• Please review our updated Terms and Rules here

Atari 1050

JammyJules said:
ok - Now I don't think it is the reset circuit. I made up a lead with a 10k resistor. Connected the RES on the 6507 to 5v, and then grounded the pin for a second. Didn't change anything. Still no activity on address or datalines from the 6507.

Onwards and upwards as they say ..
Click to expand...
Hmmm
 
I have a development. I removed the WD2793A-PL02 (Floppy disc controller) and the 6532 (RIOT PLA?), and now I see activity on the following pins of the 6507 :

D0 to D7 - activity
A12 - activity
A9 to A11 - nothing (low)
A0 to A8 activity.

If I put either IC back in, it goes back to 6507 not running - so I guess I need to trace around and see if I can work out a link.
 
MiaM said:
Related tangent:
Note that it's possible to do some simple testing of an Atari 8-bit disk drive even without a bootable DOS disk.
I.E. it's possible to enter a short basic program that formats a disk and that reads/writes raw sectors of a formatted disk.

Also if you haven't got a bootable DOS disk, you might anyway want to build a simple interface between a modern computers serial port (or FTDI style USB TTL serial) and Atari SIO. I would build that interface with a switch that allows you to have it either emulate a SIO device (I.E. emulate a disk drive) using a modern computer, or act as a SIO controller (I.E. talk to a disk drive) using a modern computer.
This will both allow you to use your 800 even while your 1050 is broken, but also allow you to try to talk to the 1050 from a modern PC, I.E. you wouldn't need to type in commands over and over after each power cycle.
(this will of course not be of much use if the CPU in the 1050 doesn't even run code correctly, but still).
Click to expand...
This is a great shout. I didnt know this was a thing. I have ordered the bits to make one. I should be able to 3D print an SIO socket. Thanks for that.
 
JammyJules said:
I have a development. I removed the WD2793A-PL02 (Floppy disc controller) and the 6532 (RIOT PLA?), and now I see activity on the following pins of the 6507 :

D0 to D7 - activity
A12 - activity
A9 to A11 - nothing (low)
A0 to A8 activity.

If I put either IC back in, it goes back to 6507 not running - so I guess I need to trace around and see if I can work out a link.
Click to expand...
Maybe it's a RIOT or Disk controller issue. Try this: Remove one keep the other in, check, repeat. Also try the thing MiaM said.
 
JammyJules said:
I have a development. I removed the WD2793A-PL02 (Floppy disc controller) and the 6532 (RIOT PLA?), and now I see activity on the following pins of the 6507 :

D0 to D7 - activity
A12 - activity
A9 to A11 - nothing (low)
A0 to A8 activity.

If I put either IC back in, it goes back to 6507 not running - so I guess I need to trace around and see if I can work out a link.
Click to expand...
I don't know if the 6507 has some sort of "halt" input.
A regular 6502 doesn't and thus if there is no activity on the lower address line (and IIRC a signal that might be called sync or something similar?) and the signals aren't grounded then the CPU has ran into one of the illegal instructions that crashes it.

Do you se any short burst of signal immediately after a reset? Might be wroth temporary adding a reset button to easily repeat that. If so then the processor actually runs code but crashes.

I haven't had a look at the schematics (my 1050 seems to work fine and I'm not super familiar with Atari hardware), but I would assume that the 6532 is the only RAM in the 1050. Without it I would think that any reads from the RAM area reads back as $FF. At least in theory there might actually be some problem with the ROM or CPU that causes a crash if it reads something else than $FF from what is supposed to be RAM. Not super likely but if you have a way to check the ROM content (eprom programmer or for that sake anything else where the ROM fits) it might be worth checking that.

I wonder if a ROMulator can be used with a 6507 with some DIY adapter? If so that would be a great way both to inject test software and to replace the RAM in the 6532.

It's at least fortunate that the 1050 uses a 6532 rather than a 6530 (which also contains ROMs and thus the chips are application specific).

Worth knowing is that at least some of the chips/ROMs used in home computers and video games back in the days didn't have the same quality level as chips that went into generic electronic devices, and thus the chips might have gone bad just from aging badly. Don't know who made the chips for Atari at this time, but slightly surprising MOS (I.E. Commodore) produced a bunch of ROM for Atari earlier on, and the MOS ROMs are known to not have the best quality. It was probably a matter of what the customer paid for, I.E. the customer were probably offered different pricing dependent on where they would set the rejection level for wafers that seem marginal. This is just a guess, but seems likely.
 
MiaM said:
I don't know if the 6507 has some sort of "halt" input.
A regular 6502 doesn't and thus if there is no activity on the lower address line (and IIRC a signal that might be called sync or something similar?) and the signals aren't grounded then the CPU has ran into one of the illegal instructions that crashes it.

Do you se any short burst of signal immediately after a reset? Might be wroth temporary adding a reset button to easily repeat that. If so then the processor actually runs code but crashes.

I haven't had a look at the schematics (my 1050 seems to work fine and I'm not super familiar with Atari hardware), but I would assume that the 6532 is the only RAM in the 1050. Without it I would think that any reads from the RAM area reads back as $FF. At least in theory there might actually be some problem with the ROM or CPU that causes a crash if it reads something else than $FF from what is supposed to be RAM. Not super likely but if you have a way to check the ROM content (eprom programmer or for that sake anything else where the ROM fits) it might be worth checking that.

I wonder if a ROMulator can be used with a 6507 with some DIY adapter? If so that would be a great way both to inject test software and to replace the RAM in the 6532.

It's at least fortunate that the 1050 uses a 6532 rather than a 6530 (which also contains ROMs and thus the chips are application specific).

Worth knowing is that at least some of the chips/ROMs used in home computers and video games back in the days didn't have the same quality level as chips that went into generic electronic devices, and thus the chips might have gone bad just from aging badly. Don't know who made the chips for Atari at this time, but slightly surprising MOS (I.E. Commodore) produced a bunch of ROM for Atari earlier on, and the MOS ROMs are known to not have the best quality. It was probably a matter of what the customer paid for, I.E. the customer were probably offered different pricing dependent on where they would set the rejection level for wafers that seem marginal. This is just a guess, but seems likely.
Click to expand...
Do you think it could be RIOT or floppy controller? He had activity on the data bus after removing those ICs. Also, are you an expert or someone who can at least tell if a PCB would work? I need to see if my homebrew would work, and I don't want to order it and solder just for it to not work. If you can tell if a PCB would work, I can DM you my PCB design.
 
xp0nentz said:
Do you think it could be RIOT or floppy controller? He had activity on the data bus after removing those ICs. Also, are you an expert or someone who can at least tell if a PCB would work? I need to see if my homebrew would work, and I don't want to order it and solder just for it to not work. If you can tell if a PCB would work, I can DM you my PCB design.
Click to expand...
I wouldn't call myself an expert, and that would be a bit too much for me to take on.
What are you designing? Maybe start a separate thread about your design?
 
MiaM said:
I wouldn't call myself an expert, and that would be a bit too much for me to take on.
What are you designing? Maybe start a separate thread about your design?
Click to expand...
It's an 8052 based SBC, you don't have to be an expert, I just need to know if my PCB would work. I can DM it to you.
 
Last edited:
MiaM said:
I don't know if the 6507 has some sort of "halt" input.
A regular 6502 doesn't and thus if there is no activity on the lower address line (and IIRC a signal that might be called sync or something similar?) and the signals aren't grounded then the CPU has ran into one of the illegal instructions that crashes it.

Do you se any short burst of signal immediately after a reset? Might be wroth temporary adding a reset button to easily repeat that. If so then the processor actually runs code but crashes.

I haven't had a look at the schematics (my 1050 seems to work fine and I'm not super familiar with Atari hardware), but I would assume that the 6532 is the only RAM in the 1050. Without it I would think that any reads from the RAM area reads back as $FF. At least in theory there might actually be some problem with the ROM or CPU that causes a crash if it reads something else than $FF from what is supposed to be RAM. Not super likely but if you have a way to check the ROM content (eprom programmer or for that sake anything else where the ROM fits) it might be worth checking that.

I wonder if a ROMulator can be used with a 6507 with some DIY adapter? If so that would be a great way both to inject test software and to replace the RAM in the 6532.

It's at least fortunate that the 1050 uses a 6532 rather than a 6530 (which also contains ROMs and thus the chips are application specific).

Worth knowing is that at least some of the chips/ROMs used in home computers and video games back in the days didn't have the same quality level as chips that went into generic electronic devices, and thus the chips might have gone bad just from aging badly. Don't know who made the chips for Atari at this time, but slightly surprising MOS (I.E. Commodore) produced a bunch of ROM for Atari earlier on, and the MOS ROMs are known to not have the best quality. It was probably a matter of what the customer paid for, I.E. the customer were probably offered different pricing dependent on where they would set the rejection level for wafers that seem marginal. This is just a guess, but seems likely.
Click to expand...
I have checked the rom by inserting it into my t48, loading Atari 1050 vA.rom and doing a verify. It passed, and then as a check, if failed against Atari 1050 vB.rom. Roms were downloaded from here :

http://www.atari8.us/Atari 8-bit Computers/Collections/Utilities/Roms/Roms/Disk Drive Roms/

When it's crashing, I am not sure if I see a "short burst" or not, as it must be extremely short, and all the addresss and data lines go high.

The romulator is a great shout - I used one to fix my PET and didn't think about using it for this. I will look at building an adaptor and give that a try. I have also ordered another RIOT as they are cheap enough. If it's the floppy controller it gets a bit expensive.
 
JammyJules said:
I have checked the rom by inserting it into my t48, loading Atari 1050 vA.rom and doing a verify. It passed, and then as a check, if failed against Atari 1050 vB.rom. Roms were downloaded from here :

http://www.atari8.us/Atari 8-bit Computers/Collections/Utilities/Roms/Roms/Disk Drive Roms/

When it's crashing, I am not sure if I see a "short burst" or not, as it must be extremely short, and all the addresss and data lines go high.

The romulator is a great shout - I used one to fix my PET and didn't think about using it for this. I will look at building an adaptor and give that a try. I have also ordered another RIOT as they are cheap enough. If it's the floppy controller it gets a bit expensive.
Click to expand...
Good, we're somewhat narrowing it down. So just to make sure, one ROM works and one doesn't?
 
xp0nentz said:
Good, we're somewhat narrowing it down. So just to make sure, one ROM works and one doesn't?
Click to expand...
One rom verified in the ready, the other didnt (it was a few bytes different).

The rom data starts at 0x800 - with FF all before that. I haven't been able to check if that is correct yet.
 
JammyJules said:
One rom verified in the ready, the other didnt (it was a few bytes different).

The rom data starts at 0x800 - with FF all before that. I haven't been able to check if that is correct yet.
Click to expand...
Scope the rom pins, maybe that might make a development.
 
JammyJules said:
I understand you are trying to help - and I appreciate that, but I get the feeling that you are not reading my posts, and I have to repeat myself
Click to expand...
Oh, sorry about that. Which ROM starts at 0x800? Is it the one that isn't working quite right or the one that is verified. Also sorry if I can't read that much of your posts, I'm at school from 8 am to 3 pm, and I often read latest posts in school to see if anything happens. By time I get home, I forget most. So apologies if I can't read all your posts. If you can, please fill me in on what the latest developments are.
 
Torch said:
Am I the only one getting vibes of a Talking Eliza chatbot? :p
Click to expand...
I'm not an AI. I get why you got that though. It does look like I'm looking for keywords inside of messages then just replying to that. Also, tomorrow I won't be able to reply for the day, I'm transferring my breadboard design to a PERF board. Good day!
 
Last edited:
Torch said:
Am I the only one getting vibes of a Talking Eliza chatbot? :p
Click to expand...
I don't understand why this person is bloating this thread. I came here for advice, this person's first comment was "don't listen to what I say" and has since been asking me to repeat myself over and over again. Even when i told them that, they asked me to do it again. I have used this forum previously to fix my CBM PET, and had fantastic advice. I feel like I am being spammed lol.

I don't think I will post on this thread again.
 
Oh I just found out you can ignore a member. I will continue to post as it maybe useful to others.
 
Sorry for bloating the thread, I won't post here again if it helps. I hope you succeed.
 
Well I have some progress.

I swapped out the riot chip and now the head tries to Initialise (or doesn't - not sure!)

No spinning when a disk is inserted though. I did a continuity check on the switch by the eject are and it seems ok. Not sure about the positions I found with the attached plugs. They do not seem to reflect what is on the pcb.
 

Attachments

You must log in or register to reply here.
Back
Top