space Duel and the use of the current revision in this game due to timing issues since the LSB was not delayed enough to address all speed grades on Space Duel. To date Steven has not done this and I will have to update my page to reflect this issue.

The originial replacements worked fine with the 450 ns (TI I believe) 2114s but we encountered two instances where there were stray vectors on 600 ns boards. Xilinx changed to an RoS part whose timing was not as advertised (-20 parts) so I provided Arcadeshop with a reprogrammed batch (same chips) which worked on 600 ns parts in my test bench and still work on all other Atari games.

The current rev works great on all games EXCEPT some Space Duels so I would recommend against using them on this game. I am not making anymore of these at this time until I get the time to resolve the issue properly.
 
While this subject is on the top of the list, what are the symptoms of a bad AVG in a SW cabinet? I have a PCB that works for about 5 min then resets could this ne a faulty PCB?
 
Thanks for all the replies guys. Glad to have the SD problem with the Arcadeshop boards out in the open as well as I was going to use this in a Space Duel.
 
Thanks for clarifying. I will go back and edit my post. It looked like it might have been Jrok's work.

It looked similar ;) I used the same socketed PLCC as Omar's. I never made more than 2 which I'm using on my boards, as this was right when Omar did his run of board.

xilinx_avg.jpg


- James
 
Cool! I should have know Omar's was not yours though even though they do look very similar... I really like how you use the negative space when you layout the board. :) I presume that makes it easier to etch since your removing less?

It looked similar ;) I used the same socketed PLCC as Omar's. I never made more than 2 which I'm using on my boards, as this was right when Omar did his run of board.

- James
 
Back
Top Bottom