Centipede trackball control affects horozontal position of the centipede

Look at the screen:
Clearly it's at least one bad 2101.
Replace the pokey... run self test... listen to the beep codes.

From the pics, I'd guessing L7 and N7 are bad (or the 153s that read them out -- but that wouldn't explain the sprite issues)

View attachment 790465
What do you see in the image that indicates a bad 2101? What am I missing? What don't I understand? It must be something obvious with the way you typed "Clearly" with such a tone.

The Pokey can sit in the PCB and rot at this point. mecca already declared Pokey so you're adding nothing new here. Besides, you already said this bug had nothing to do with the Pokey. I started this thread to discuss with this community about a bug that intrigued me. I'll replace the Pokey when I decide to fix the board.

Came into this thread expecting a weird and interesting centipede issue.

Other members of this community are interested about this bug as well. We could fix the board but what fun would that be? That's been done at least 1000 before here. The community has spoken, they demand interesting bugs! Our avatars will be forever cast in shame if we do not oblige at this point. Focus man! They may eat us alive!

Beep codes... You want the beep codes from a board with dead audio... My gnomes have been working tirelessly to put my lab together but they are a long ways off still. Let me see what I can cobble together for you.

I'll post a new testing video for everyone tomorrow. I would be appreciative of your input.
 
I suggested POKEY, yes. I was the subject of insults and ridicule because I don't know how to fix anything according to some engineering wonk whose relevance in actually doing anything apparently expired in the year 2001. I don't specialize in 1 or 2 things, I have to fix any and all games in arcades (ie. not talk trash on the internet from my mom's basement or cubicle) and I'm actually very good at what I do. for those wondering though, the rare video game breaks I've deal with the last 10 days were vectors, hence why I suggested POKEY. I don't obsess over these items exclusively to tear other people down.

the reasoning for why I said POKEY is because it's a common issue on these games when you combine the possibilities of knuckleheads prying chips out one sided and the dodgy socket quality on certain Atari games, you might have bad connections in the chain somewhere. this isn't a "omg single wipe sockets are bad" recommendation, I'm just saying properly remove the chip and ensure the legs don't have 45 degree bends and that they're all straight and uniform and try again. if you have an equivalent part on hand, do the same straightening procedure and try that in place of the suspected bad one (obviously use a known good part). if that doesn't get it then you escalate to socket replacement if you can solder. if none of these measures work then you farm it out to someone else to fix for you. all of these items apply to any socketed chips on the board, if you have program-related strangeness then I would suggest doing the roms.

every day though it's heartbreaking knowing that if I could just have a 220 IQ, an engineering degree from MIT, the ability to read schematics, and virtually no social skills I could've picked up more chicks throughout my life like Mark Spaeth if I just knew how all the bits talk to each other.
 
  • Like
Reactions: r3v
I suggested POKEY, yes. I was the subject of insults and ridicule because I don't know how to fix anything according to some engineering wonk whose relevance in actually doing anything apparently expired in the year 2001. I don't specialize in 1 or 2 things, I have to fix any and all games in arcades (ie. not talk trash on the internet from my mom's basement or cubicle) and I'm actually very good at what I do. for those wondering though, the rare video game breaks I've deal with the last 10 days were vectors, hence why I suggested POKEY. I don't obsess over these items exclusively to tear other people down.

the reasoning for why I said POKEY is because it's a common issue on these games when you combine the possibilities of knuckleheads prying chips out one sided and the dodgy socket quality on certain Atari games, you might have bad connections in the chain somewhere. this isn't a "omg single wipe sockets are bad" recommendation, I'm just saying properly remove the chip and ensure the legs don't have 45 degree bends and that they're all straight and uniform and try again. if you have an equivalent part on hand, do the same straightening procedure and try that in place of the suspected bad one (obviously use a known good part). if that doesn't get it then you escalate to socket replacement if you can solder. if none of these measures work then you farm it out to someone else to fix for you. all of these items apply to any socketed chips on the board, if you have program-related strangeness then I would suggest doing the roms.

every day though it's heartbreaking knowing that if I could just have a 220 IQ, an engineering degree from MIT, the ability to read schematics, and virtually no social skills I could've picked up more chicks throughout my life like Mark Spaeth if I just knew how all the bits talk to each other.
As always mecha I'm appreciative of your input. Historical data like yours and yaryar's is very important. I'm sure you've accumulated quite a bit over the years. As an example, knowing something like the failure rates and failure modes of specific ICs isn't going to be printed in any schematic. I'll recheck the Pokey as you suggested when the time comes.

Don't be so hard on yourself mecha. I wasn't given the MIT package in life either. Sometimes people get so smart they forget how to tie their shoelaces...
You don't want that man. You would be tripping all over yourself. It's just not a good image.
 
As always mecha I'm appreciative of your input. Historical data like yours and yaryar's is very important. I'm sure you've accumulated quite a bit over the years. As an example, knowing something like the failure rates and failure modes of specific ICs isn't going to be printed in any schematic. I'll recheck the Pokey as you suggested when the time comes.

Don't be so hard on yourself mecha. I wasn't given the MIT package in life either. Sometimes people get so smart they forget how to tie their shoelaces...
You don't want that man. You would be tripping all over yourself. It's just not a good image.
I've had no dealings with @yaryar but I've read many great things. @gamefixer is legendary also but typically busy with the job. our occupation is different. lol
 
I've read that replacing the existing ball with a new slightly "more white" ball or replacing the CPO can fix this issue.

Thanks,
Jason
 
I've read that replacing the existing ball with a new slightly "more white" ball or replacing the CPO can fix this issue.

Thanks,
Jason
When I got my centipede it had an 8 ball for a track ball and it still worked. So I would lean more towards the cpo
 
What do you see in the image that indicates a bad 2101? What am I missing? What don't I understand? It must be something obvious with the way you typed "Clearly" with such a tone.
It is pretty fucking obvious to anyone who's ever played centipede.

The Pokey can sit in the PCB and rot at this point. mecca already declared Pokey so you're adding nothing new here. Besides, you already said this bug had nothing to do with the Pokey. I started this thread to discuss with this community about a bug that intrigued me. I'll replace the Pokey when I decide to fix the board.
The problem you posted about has nothing to do with the pokey, kiddo.

The lack of sound and the bad PRNG are a different problem -- the pokey.

You clearly have no interest in actually learning anything and think it's useful to attack me, so I'm done with this thread.

(And from later evidence, there aren't going to be any error beeps anyway, and the problem is further downstream from the RAMs).
 
Last edited:
I suggested POKEY, yes. I was the subject of insults and ridicule because I don't know how to fix anything according to some engineering wonk whose relevance in actually doing anything apparently expired in the year 2001.
That's 2008... when I left school, got a real job, and didn't need to fix other people's toys to pay the mortgage.
My employer would be shocked to learn that I don't do engineering work.

the reasoning for why I said POKEY is because it's a common issue on these games when you combine the possibilities of knuckleheads prying chips out one sided and the dodgy socket quality on certain Atari games, you might have bad connections in the chain somewhere.
Your reasoning is that you couldn't even bother to look at the schematics and see that the pokey has nothing to do with the trackball.
You don't care about actually providing useful information, just the appearance of it to people who don't know any better.

every day though it's heartbreaking knowing that if I could just have a 220 IQ,
You could start with clawing your way up to triple digits.
Look! I can make sophomoric insults too!
 
It is pretty fucking obvious to anyone who's ever played centipede.


The problem you posted about has nothing to do with the pokey, kiddo.

The lack of sound and the bad PRNG are a different problem -- the pokey.

You clearly have no interest in actually learning anything and think it's useful to attack me, so I'm done with this thread.

(And from later evidence, there aren't going to be any error beeps anyway, and the problem is further downstream from the RAMs).
Hudson this thread is about analyzing and understanding a bug not fixing a board. At this point you've added absolutely zero towards that discussion. I can tell by your comments you didn't even take the time to watch the video. Therefor you must have no real interest in understanding what's going on here. If you're not going to contribute towards that discussion it probably would be best if you remained silent.

You're complaining about how you're being talked to? Didn't anyone ever teach you the golden rule? Do unto others as you would have them do unto you. Let me break that down for you in terms you can better understand. Your input affects my output. You must've been asleep that day in class.

You've messed up here Hudson...

You've personally attacked people I hold in high regard in a public forum.

You've tried to squelch other people's voices through intimidation. That type of stuff kills discussion forums. People become fearful of saying something stupid and being lambasted for it so they don't contribute at all. Example: You expect a board with timing issues to do a propper RAM test?! What foolishness!!! Durp.

You came to this thread to teach while swigging around like big D Mcgee. We're all here to discuss and learn about a bug.

I threw down the gauntlet and started this challenge and you threw your hat into the ring. Your words are public record Hudson. This bug is going to make you look like a fool for typing out of your ass, not me. I just gave you enough rope. Now go clean your keyboard. ffs

You underestimated me.
 
You clearly have no interest in actually learning anything and think it's useful to attack me, so I'm done with this thread.

Don't let the door hit you in the ass on the way out.

It's hilarious to see you whining about people attacking you, when all you do is insult people who you perceive as being less knowledgeable than you.
 
  • Like
Reactions: r3v
The Pokey can sit in the PCB and rot at this point. mecca already declared Pokey so you're adding nothing new here. Besides, you already said this bug had nothing to do with the Pokey. I started this thread to discuss with this community about a bug that intrigued me. I'll replace the Pokey when I decide to fix
the board.

Beep codes... You want the beep codes from a board with dead audio... My gnomes have been working tirelessly to put my lab together but they are a long ways off still. Let me see what I can cobble together for you.


What is your aversion to replacing the Pokey?

If you have bad RAM, that could look like a lot of other things. These boards are garbage in, garbage out. So it may look like a problem in a given circuit, but you can't be sure of that when that circuit is being fed bad data. So Troubleshooting 101 dictates that you start with the bugs you can fix, and let the board help you where it can.

You want to use the self-test to determine if there is bad RAM. Self-test indicates bad RAM via beep codes, not anything visual. But if your Pokey is bad, you may not hear them. So fix the audio, so you can hear if the board is telling you if you have bad RAM.

Dead audio on these is easy to fix, as it's one of two things: A bad Pokey, or a bad LM324 preamp on the game board. So if you have no audio, you want to start there in this case.
 
You want to use the self-test to determine if there is bad RAM. Self-test indicates bad RAM via beep codes, not anything visual. But if your Pokey is bad, you may not hear them. So fix the audio, so you can hear if the board is telling you if you have bad RAM.

Hey clown -- try to keep up.

He posted a video showing self test getting to the final screen, which (shockingly) means there weren't any beep codes.

1736258403094.png

...which I posted about on Sunday:
(And from later evidence, there aren't going to be any error beeps anyway, and the problem is further downstream from the RAMs).

This is exactly the behavior you attacked ATGW over in another thread a few days ago, but I'm sure you lack the self awareness to understand that.
 
Seeing the character set on the right of the screen, that doesn't appear to be a video ram failure, the repeating pattern is the same across all 4 columns, in order for that to happen 2 or more 2101s would have to fail in the same way, same bits, same everything. Kind of unlikely. Check the multiplexer after the ram for the background. Motion objects appear to be the correct groahics on the clip I saw. All graphics appear correctly drawn, whether they're supposed to be where they are or not, indicating that the graphics roms are fine.

Pic1-3 may be stuck
 
In the second video you will see that all the bugs on the game board can be related to the timing on 256HD.

1) Centipede memory location and the archer memory location are being selected at the wrong time to be read or written to.
2) The wrong characters are being fetched from the ROMs
3) Vertical line 1 is firing early and is being cropped off.
4) We have more black space horizontally to the right compared to the left.
5) The centipede for player 2 in cocktail mode starts in the middle of the screen instead of the 256HD location.
6) Good enough 256H to build H sync and V sync for the monitor to sync but poor 256HD generation. Video is not distorted or crushed.
7) The board thinks it's in cocktail mode as if reading the wrong bit.
8) Some of the dip switch values will occasionally jitter on the test screen (info not previously included in a video or this thread).

I'll gladly go out on a limb here and give you my current theory.

Something is dirty with the oscillator circuit. Maybe Q1 or the crystal itself. A dirty good enough clock will affect higher counts yet still be capable of providing V sync and H sync.

My second guess would be The IC at location M4 (LS175) generating 256HD is faulty or something within the circuit that generates it.

We also get a RAM beep of one. However, you can not expect a board to do a proper RAM test with timing issues. A RAM beep of one should be expected from a board suffering from a timing issue such as this.

Why I don't think it's RAM.
1) No random characters or fluctuating changing ones.
2) No glitches limited to a specific area on the screen.

We will figure this out gentlemen. The board will reveal all eventually. This is an interesting one for sure.

Thank you to everyone who has contributed technically. This is what I came here for.

I'll be doing further testing and post another video this weekend.

I would love to hear everyone's opinion of my theory in the meantime.
 
I would love to hear everyone's opinion of my theory in the meantime.
Kiddo, your'e a dumb clown, don't know what you are talking about and you probably hang the toilet roll with the end coming off the back.

Also, it doesn't help when you perfectly explain your thought process to everyone without using a green lantern decoder ring.

That should cover it so the experts won't have to waste time they could be using to belittle others to chime in here.

Hope that helps.
:cool::cool:;);)
 
You're overthinking this.

Replace the RAM at H2 then see where you're at.

It sounds like something is messing up the data bus. (Or at least the CPU thinks so.) A bad RAM can do this. And if it isn't that, it could be a number of other things.

Also, if you haven't done so already, the first thing you should do is VISUAL INSPECTION. Specifically, look for places on the parts side where tin may have flaked off of one trace, and shorted to an adjacent trace. This is common on these boards.

Look especially under all socketed chips, as it's common for people to use screwdrivers to remove the chips, and scrape the traces underneath, smearing tin from one trace to another. This is a common issue because Atari didn't use solder mask on the parts sides of these boards. Also look for debris around and under the ROMS and RAM, where the bus traces run under the chips, and the bus traces run between chip pins.

Hold the board up to a bright light, putting the board between you and the light, and look THROUGH the board. Especially at the areas under the chips, where the traces run between the legs. This technique will make shorts and other debris a lot more visible. You can get tin whiskers on these that are almost invisible to the naked eye.

There's a process to troubleshooting these, and it's been covered in hundreds of threads here. Start with the simple things, and let the board help you. Don't overcomplicate.
 
dear @HudsonArcade after near 18 years of coexisting with you on here under whatever name you use, I've placed you on ignore. you can stop directing quote posts at me because I'm not going to read them and whatever insight you believe you're contributing, that's probably not really noteworthy or helpful either. have a great day
 
Back
Top Bottom