ATR-8000 (and other) CP/M System disks here

[a8isa1]

@sup8pdct

 

James, could you suggest a method to take Don Maslins CP/M disk collection (or a subset thereof) and convert them to 256bytes per sector images so that they can then be converted to ATRs?

 

I assume there are CP/M emulators but perhaps there's an easier method.

 

-SteveS

[ricortes]

Every looks great except the program MAKESYS.COM seems to hang. I don't know what it's supposed to do so perhaps I'm not using it correctly.

 

-SteveS

Typo? There is no MAKESYS.COM on the disk.

 

Just a heads up, if you are using the AUTOTERM boot disk, you are not seeing anything past column 40 with DIR and some of the programs. It doesn't wrap, just disappears. It makes it a little hard to follow prompts in some programs. Makes me think we should be running on the RS232 connection until a suitable solution is found. TYPE works fine.

[a8isa1]

Typo? There is no MAKESYS.COM on the disk.

 

Just a heads up, if you are using the AUTOTERM boot disk, you are not seeing anything past column 40 with DIR and some of the programs. It doesn't wrap, just disappears. It makes it a little hard to follow prompts in some programs. Makes me think we should be running on the RS232 connection until a suitable solution is found. TYPE works fine.

I'm using Claus Buchholz's DT-80 image (which he posted somewhere in these forums) on an Atarimax cartridge, instead of Autoterm. That left 40 columns/right 40 columns feature of Autoterm was always too annoying.

 

-SteveS

Edited May 23, 2014 by a8isa1

[a8isa1]

Just a heads up, if you are using the AUTOTERM boot disk, you are not seeing anything past column 40 with DIR and some of the programs. It doesn't wrap, just disappears. It makes it a little hard to follow prompts in some programs.

With Autoterm you can toggle back in forth between the left and rightmost 40 columns by typing CTRL+DEL.

 

-SteveS

[sup8pdct]

That is good. only 1 mistake on a manually assembled cpm data disk. An easy fix.

Now just to write up some instructions and do up a zip file.

There are 6 disks in total in the archive.

1 is broken as only 1/2 of the data area is there, but all boot data is there.

one is an older version of the of the CPM bios.

one has nothing at all.

the other 2 have source code.

It will take time to extract the files however.

 

James

[ricortes]

I'm using Claus Buchholz's DT-80 image (which he posted somewhere in these forums) on an Atarimax cartridge, instead of Autoterm. That left 40 columns/right 40 columns feature of Autoterm was always too annoying.

 

-SteveS

Another DOH! I was leaning on the space bar for 50 characters to get the windows to switch. thx for the heads up.

[AtariGeezer]

I have hand assembled this CPM disk with some files on it. No idea if this will work.

 

Just disk copy the atr to a real floppy on a DD drive and make sure it is single sided . Wipe the disk before you format and write this out.

 

Am hoping the CPM files will load and work ok.

They are DDSYSGEN for the latest version only (27-april-1984).

DDINIT to format disks

SYSTEM.SWP that DDSYSGEN can use if asked to.

 

Feedback please.

 

James

James,

 

Which DD drive and Disk Copy program are you using to copy the ATR image to?

 

Thanks,

Jay

[sup8pdct]

@sup8pdct

 

James, could you suggest a method to take Don Maslins CP/M disk collection (or a subset thereof) and convert them to 256bytes per sector images so that they can then be converted to ATRs?

 

I assume there are CP/M emulators but perhaps there's an easier method.

 

-SteveS

I am doing it by hand using WINHEX.

 

The first disk, i just used the files that were zipped separately in that archive. The others tho Are more problematic. The FDI images i have found from the start of the directory till the end of the file have the sectors stored in this order 1,3,5,2,4 @ $400 per sector.

22NICE was a bit hit and miss. Some files extracted ok. Some didn't, depending on the block size used and if sectors are in the right order.

 

What i am doing is to create a blank SS/DD atr in an emulator. Load it and an IndusGT cpm disk to see where things are. Load each file and copy and paste to the correct position on the blank ATR file and update the directory. I also created an excel spread sheet to keep track of block numbers used, position in file etc.

 

A specially written program would be easier but it would need to be tailored to each disk.

Don't even bother with ATR8K484.tdo The whole 2nd side isn't there.

Have found read.me in atr8k382 has the high bit set random places through out the file. Don't know about the rest of the disk tho.

 

James

[sup8pdct]

James,

 

Which DD drive and Disk Copy program are you using to copy the ATR image to?

 

Thanks,

Jay

I'm not. I have to rely on others to test what i am doing as my ATR8000 isn't working properly. A real community effort

Any SS/DD drive should be ok as well as any disk copy program that copies all sectors and can handle 256 byte sectors.

Using the mech on on the ATR8000 would be the best to use.

 

James

[sup8pdct]

I had better step through the disks on Don Maslins CP/M disk collection.

 

ATR8K382
is the old version 03-1982. this has problems with different format disks, namely single v double sided. Some files of the same name are in the image below but are only meant for this version. Don't ever mix them.


ATR8K484
is seriously damaged. only good part is the system tracks, tho CPM starts at $D200 compared to the default which is $D400

ATR8KATM
so damaged that the disk is blank.

ATR8KBIO
has 2 files.
AUTOBIOS.MAC Is the source file of the Bios of the 04-1984 version. It does a .include for DEBLOCK.MAC as part of its assembly. I have this as a text file.
DEBLOCK.MAC is the source file of deblocking routines for 03-1982 version.Someone has done a oops here. Also have this as a text file.

ATR8KMDM
has 2 files
SERIAL.MAC. source for serial routines for modem37? in z80 opcodes. I have this as a text file
MODEM37.MAC. source for modem37 in 8080 opcodes. I also have this as a text file.

ATR8000
All files are on the atr image i made. System tracks created by the ATR.EXE file i made. Loads at $D400 (default)

 

 

Does anyone want any of the other disks done so that they can be loaded onto the CPM box? I can see no reason to do so apart from the early version for hysterical value.

 

James

 

[AtariGeezer]

I'm not. I have to rely on others to test what i am doing as my ATR8000 isn't working properly. A real community effort

Any SS/DD drive should be ok as well as any disk copy program that copies all sectors and can handle 256 byte sectors.

Using the mech on on the ATR8000 would be the best to use.

 

James

Gotcha, what kind of problems are you having with your ATR8000?

 

My status update: I found my cable that connects the ATR8000 to a ST SF314 and connected it all up. Turned on the 800 and the drive came alive so I turned off the SF314 and booted Mydos 3.18 off the SIO2SD. After loading I disabled D1 and turned the SF314 back on and tried to access D1 on the ATR8000, the drive was still responding. Then I tried to format a SSDD disk in the SF314, but that didn't work even after 5 times. After that the SF314 was no longer working as far as booting or trying to format again. If you press the reset button on the ATR8000, the SF314 comes on for a bit like it should, but that's all...

 

I'm thinking I either have a bad power supply or an intermittent chip in it... More trouble shooting tomorrow...

[ricortes]

Gotcha, what kind of problems are you having with your ATR8000?

 

My status update: I found my cable that connects the ATR8000 to a ST SF314 and connected it all up. Turned on the 800 and the drive came alive so I turned off the SF314 and booted Mydos 3.18 off the SIO2SD. After loading I disabled D1 and turned the SF314 back on and tried to access D1 on the ATR8000, the drive was still responding. Then I tried to format a SSDD disk in the SF314, but that didn't work even after 5 times. After that the SF314 was no longer working as far as booting or trying to format again. If you press the reset button on the ATR8000, the SF314 comes on for a bit like it should, but that's all...

 

I'm thinking I either have a bad power supply or an intermittent chip in it... More trouble shooting tomorrow...

 

Are you much of a hardware guy? Reason I ask is because I spent a couple of days staring at the schematics and poking around with a DVM and I'm at the 90% confidence level for '"What is wrong with our ATR8k's."

 

Not to say there aren't the normal problems with dirty connectors; the edge connectors on mine look like someone coated them with candle wax.

 

That 4069 on the SIO bus is a flawed design. I don't mean just from choice of parts because it could be made right, implementation is a problem. I'm getting a little old so short term memory and eye sight has decayed a bit. Looking at the schematic I was thinking the 4069 would be powered up with the inputs floating<horrible design choice if I am right> so I popped out the old DVM and started poking around. I was thinking there were a couple of unused inverters in the 4069 and it could be I missed them/used somewhere else on the schematic so I checked them out. Found .55 VAC/1.29 VDC on them, pin #3 & #11. This is a pretty clear indication they are not terminated and in a race condition. I doubled check with the power off, just looking at resistance at 2 MOhms and nothing => they be floating. I almost brought out the diagonal snipers right there.

 

Brief synopsis for anybody not familiar with it, I've seen this type of race conditions going like a free running oscillator at ~14 MHz. They spike the hell out of an entire system and draw more current then would otherwise be necessary. You can't just do something like add a cap because they will just spike the system harder. Only fix is to terminate the inputs properly. This problem will be there no matter what you do. If you unplug the SIO2PC, if you unplug the Atari, if you switch to running the ATR8k as an RS232 CPM box. The glitch is always on when the ATR8k is on.

 

I can't be sure this is what is causing the system to flake out whenever something else is hooked to the SIO bus although it is a leading suspect. I'm saying it probably should be fixed no matter what. Even when it is just sitting there with nothing but the Atari hooked to it, mine occasionally goes nuts on its own. Just suddenly starts doing the head load and seek thing for no reason. It's too bad that sucker is soldered directly to the mother board. I am ok with modifications on mine but I worry some of you may want to keep them original.

 

Just a little link for anyone that wants to see how well known of a design problem it is.

http://www.interfacebus.com/IC_Output_Input_Pullup_Resistor_Values.html

[sup8pdct]

Gotcha, what kind of problems are you having with your ATR8000?

 

 

Problem 1. It won't format a disk, yet it can read and write sectors. I suspect the FDC.

Problem 2.After a very short time with CP/M running, it refuses to accept text input or mirrors it back. Once an app is running that outputs only, it runs fine. Maybe the CTC is acting up?

 

James

[AtariGeezer]

 

Are you much of a hardware guy? Reason I ask is because I spent a couple of days staring at the schematics and poking around with a DVM and I'm at the 90% confidence level for '"What is wrong with our ATR8k's."

 

Not to say there aren't the normal problems with dirty connectors; the edge connectors on mine look like someone coated them with candle wax.

Yep, I'm an electronics technician by trade. Got interested by the age of 5, built many kits by the age of 18. In high school I took 4 years of basic electronics and 2 years of vocational electronics. While in the Air Force I worked on Microwave Intrusion systems, Microwave and Satellite Radio Communications, plus Air Traffic Control Communications and systems. In civilian life I've worked on HF & VHF radios at Datron and many components in Laser systems at Cymer.

 

So I know my way around circuit boards and test equipment

 

I'm close to 55 now and my eye sight has diminished as well, I use a combo of reading glasses and bifocals plus a magnifying lens to work on stuff now

 

Hopefully it will be an easy fix as it worked 100% 20 years ago when I boxed it up and moved back to California...

[AtariGeezer]

 

Problem 1. It won't format a disk, yet it can read and write sectors. I suspect the FDC.

Problem 2.After a very short time with CP/M running, it refuses to accept text input or mirrors it back. Once an app is running that outputs only, it runs fine. Maybe the CTC is acting up?

 

James

Ahhh, ok. I think my FDC should be ok, shouldn't have died from just sitting in a box? We will see what's up...

[a8isa1]

I am doing it by hand using WINHEX.

 

The first disk, i just used the files that were zipped separately in that archive. The others tho Are more problematic. The FDI images i have found from the start of the directory till the end of the file have the sectors stored in this order 1,3,5,2,4 @ $400 per sector.

22NICE was a bit hit and miss. Some files extracted ok. Some didn't, depending on the block size used and if sectors are in the right order.

 

What i am doing is to create a blank SS/DD atr in an emulator. Load it and an IndusGT cpm disk to see where things are. Load each file and copy and paste to the correct position on the blank ATR file and update the directory. I also created an excel spread sheet to keep track of block numbers used, position in file etc.

 

A specially written program would be easier but it would need to be tailored to each disk.

Don't even bother with ATR8K484.tdo The whole 2nd side isn't there.

Have found read.me in atr8k382 has the high bit set random places through out the file. Don't know about the rest of the disk tho.

 

James

Oops wrong CP/M disk collection. Don Maslin's stuck in my mind.

 

For giggles I sometimes get the urge to fire up random CP/M apps. I don't know why. I just do.

 

Unfortunately these days I don't have any 5.25" drives on PCs with which to generate disks so I was wondering if there's an easy way to convert CP/M images to ATRs and use the Atari to make the CP/M disks. They don't need to be bootable. I have two each of SSDD, DSDD, and DSHD 5.25" drives.

 

BTW, thanks for having a look at the ATR8000 images in Don's collection. Would have left me scratching my head if I thought I needed them and couldn't find a way to use them. Perhap it actually did drive me crazy 20+ years ago.

 

-SteveS

[a8isa1]

 

Are you much of a hardware guy? Reason I ask is because I spent a couple of days staring at the schematics and poking around with a DVM and I'm at the 90% confidence level for '"What is wrong with our ATR8k's."

 

Not to say there aren't the normal problems with dirty connectors; the edge connectors on mine look like someone coated them with candle wax.

 

That 4069 on the SIO bus is a flawed design. I don't mean just from choice of parts because it could be made right, implementation is a problem. I'm getting a little old so short term memory and eye sight has decayed a bit. Looking at the schematic I was thinking the 4069 would be powered up with the inputs floating<horrible design choice if I am right> so I popped out the old DVM and started poking around. I was thinking there were a couple of unused inverters in the 4069 and it could be I missed them/used somewhere else on the schematic so I checked them out. Found .55 VAC/1.29 VDC on them, pin #3 & #11. This is a pretty clear indication they are not terminated and in a race condition. I doubled check with the power off, just looking at resistance at 2 MOhms and nothing => they be floating. I almost brought out the diagonal snipers right there.

 

Brief synopsis for anybody not familiar with it, I've seen this type of race conditions going like a free running oscillator at ~14 MHz. They spike the hell out of an entire system and draw more current then would otherwise be necessary. You can't just do something like add a cap because they will just spike the system harder. Only fix is to terminate the inputs properly. This problem will be there no matter what you do. If you unplug the SIO2PC, if you unplug the Atari, if you switch to running the ATR8k as an RS232 CPM box. The glitch is always on when the ATR8k is on.

 

I can't be sure this is what is causing the system to flake out whenever something else is hooked to the SIO bus although it is a leading suspect. I'm saying it probably should be fixed no matter what. Even when it is just sitting there with nothing but the Atari hooked to it, mine occasionally goes nuts on its own. Just suddenly starts doing the head load and seek thing for no reason. It's too bad that sucker is soldered directly to the mother board. I am ok with modifications on mine but I worry some of you may want to keep them original.

 

Just a little link for anyone that wants to see how well known of a design problem it is.

http://www.interfacebus.com/IC_Output_Input_Pullup_Resistor_Values.html

Wow! Finally a good explanation. Over my head but you explaned it well enough.

 

Now I don't have to wave a chicken over my head when I want to format disks.

 

Seriously though, only by plenty if trial and error I learned that the best way to get anything accomplished on the ATR8000 is to use it all by its lonesome, at least when I can do that.

 

-SteveS

[ricortes]

Yep, I'm an electronics technician by trade. Got interested by the age of 5, built many kits by the age of 18. In high school I took 4 years of basic electronics and 2 years of vocational electronics. While in the Air Force I worked on Microwave Intrusion systems, Microwave and Satellite Radio Communications, plus Air Traffic Control Communications and systems. In civilian life I've worked on HF & VHF radios at Datron and many components in Laser systems at Cymer.

 

So I know my way around circuit boards and test equipment

 

I'm close to 55 now and my eye sight has diminished as well, I use a combo of reading glasses and bifocals plus a magnifying lens to work on stuff now

 

Hopefully it will be an easy fix as it worked 100% 20 years ago when I boxed it up and moved back to California...

That's great news. I'm pure hobbyist so it is good to have people around who have made their bones.

 

Have you looked at the ATR8k schematics? There is a ~4 meg file floating around labeled ATR8000.ZIP<like 2 other unrelated files> that has the schematics drawn by C. Calvo. The more I poke around, the more sure I am in over my head.

 

For one thing, in addition to the 4069, there appears there is at least one other chip, a 74ls16, with the same problem. Less of a problem I should say since it isn't CMOS and is open collector. Makes me suspect they didn't do it anywhere.

 

That got me on the focused on the problem of termination/pull ups/downs. If you look at the schematic, practically nothing going to the outside world is terminated. By that I mean if you don't have a disk drive or printer plugged in, all the I/O is just floating!

 

I decided to take a look at the FDD connector since it is our bread and butter behind the SIO plug. Looks like AC on many of the pins. I know it isn't done on the IBM side with drives, but shouldn't there be a terminating resistor plug like they make for SCSI for unused inputs when you have no drives attached? Same hold trues for the printer port. Looks OK to me until you don't have a printer plugged in, then everything floats. I'm contrasting this to anti noise hardware on the Atari for everything from SIO buss to joysticks. I'm starting to believe the successful running of an ATR depends on how many florescent lights you have and distance to the nearest radio station.

[sup8pdct]

Oops wrong CP/M disk collection. Don Maslin's stuck in my mind.

 

Unfortunately these days I don't have any 5.25" drives on PCs with which to generate disks so I was wondering if there's an easy way to convert CP/M images to ATRs and use the Atari to make the CP/M disks. They don't need to be bootable. I have two each of SSDD, DSDD, and DSHD 5.25" drives.

 

-SteveS

It would be possible. A program will need to be written. Another way is if you have and INDUSGT with ramcharger and CP/M, you can transfer them over using the indus CP/M software onto a disk. The ATR8000 cpm should be able to read it.

One thing to consider with a custom written program would to invert the data being transferred.

 

James

[sup8pdct]

Here is a ZIP archive of files etc that i have created that may interest people and includes files and instructions on how to create a bootable CPM disk on/for the ATR8000. I have included quite a bit. except for the ATR8000 manual. I would like the CPM manual for the ATR8000 but noone has scanned it yet.

 

James

ATR8000 boot.zip

[AtariGeezer]

That's great news. I'm pure hobbyist so it is good to have people around who have made their bones.

 

Have you looked at the ATR8k schematics? There is a ~4 meg file floating around labeled ATR8000.ZIP<like 2 other unrelated files> that has the schematics drawn by C. Calvo. The more I poke around, the more sure I am in over my head.

 

For one thing, in addition to the 4069, there appears there is at least one other chip, a 74ls16, with the same problem. Less of a problem I should say since it isn't CMOS and is open collector. Makes me suspect they didn't do it anywhere.

 

That got me on the focused on the problem of termination/pull ups/downs. If you look at the schematic, practically nothing going to the outside world is terminated. By that I mean if you don't have a disk drive or printer plugged in, all the I/O is just floating!

 

I decided to take a look at the FDD connector since it is our bread and butter behind the SIO plug. Looks like AC on many of the pins. I know it isn't done on the IBM side with drives, but shouldn't there be a terminating resistor plug like they make for SCSI for unused inputs when you have no drives attached? Same hold trues for the printer port. Looks OK to me until you don't have a printer plugged in, then everything floats. I'm contrasting this to anti noise hardware on the Atari for everything from SIO buss to joysticks. I'm starting to believe the successful running of an ATR depends on how many florescent lights you have and distance to the nearest radio station.

I took a quick glance at the PDF schematics I can see that mine is a bit different. Mine has a MC14069 for U10 (Not U9 as in the schematics) and U9 is something different, so I'll have to print those out and draw in the differences to see what's different.

 

The SIO lines on the 810's have a resistor in series then going to a 3086.

 

The power supply in the ATR could use another diode bridge rectifier before going to the 5 volt regulators. As it is you have 8 VAC going straight to them, not a good design...

 

You would think there would be an optional termination plug on the floppy connector if no standard drives were connected, but the makers at SWP were probably thinking that anyone who bought one would be using it for CP/M or to have more storage space other than a 90k 810

[ricortes]

I took a quick glance at the PDF schematics I can see that mine is a bit different. Mine has a MC14069 for U10 (Not U9 as in the schematics) and U9 is something different, so I'll have to print those out and draw in the differences to see what's different.

 

The SIO lines on the 810's have a resistor in series then going to a 3086.

 

The power supply in the ATR could use another diode bridge rectifier before going to the 5 volt regulators. As it is you have 8 VAC going straight to them, not a good design...

 

You would think there would be an optional termination plug on the floppy connector if no standard drives were connected, but the makers at SWP were probably thinking that anyone who bought one would be using it for CP/M or to have more storage space other than a 90k 810

 

Looked at the power supply schematics based on your observations. Had an OMG! Moment too. The 12V supply is robust, just isn't used for anything other then RS232. Maybe they had the MS DOS/Hard Drive in mind or an internal floppy. Odd design choices.

 

I have a 4069 in U9, but on my MB it just goes to U12 without passing U10 or U11 and collecting $200. U22 is AWOL, just a bunch of holes in the MB. On the bright side, it does look like there is room/provision to mount a 34 pin IDE type connector next to the 34 contact edge connector. I may make that change just to get better connections.

 

My ROM socket is 28 pins <ibid> but near as I can tell, no option for using a bigger ROM. They have the option to use 2716 or 2732, nothing else.

 

On the 4069 I am still asking why on a number of levels. I mean I can see where using the 74ls16 because of its open collector output. Makes interfacing to the Atari simple. Question becomes why use a 4069 when a pin compatible 74LS14 could have been used. I mean if you wanted to square up inputs, prevent false triggering, go with a Schmitt trigger. If you wanted it in CMOS, they even make a 40106 which is same pin out as 4069, 74ls14, but with Schmitt trigger input.

 

I'm probably going to use the extra inverters on the 4069 tied to the lines with LEDs just to get an idea of what is going on. It will look pretty.

 

I keep trying to follow the logic of the CTC in the bit bang serial scheme. They seem to have thrown a handful of chips at a problem that could have been done with a single UART. I've noticed the CTC in my system is getting pretty hot. I may start substituting low power chips where I can.

Edited May 26, 2014 by ricortes

[AtariGeezer]

 

Looked at the power supply schematics based on your observations. Had an OMG! Moment too. The 12V supply is robust, just isn't used for anything other then RS232. Maybe they had the MS DOS/Hard Drive in mind or an internal floppy. Odd design choices.

 

I have a 4069 in U9, but on my MB it just goes to U12 without passing U10 or U11 and collecting $200. U22 is AWOL, just a bunch of holes in the MB. On the bright side, it does look like there is room/provision to mount a 34 pin IDE type connector next to the 34 contact edge connector. I may make that change just to get better connections.

 

My ROM socket is 28 pins <ibid> but near as I can tell, no option for using a bigger ROM. They have the option to use 2716 or 2732, nothing else.

 

On the 4069 I am still asking why on a number of levels. I mean I can see where using the 74ls16 because of its open collector output. Makes interfacing to the Atari simple. Question becomes why use a 4069 when a pin compatible 74LS14 could have been used. I mean if you wanted to square up inputs, prevent false triggering, go with a Schmitt trigger. If you wanted it in CMOS, they even make a 40106 which is same pin out as 4069, 74ls14, but with Schmitt trigger input.

 

I'm probably going to use the extra inverters on the 4069 tied to the lines with LEDs just to get an idea of what is going on. It will look pretty.

 

I keep trying to follow the logic of the CTC in the bit bang serial scheme. They seem to have thrown a handful of chips at a problem that could have been done with a single UART. I've noticed the CTC in my system is getting pretty hot. I may start substituting low power chips where I can.

The reason why they used a 4069 may have just been a cost factor back then?, otherwise I couldn't tell ya why.

There's a 74LS367AN in u9 , u11 and u12 goes to the ram chips on mine.

 

My ATR doesn't have the 34 pin mounting holes, wish it did

Looks like a lot of difference in revisions...

 

I'll have to look at the CTC circuit...

[AtariGeezer]

After looking closer, I have a Rev-A board...

[sup8pdct]

 

Looked at the power supply schematics based on your observations. Had an OMG! Moment too. The 12V supply is robust, just isn't used for anything other then RS232. Maybe they had the MS DOS/Hard Drive in mind or an internal floppy. Odd design choices.

 

12V is also on the FDC.

 

Check out the schematics in my archive. I have fixed them to make them match mine. Hope i got all of the errors.

 

James