40 replies to this topic

[Anthony John Morley]

Hi All.

I'm new to the forum but not so new to the Atari 2600 after having grown up in the 80's with one and a whole host of classic games.

I recently borrowed an Atari 2600 (Vader) off my uncle who kept his daughters Atari in the loft for years. It works fine, plays any game you put in it unless they're dead! (Obviuosly)

My problem is, I got myself an Atari CX2600 Woody a couple of days ago. It works fine on most games but will not, regardless of cleaning, jiggling cartidges, etc, play 4 or 5 games that play fine in the Vader model.

Has anyone else come across this and is there anything that can be done to make these games, (that do work fine) work in this model of the Atari?

A shame if not as it's in great condition, been well looked after and is completely boxed, a very good condition box too.

Any advice welcomed.

Cheers

Edited by Anthony John Morley, Thu Apr 19, 2012 5:15 AM.

[Pioneer4x4]

Are they activision carts?  They have thinner boards and can be more hassle to get clean.  Since the 2600 does work, keep trying reinserting the carts, that will clear off the contacts as well.
What actual games are they?

[Anthony John Morley]

I read about the Activision issue but I don't think that applies. Only River Raid is Activision of the ones that wont play in this machine. 3 different PacMac carts wont play but they all play in the Vader. 2 Centipedes wont play, again they both work in another machine. There are a few of others, Beserk and Dodg'em are 2 of them.

It plays every other game I have, Frogger, Frostbite(Activision), DigDug, Enduro, Pole Position and a whole load of others. All 100% working games. Only Enduro and Centipede need a bit of a wiggle to get them working in the Vader. All the rest work fine.

I have cleaned the new console contacts and the carts but to no avail. I'm bewildered by it.

Edited by Anthony John Morley, Thu Apr 19, 2012 8:13 AM.

[SlowCoder]

How did you clean the console's contacts?

I like to take a cart and wet its contacts with high percentage alcohol, and insert it in the console about 20 times.  Let the cart and console dry completely and try again.  It's possible the carts that don't work have a pin that the other ones don't, and just happens to not be working on the console.

Oh, yeah ... welcome to AA!

[Anthony John Morley]

I use Isopropyl to clean games. I bought it to clean some old Sega and Nintendo games and have used it on all my Atari games. Actually brought 3 or 4 back to life with it.

To clean the contacts I used a piece of cardboard cut to the came size as the board in a cartridge and dampened with Isopropyl gently inserted into the console slot. I read about this method for Atari and other cosoles to clean the contacts. It's never failed me. I let it dry for a good hour before trying it.

It seems very strange that games like Combat and PacMan wont work it this machine when they are the 2 most sold with the consoles. I understand some games are loose and tight fitting and need a jiggle here and there to get them to work but no matter what with these certain games will they load up. Most of the time it's a blank screen and every now and then I get the bars you get if you turn on the Atari with no cart inserted.

It's a shame with it being all boxed and complete, serial nos on the box matching the console and everything inside included. I was hoping this would be a keeper.

Looks like I'll have to put this one down to experience and pass it on. It's a working machine but doesn't like some of the games I do!

*EDIT* Thank you for the welcome

Edited by Anthony John Morley, Thu Apr 19, 2012 10:00 AM.

[Rex Dart]

I wonder if all that cart jiggling might've broken some solder joints on the cart slot?

[atari2atari]

Anthony John Morley, on Thu Apr 19, 2012 9:55 AM, said:

Most of the time it's a blank screen and every now and then I get the bars you get if you turn on the Atari with no cart inserted.

This sort of "visual response" sounds like there is potential for one of the three critical integrated circuits - - the TIA, 6507 processor, or RIOT chip - - to have some "bit rot".

I have observed several situations where a console will work just fine for some games, but throws errors on others.  The errors often manifest as "bad boots" with rolling lines, or glitches and freezing in the game.

Replacing the TIA or RIOT has resolved the problem.

I attribute it to "bit rot" - - the slow degradation of integrated circuits.  Static electricity can also cause damage!

Rex Dart, on Thu Apr 19, 2012 9:56 AM, said:

I wonder if all that cart jiggling might've broken some solder joints on the cart slot?

This also sounds reasonable to me - - different carts can use slightly different cartridge pins to access the integrated circuits of the machine, and so having some carts that work and others that don't could be caused by this, too.


Time to open her up and get to work on the motherboard!

-a2a

[Anthony John Morley]

I have just been inside, had a good clean of the dead spiders and fluff. The motherboard was actually a lot better looking than I expected for its age. I checked all the circuit solder joints and everything seems to be fine (although I'm no Electrician!)

All the card slot joints are fine. I couldn't get into the actual slot part because I think you need to remove the soldered joints to take it off? But looking at the contact inside with a torch they all look fine. There is very little space between the and the all spring quite nicely.

So it's more than likely a chip issue, which I know nothing about. I never was any good at soldering either.

[mimo]

if the chips are in sockets, try pushing them down, or carefully remove/re-insert them

[Anthony John Morley]

This is how it looks inside. I tired pushing the chips down tight/removing them and trying. No joy on the unworkables.

Where would I purchase replacements and how much do they go for? Or would getting another Atari for spares be worth a try/

[jferio]

Hmm... that's not a Woody you have... that's a six switch, either a heavy or a light.

You are, at least, in luck in that the chips are all socketed. I'd say buy another system to use for "spares". If the chips in the donor aren't socketed, it just takes some care with a soldering iron and desoldering braid to remove them.

[Syntaxerror999]

jferio, on Thu Apr 19, 2012 12:48 PM, said:

Hmm... that's not a Woody you have... that's a six switch, either a heavy or a light.

You are, at least, in luck in that the chips are all socketed. I'd say buy another system to use for "spares". If the chips in the donor aren't socketed, it just takes some care with a soldering iron and desoldering braid to remove them.

you could order them from best electronics. Things like this are the reason I assembled my diagnostic kit (field service manual, shorting plugs, diagnostic test cart, and the STC cart)

Edited by Syntaxerror999, Thu Apr 19, 2012 1:13 PM.

[Anthony John Morley]

I looked at Best Electronics. I will order new chips for it this week and see how it goes with it. Hopefully it will fix it. If that fails I'll order a whole new board for it. I'm not confident enough with a soldering iron to remove and replace parts.

[BigO]

I'm having trouble seeing a scenario in which a failure in one of the chips would cause Combat to fail as described while other, presumably larger, trickier games would still work. I would have expected the opposite if it's a bad data or address line or "rotted bit". I'd still be stuck on aggressively cleaning the cartridges and/or slot.

Having said that, I'd also readily experiment with swapping chips if I had spares available.

[SlowCoder]

BigO, on Sat Apr 21, 2012 7:16 AM, said:

I'm having trouble seeing a scenario in which a failure in one of the chips would cause Combat to fail as described while other, presumably larger, trickier games would still work. I would have expected the opposite if it's a bad data or address line or "rotted bit". I'd still be stuck on aggressively cleaning the cartridges and/or slot.

Having said that, I'd also readily experiment with swapping chips if I had spares available.

I was thinking the same.

[Anthony John Morley]

BigO, on Sat Apr 21, 2012 7:16 AM, said:

I'm having trouble seeing a scenario in which a failure in one of the chips would cause Combat to fail as described while other, presumably larger, trickier games would still work. I would have expected the opposite if it's a bad data or address line or "rotted bit". I'd still be stuck on aggressively cleaning the cartridges and/or slot.

I thought the same. Combat and Pacman are games that you would assume would play straight away but dont. Frostbite, Frogger play fine and they're way advanced in terms of graphics and memory.

The games that don't work in this machine have been cleaned thoroughly. They work flawlessly in other machines. I borrowed my brothers Atari Jnr to test out all my games on this afternoon and they all work on that fine.

Like I said in an earlier post. I'm bewildered by it. But for the sake of £10 for new chips it's worth a try I guess. Or just spend a few extra £'s and get a whole new board for it, ready to go.

I've had a few Atari's over the years and cleaned them up, brought a few back to life from completely not working from cleaning the cart slot. Never experienced one play some games and not others like this.

Any other suggestions welcomed though......

Edited by Anthony John Morley, Sat Apr 21, 2012 10:16 AM.

[Anthony John Morley]

SlowCoder, on Thu Apr 19, 2012 9:21 AM, said:

I like to take a cart and wet its contacts with high percentage alcohol, and insert it in the console about 20 times.  Let the cart and console dry completely and try again.

I tried this method also after reading it and still had the same result with the same games. All the other worked fine as normal.

Edited by Anthony John Morley, Sat Apr 21, 2012 10:20 AM.

[Nukey Shay]

Actually, the Combat game program does do something uncommon (in relation to more-modern games). It reconfigures data direction for one of the RIOT addresses. I can't recall if the Pac-Man game program does anything out of the ordinary, tho.

Anybody recall if some other early and common game resets DDR?

EDIT: Star Wars:Empire Strikes Back does it, too.  So does Cosmic Ark (tho that one is definately less-common than either of the others).

Edited by Nukey Shay, Sat Apr 21, 2012 10:48 AM.

[atari2atari]

Nukey Shay, on Sat Apr 21, 2012 10:37 AM, said:

Actually, the Combat game program does do something uncommon (in relation to more-modern games). It reconfigures data direction for one of the RIOT addresses. I can't recall if the Pac-Man game program does anything out of the ordinary, tho.

Anybody recall if some other early and common game resets DDR?

EDIT: Star Wars:Empire Strikes Back does it, too.

Interesting!  Thanks, Nukey - - your posts always provide some good perspective.

As a side note, the RIOT chip tends to be the one I replace the most on 2600 consoles, followed by the TIA.

-a2a

[Nukey Shay]

The only thing that I notice about the Pac-Man code is that the interrupt vector is set to be zero.  Typically, game programs at the time which did not use BRK interrupts would put an extra copy of the reset vector into that space (as a failsafe?  I dunno).  I have no idea how this would mesh with Combat's problem...if RIOT is indeed the cause.

[Nukey Shay]

Duh! I might have it backwards! Combat uses the interrupt vector space to hold a 2-byte data table. So it could be that the reset vector is not being read properly by the console (in such a case, the game crashing by subsequently trying to execute any of the hundreds of zeros in game code calls up the interrupt vector - rebooting the game correctly if the reset vector also exists in the IRQ).


Simply put...Pac-Man and Combat both don't have a valid executable address in the last 2 bytes of rom.  The same is true of River Raid, Dodge'Em, and Berzerk (all zeros as the IRQ vector). That much, they have in common with each other.

Edited by Nukey Shay, Sat Apr 21, 2012 11:23 AM.

[Anthony John Morley]

OK. I've just been through all the games I have here that work and some borrowed games that work on an Atari Jnr (Long Rainbow) and Black Vader models but not on this machine CX2600U Sixer. (I'm assuming it's the heavy version with the big block of metal inside covering the circuitry?)

Working Carts.....

Venture
Wizard Of Wor
Pole Position (2 carts)
Frogger
Space Invaders (3 carts)
Frostbite
Donkey Kong (2 carts)
Soccer (2 carts - One called Football)
Moon Patrol
Yar's Revenge
Gorf
Missile Command
Space War
Outlaw
Pro Wrestling
Q-Bert
Dig Dug (1985 Cart)
Chopper Command
Enduro

Carts that do not work in this machine but work in other machines.....

Centipede (1982 +1988 Carts)
Dodg'em
Beserk
Warlords
River Raid
PacMan (3 carts all 1981)
Slot Racers
Circus Atari
Combat (CX-2601 and CX-2601-P)
Blackjack
Human Cannonball
Star Master
Spiderman
Ghostbusters
Mousetrap
Dig Dug (1987 Cart)

Hope this helps those that know more than most

Thank you for all the help so far. You're all diamonds

Edited by Anthony John Morley, Sat Apr 21, 2012 11:20 AM.

[Nukey Shay]

I don't have disassemblies of all of those, but the same pattern is emerging for every one I do.  The failing games either use zero or unrelated data in the IRQ vector space...instead of a copy of the reset vector (or a valid BRK interrupt routine) that the working games are using.

Could this be the cause?  Is the console somehow masking the reset vector with bad data causing it to "fall into" it's interrupt and try to boot that way?

[Anthony John Morley]

Nukey Shay, on Sat Apr 21, 2012 11:06 AM, said:

Simply put...Pac-Man and Combat both don't have a valid executable address in the last 2 bytes of rom.  The same is true of River Raid, Dodge'Em, and Berzerk (all zeros as the IRQ vector). That much, they have in common with each other.

Very interesting as the games you mention are on my list of none working in this machine. So if the machine is not reading the carts correctly on load up, is this down to the RIOT chip being faulty?

[Nukey Shay]

I don't know.  I couldn't tell you hardware from a hole in the ground.  But that is the software pattern I'm seeing in every disassembly I have currently on hand.