[Solved] Serial puzzle (was DISKREAD / RAWDUMP)

[dave]

I'm curious whether anyone has experienced any issues using OSI diskread or rawread. I recall using rawread years ago and finding it pretty straightforward. Until now, though, I haven't had a full working system to make disk images with. I recently repaired a B51 (the control electronics that is; it's still making some concerning squeaky sounds, but seems to read disks just fine.) I tried Ed's rawdump utility, then the diskread utility from winOSI, derived from Ed's original.

I have had no problems booting up the C4P I'm able to boot OS65D and play around without crashing or other issues. I'm able to send data from the C4 to the terminal program and from the C4 to the terminal program (Coolterm) as well, from the monitor by modifying the contents of $FC01. However, the rawdump and diskread programs seem to hang up while trying to save track 0. I get the message

READING TRACK 0 ...SAVING

and then a hang.

I'm not seeing any activity on the serial line, either. I thought it may be that the code is assuming a C1 and writing to the wrong address, but I looked at at the source and can't see any red flags. It should detect the C2/4 ACIA, and should set up the normal 16x clock by default.

Before I dig deeper, is there something obvious I'm missing?

Dave

[nama]

I've sent you a copy of DISKREAD which is what I (used to) use.
Mike has an improved version called DUMPVIEW which is...better.

Phil

[nama]

Me again,
Doesn't this sound similar (identical) to the issue I was having with my C2?
If you recall I was able to receive, but using DISKREAD would hang on the beginning of tk0....identical to yours.
If you recall if I wiggled the serial cable I was able to get the dump to flow to the screen, but not through the serial. I did get it working in the end, but It was awhile ago now.

Again, I can't quite recall how I got it working, so it may pay to go back through our email threads on the subject.

Phil

[dave]

I spent some more time with the C4P today:

Since slow transmits worked, but fast transmits didn't, I suspected an issue with the TDRE bit. Sure enough, scoping the chip shows an endless loop waiting on the ACIA:

loop_0.png (47.48 KiB) Viewed 3331 times

D8 (red) is the /CS line. The yellow trace is the serial transmission (no activity).

You can see the endless loop:

LDA $FC00 (AD 00 FC) - Read the ACIA control register
(08) - read $08 ( = CTS high. This inhibits TDRE)
4A, 4A (LSR, LSR) Shift bit 1 (TDRE) into the carry flag. For $08, this clears the carry.
90 F9 (BCC -5) Branches back to control register read.
68 - prefetch of next instruction (PLA), gets discarded.

It turns out, the 505 is pre-wired for modem handshaking, and the CTS line is high unless pulled low by the modem. A couple of jumpers fix DCD and CTS lines:

Screen Shot 2015-03-21 at 8.23.10 PM.png (43.93 KiB) Viewed 3331 times

I cut the default traces, and jumpered DCD and CTS to ground, and finally, the C4P is able to transmit over the serial port Now I'm kicking myself for not looking at the FC00 port in the ROM monitor. Perhaps I would have seen the 08 there as well, and figured this out sooner. I plan to make a separate thread on configuring the 505 board for serial I/O, since the default configuration is not suitable for serial comm with a host PC.

So far I've dumped a C2/4 5.25" DOS 3.3 working disk that has an alternate BEXEC*. It's an OSI brown disk and has a crude command interpreter instead of the usual numbered menu. Not really much of an improvement, but different. I don't know if it does actually have any improvements.

I have a set of C1 disks I plan to dump. I'm aware that there is a set of disks that comes with the WinOSI image, but I hate to waste these, and rather preserve them in case there's some difference.

I will go through whatever 5.25" disks I can salvage next, then move on to the 8" disks.

Dave

[MK14HAK]

Awesome, just quietly I knew YOU could sort it.
Hope you have something interesting other than just DOS disks Dave , please ...

[dave]

Sadly, I so far only have mostly various OSI demo and distribution disks, a few games, and a few really lame programs. I do have some unmarked disks, one of which had OS65D on it, so perhaps I can find a backup of something interesting. But I'm pretty sure my development disks were labeled, and I cannot find any. I found one disk cryptically labeled "TIMDOM" with my handwriting, but it's not readable. I have 3 disks labeled "CC1" - "CC3" which I'm assuming means "Colossal Cave", but those would just be text in a random-access database format. I can't find the (unfinished) game disk, or the utility programs used to edit the database.

I've gone through all the readable 5.25" disks now, and will work on posting them in the next week or two. I will clean up the software section of the archive as well. Now it's time to focus on getting another system (or any CPU/disk interface) working with an 8" setup, so I can rip the 8" disks, which are more interesting.

Perhaps it's time to start working on some new software for OSI?

Dave

[MK14HAK]

Well that's a blow. I was thinking you were sitting on a goldmine of contributed OSI disks etc..
Would like to know what you have on 8" Dave ?
I have a few more programs to contribute soon, utilities mainly.
Maybe some PET or TRS80 BASIC programs could be converted to OSI and there is probably some C4<>C1 conversions still needed...
Mike

[MK14HAK]

Mark Spankus perhaps remains a hope of further contributions ? I know he has Forth and 65U

[dave]

I never did buy much software back in the day--all I had was what I wrote. I haven't given up on building up a software library, though. Perhaps with Stephan's cc65 efforts, and Vince Briel's SBIII, and more people getting their systems working, we will see some more home grown software. I think 2015 will be a good year for OSI.

I do have some good stuff on 8" floppies. For one, the UCSD p-system with, I think, Pascal and FORTRAN. Also another word processor, OS65U, and perhaps more (I think a database?). These were generously contributed by forum member W. Falicoff, who used his specialized machine for his research and writing.

I also have an 8" image of a programming language called Beta-65. I sent the disk to David Gesswein a few years back, and he sent me the disk image. I will have to hunt down a manual, though.

So, sparse on software. The good news is, it's an authentic OSI experience

Dave