• Please review our updated Terms and Rules here

Transferring text files from Kaypro II to Windows machine.

The CP/M version 2.2f and beyond are for the Kaypro /84 models. Your "Kaypro II CP/M 2.2" is correct for what you have. There is also a "Kaypro IV CP/M 2.2", but I don't think the single/double sided is handled by the CP/M version, only the ROM version.
 
I take that back. In spite of the Kaypro Technical document saying:
1783739559946.png
I have a floppy image that calls itself "2.2F" ("KAYPRO 10 CP/M Version 2.2F") and is for the Kaypro 10, which is a /84 model as far as hardware is concerned. This floppy does not boot on a Kaypro II (or IV), at least in my simulator. So, I think there's a mistake in the Kaypro Technical manual that's been propagated to other sources. Or else Kaypro made two different "2.2f" versions, but the one for the /83 models was only ever called "2.2f" internally.
 
Probably another rabbit hole, but interesting, nonetheless. The Kaypro CP/M master diskettes (R/O since they are without cut outs) are, as far as I know, numbered in series; but there is no way of associating that number with the serial number of the machine. The current practice of selling machines without any of the diskettes that were sold with them begs for trouble which is far more serious than having to source your own keyboard or keyboard cable. The machine I acquired this year seems to have had little use; yet someone had had it to pieces. In the process he had lost quite a few UNC screws (not common animals in the EU) for the lid and holding the mainboard in place, and had exchanged the A: and B: drives. While the seller separately offered me some diskettes, I have no reason to believe that they were associated with the machine as originally manufactured, and although the machine was described as in top condition, the seller was unaware that the keyboard was unable to communicate with the mainboard on account of several dead chips around the SIO that I had to identify and replace.

All of which is a long way of asking whether the CP/M master diskettes that I have are correct for the machine, or whether they stand in the way of using PIP to transfer even ascii files successfully over an otherwise working serial connection.
 
DavidWien said:
All of which is a long way of asking whether the CP/M master diskettes that I have are correct for the machine, or whether they stand in the way of using PIP to transfer even ascii files successfully over an otherwise working serial connection.
Click to expand...
I seriously doubt you have the wrong software for your ROM/mainboard. I've used the disk images you referenced in my simulator, and I have a high degree of confidence that (at least in this regard) the behavior of my simulator is closely aligned with real hardware.

I forget if you said you had read the ROM and compared with "known good" images, but if not then that's one possibility (that the ROM has faded - but that's only possible if the ROM is an actual EPROM (with the "erase window") and not a "mask programmed" ROM. Another possibility is that there is something very unusual wrong with the Z80-SIO (again, I forget if you tried swapping that out, in which case nevermind). Or else there is something weird going on with Tera Term such that ^Z is not actually getting sent to the Kaypro (assuming all the other characters ARE getting sent).
 
I took a quick look at the source code for Tera Term, at least that which is posted on github. I don't see any obvious mishandling of ^Z, but unfortunately a large number of comments (both in the code and in git commits) are in Japanese (which I don't read) and so I don't really have any confidence in that assessment. This is just throwing stones, but my concern would be that something was done to accommodate international keyboards or character sets that has affected ^Z. That's probably a stretch, but seems at least remotely possible. We'd have to see the results of the "PIP KAYPRO.TXT=RDR:", RESET, "SAVE 40 KAYPRO.TXT", "DUMP KAYPRO.TXT" experiment to know whether Tera Term is sending something else in place of ^Z - or PIP is receiving any characters at all.

In your Tera Term / TERM.COM experiment, I guess the one thing not tested was whether you could type ^Z in Tera Term and that was received by TERM.COM. This control character is the Clear Screen code for the Kaypro, so if TERM.COM did receive it (and doesn't make it visible like printing "^Z") it should clear the screen on your Kaypro. That would be proof that the ^Z was making it all the way through.
 
Another data point. Looking at a disassembled copy of TERM.COM I see that it directly accesses the Z80-SIO, instead of making calls to CP/M (ultimately landing in the ROM). So, if the ROM was faulty, TERM.COM would work but PIP (RDR: ) would not.
 
durgadas311 said:
Another data point. Looking at a disassembled copy of TERM.COM I see that it directly accesses the Z80-SIO, instead of making calls to CP/M (ultimately landing in the ROM). So, if the ROM was faulty, TERM.COM would work but PIP (RDR: ) would not.
Click to expand...
It is extremely hot here without A/C, but I will try your earlier suggestion later. I am not sure which ROM chip you are referring to. As I understand it, the character ROM is only responsible for putting the images on the screen and not for actually clearing the screen.

Are these EPROMs (with the "erase window") or "mask programmed" ROMs? I have no idea how to "read" them.

L1010425_Char-ROM.jpg


I dont feel confident with TeraTerm. What alternative program should I try to acquire for the pc?


L1010425_Mon-ROM.jpg
 
I'm referring to the 81-232A. It does appear to be an EPROM (UV-erasable). While the label looks to be factory and sufficient to block ambient light, it's always possible that something else has compromised it - or that the initial programming was sub-standard. These ROMs are effectively programmed by "charging capacitors" and so they won't retain programming forever. But I'm not sure that 40 years is the limit either. It all depends on conditions, and you did mention that there were some bad chips and so possibly this machine has been subjected to more severe conditions than normal. Of course, it is possible that a PROM programmer won't read the ROM under the exact same timing as the Kaypro, and thus might not show the same symptoms. But let's start by assuming a read of it returns what the Kaypro sees.
 
Oh, you don't have access to a PROM programmer to read them? It is possible to get a copy into a file on disk (I did this when I first got my 2X), although not "easy". But, then it's a matter of comparing that file to a known good image, so we're back to transferring files to/from the Kaypro. Perhaps there's a way using the Gotek?
 
durgadas311 said:
Oh, you don't have access to a PROM programmer to read them?
Click to expand...
You are kidding: that would be another rabbit hole!
durgadas311 said:
It is possible to get a copy into a file on disk (I did this when I first got my 2X), although not "easy". But, then it's a matter of comparing that file to a known good image, so we're back to transferring files to/from the Kaypro. Perhaps there's a way using the Gotek?
Click to expand...
It is not a Gotek, but an HXC2011 emulator.

I dont feel confident with TeraTerm. What alternative program should I try to acquire for the pc?
 
DavidWien said:
...
It is not a Gotek, but an HXC2011 emulator.

I dont feel confident with TeraTerm. What alternative program should I try to acquire for the pc?
Click to expand...
I assume the HXC2011 has some external storage, like a USB thumb drive, and you could add/extract images from it. It would be a matter of extracting the image where you saved the file and then using tools on the PC to extract the file. Not sure how that's done, though.

You could use Kermit95 instead of Tera Term, but it's more setup commands, as mentioned by https://forum.vcfed.org/index.php?t...ro-ii-to-windows-machine.1257371/post-1509581
 
A quick test you could do now, that would tell us whether the ROM is the suspect or Tera Term, is to use DUMP (not TYPE) to examine the SAVE 40 file. Use ^S/^Q to pause/resume until you can see the memory at offset 1ECA and see if any of your text was saved. If you see your text there, then that points to Tera Term's handling of ^Z. If no text is seen, then it points to the ROM not doing the receive correctly (possibly corrupted/faded).
 
DavidWien said:
Doug, This may say something to you. Also this.

We also havent tried this.

D
Click to expand...
It looks like PCGET/PCPUT go directly to the Z80-SIO(s), so if your ROM is corrupted it would bypass that. However, you still need something on the PC side that talks XMODEM in order to make it work. You'd also need to make sure and use the "modem" port option, as the /83 models don't have a serial printer port (second SIO). Plus, you're still stuck behind the "bootstrap" procedure which has you running PIP (on RDR: / TTY: ).

Not sure what those other links will do for you.
 
Just as supplemental information, since it was mentioned earlier, here is a transcript of the procedure to capture a copy of the ROM image on a Kaypro II/IV (/83) - using only what's available on the distribution disk. The one missing piece is how you can verify this image against the official one. Since we'd be looking for as little as one byte difference, it probably won't work to compare "by hand" (eye). But, this transcript contains the full dump of the "known good" image, so you could compare that against a capture of the DUMP output (since you said you can use ^P to capture console output via the printer port).

Out of "healthy paranoia" I take extra steps along the way in this example, to check my work. Those are not strictly necessary, but are helpful to ensure you get a true image of the ROM.
 

Attachments

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