+ 1 !
- VCF East - April 12 - 14, Infoage Museum, Wall NJ
- VCF South West - June 14 - 16, Davidson-Gundy Alumni Center at University of Texas at Dallas
- VCF West - Aug 2 - 3, Computer History Museum, Mountain View, CA
- VCF Midwest - Sept 7 - 8 2024, Schaumburg, IL
- VCF SoCal - Mid February 2025, Location TBD, Sothern CA
-
Please review our updated Terms and Rules here
Model 2 emulator?
- Thread starter Dustym
- Start date
+ 1 !
lowen
Veteran Member
Everyone,
The community is small, but there are at least 5 people on the board that I know of who have unique 8in media that has not been imaged. I'm sorry if that sounds curmudgeonly, but I'm not wellBy my count at least 95% of the images in pski's model2archive came from me, beit directly or indirectly. Please people image and share! *cough* *sneeze*
I am sending my LS-DOS 6.2 disk to Pete, along with the Model II extender (as I have no need for it, and the Hans03/04 set will do what I need). Kelly did the Xenix disks a long time ago, and scanned the 6000 service manual (the best documentation out there on the last iteration of the Model 12's mainboard). I did two of the three LS-DOS 6.3.1A disk images out there (thanks to Tim Mann for the imaging). Bill Degnan did .IMD's of a number of things a year and half or so ago (thanks, Bill!). You did a number of images, and Pete put together the github. Sounds like a good start to a collaboration!
I had started a project a couple of years ago to reconstruct the 6.3.1A Model II sources, but it's stalled at the moment (tracing and disassembling running code is part of it, and then testing the re-assembly of the resulting code, requires either hardware or a working emulator, and I have had neither to work with until recently). I would like to see a collaboration here, as some of us have different pieces disassembled/patched/ and otherwise reverse-engineered. A side effect could be LS-DOS hard disk booting on the II, including with the Lo-Tech/FreHD and the Hans-01 board. (I also have a project, also stalled at the moment, to work on LDOS 5.3.1 for Max-80.... does anyone have a MaxDOS disk to image (MaxDOS = LS-DOS 6 for Max-80)
Frank is very busy with work, and I respect him enough to not nag him about it. His files are, to the best of my knowledge, 6.3.0 'maybe' (as there was no official 6.3.0 release for the II). The 6.3.1A sources are all Misosys' work (Roy Soltoff), and Frank has the official 6.3.0 sources. So the Model II sources Frank has are probably somewhere between the 6.2.0AC sources and proto-6.3; the date patches may not be implemented in the Model II stuff Frank has, or they might be implemented, I don't know yet, since Frank hasn't yet shared the files with me.
EDIT: Forgot to reply to this tidbit.... The II-series are the only TRS-80's that use Z80 Mode 2 vectored interrupts and that are multithreaded (Z80DMA can be thought of as an execution thread for the floppy......) This causes both the machine and its OSes to be more complex, making emulation even more complex (which is part of the reason MAME is pretty slow in its emulation).
And Frank has already posted volumes on Tandy's fixation of fixing hardware with software.....
That's another thing I won't touch..... along with Windows programming. I have half a chance of understanding the MAME emulator framework, since I run it on Linux. I do have Windows 7 as a VM, and I did purchase Matt Reed's emulators and have them on it, but it's not my preferred environment.The Arcnet process is commented out to a stub, because that board's wacky interfacing to its memory and the 80C26 had so many many problems I gave up (I will deal with it later). Ask Frank about Arcnet boards - I dare you!
No, I think I've read his opinion on them before and I'll not wake up that particular 'sleeping dog.'
(Frank, if you happen to read this, that's said with utmost respect but also with humor......)
Last edited:
lowen
Veteran Member
I'm getting ready to find out, as I've just successfully desoldered the 8021 in my Model II keyboard (I have a 6000 keyboard for my 12, now, and so it's not a big loss if the M2 keyboard isn't working..... and I have the temperature-controlled solder/desolder/hot-air rework station.....) and will be building a breadboard dumper hopefully this week..... The 8021 datasheet page 5 lists the ROM Verify Mode (paragraph "8021 Testing and Debugging") and it seems pretty straightforward. I'll keep everyone informed.
Hans01
Experienced Member
@Iowen
You got the right tools to do the job, but be very careful with the PCB it is of a very bad quality and the soldermask
doesn't work anymore after all those years.
(I had to replace the XR chips on my keyboard, which was a terrible job)
I hope you can read the data and improve Mame. :D
You got the right tools to do the job, but be very careful with the PCB it is of a very bad quality and the soldermask
doesn't work anymore after all those years.
(I had to replace the XR chips on my keyboard, which was a terrible job)
I hope you can read the data and improve Mame. :D
lowen
Veteran Member
The top side doesn't even have solder mask.... and the plated-through holes are pretty fragile, too, so when the chip goes back in it will be a double-sided solder job, since three of the plated-through holes came apart, even with the right tools. But I'm less worried about getting the keyboard back together than in getting the ROM dumped.
Thanks, Hans! I'll report back once I've read it....
lowen
Veteran Member
Pete was able to successfully image the LS-DOS 6.2 disk. As far as I know it is pristine. Here's the catalog, including system and invisible files:
This is what LSI wanted TRSDOS 6.2 to look like for the Model 4; yes, DiskDISK, FED6, LED6, QuickFormatandBackup, etc were all originally slated (by LSI at least) to be in TRSDOS 6.2 for the 4. (I had a text listing one time detailing some of this, and I sent a copy to Tim Mann with the LS-DOS 6.3.1 disks he imaged back in 2001, but I've misplaced the master copy......)
(catalog was taken booted into MAME 0.175 with LS-DOS 6.3.1A, as the 6.2 disk does not completely boot. But, for the curious, here is a screenshot of the boot screen for LS-DOS 6.2):
Thanks, Pete!!
This is what LSI wanted TRSDOS 6.2 to look like for the Model 4; yes, DiskDISK, FED6, LED6, QuickFormatandBackup, etc were all originally slated (by LSI at least) to be in TRSDOS 6.2 for the 4. (I had a text listing one time detailing some of this, and I sent a copy to Tim Mann with the LS-DOS 6.3.1 disks he imaged back in 2001, but I've misplaced the master copy......)
(catalog was taken booted into MAME 0.175 with LS-DOS 6.3.1A, as the 6.2 disk does not completely boot. But, for the curious, here is a screenshot of the boot screen for LS-DOS 6.2):
Thanks, Pete!!
I guess we got lucky with the LS-DOS 6.2 disk because it squealed like a pig when I started imaging, but it did complete successfully. Not sure if I can image it again successfully, but hopefully what we got exactly was what was on the disk. I'm not sure how good the error correction and/or read verification is in ImageDisk.
AaronBrockbank
Member
- Joined
- Nov 22, 2005
- Messages
- 30
Hello all.
I got a little further than Lamar on the boot front, with it asking for, accepting and then reformatting the date - curious how it doesn't erase the date input to EOL. Spat it out there without asking for the time, as shown below:
What's interesting is that the second sector of the boot track is marked as "TRSDOS", but followed by "62Level-AH", which would concur with what lowen said:
LS-DOS 6.3.1A has in the same place (from 10h): "LS-DOS63Level-1A"
Now it will be interesting to see how much of the LS-DOS 6.2 code matches up to what's on this disk.
Aaron
That'd be from the lubricant drying out on the disk sleeve - generally the precursor to an all-out oxide dump, so you're lucky there.pski said:
I got a little further than Lamar on the boot front, with it asking for, accepting and then reformatting the date - curious how it doesn't erase the date input to EOL. Spat it out there without asking for the time, as shown below:
What's interesting is that the second sector of the boot track is marked as "TRSDOS", but followed by "62Level-AH", which would concur with what lowen said:
This is basically a beta port of TRSDOS 6.2 for the Model 4.lowen said:
LS-DOS 6.3.1A has in the same place (from 10h): "LS-DOS63Level-1A"
Now it will be interesting to see how much of the LS-DOS 6.2 code matches up to what's on this disk.
Aaron
Last edited:
lowen
Veteran Member
Once is probably enough; the sector CRCs the WD1791 lays down are pretty good, but not perfect. I ran a few of the programs from the disk in the emulator booted to 6.3.1, and nothing crashed. As I said in the email, though, you might want to run your cleaning disk through and inspect the heads for oxide. Although you can inspect the cookie through the floppy's window, turning the cookie by hand in the sleeve; any shed oxide should be pretty obvious. I've had lots of disks squeal over the years, and most of them were imaged ok (my experience being limited to a Catweasel and a 5.25 reading Model III/4 disks).
Since we have the source for 6.2 beta AC (hopefully not too different from the AH level) this to me is the baseline for a 6.3.1A for Model II source reconstruction if Frank doesn't release the Model II pieces that he has (and there is no guarantee that even if/when he does that it was brought up to 6.3.0, who knows, it might be 6.2.0 Level AH).
Last edited:
The new LS-DOS images are available at https://github.com/pski/model2archive/tree/master/Software/Other/LS-DOS
Hans01
Experienced Member
@Aaron, Iowen, Pete.
Thank you for the new LS-Dos images.
I try to run them in Mame64 Version 1.75, but only the LS-Dos 6.3.1 will boot.
To solve the problem I have setup a new installation of Mame64 / Mess version 1.70.
Now I can run all new images and much more than with version 1.75.
Now I can use the Mess GU, which is much easier to use and has the possibility to write to images without first create them.
Set the speed to 500% and create 2 or more dsdd drives, which makes live a lot easier.
Thank you for the new LS-Dos images.
I try to run them in Mame64 Version 1.75, but only the LS-Dos 6.3.1 will boot.
To solve the problem I have setup a new installation of Mame64 / Mess version 1.70.
Now I can run all new images and much more than with version 1.75.
Now I can use the Mess GU, which is much easier to use and has the possibility to write to images without first create them.
Set the speed to 500% and create 2 or more dsdd drives, which makes live a lot easier.
lowen
Veteran Member
0.170 will boot 6.2.0?
Hans01
Experienced Member
@Iowen
Yes
Yes
Hans01
Experienced Member
Use mess interface version 1.70
Load image
Go to options
Click hard reset
Click soft reset
Load image
Go to options
Click hard reset
Click soft reset
works now, thanks Hans!
lowen
Veteran Member
Hmm, the soft reset makes LS-DOS 6.2 boot with 0.175, too, for me. 0.176 is freshly out, so I'll build that next.....
EDIT: Update -- 0.176 boots 6.2 fine, using a 'soft reset' (F3 in the SDL interface with the keyboard in PARTIAL (UI Enabled) mode). 6.3.1 boots fine, too.
Last edited:
Hans01
Experienced Member
Yes, this works now too.
Mame is getting a little better every time.
'soft reset' is good to know how to use it, thanks.
Mame is getting a little better every time.
'soft reset' is good to know how to use it, thanks.
Does anyone know how we could use different M2 boot roms in mame? The current trs80m2 rom folder(see below) has a bunch of different roms which seem to be the BINs of the M2 roms. Specifically, if I want to use a particular version of the boot rom, how can I tell mame to do that? Thanks.
-rwxr-xr-x@ 1 pski staff 2048 Dec 24 1996 u54.u11
-rwxr-xr-x@ 1 pski staff 2048 Dec 24 1996 cpu_2119.u11
-rwxr-xr-x@ 1 pski staff 2048 Dec 24 1996 cpu_11be.u11
-rwxr-xr-x@ 1 pski staff 2048 Dec 24 1996 8047316.u11
-rwxr-xr-x@ 1 pski staff 2048 Dec 24 1996 8043316.u9
-rwxr-xr-x@ 1 pski staff 2048 Dec 24 1996 8043216.u11
-rwxr-xr-x@ 1 pski staff 2048 Dec 24 1996 u54.u11
-rwxr-xr-x@ 1 pski staff 2048 Dec 24 1996 cpu_2119.u11
-rwxr-xr-x@ 1 pski staff 2048 Dec 24 1996 cpu_11be.u11
-rwxr-xr-x@ 1 pski staff 2048 Dec 24 1996 8047316.u11
-rwxr-xr-x@ 1 pski staff 2048 Dec 24 1996 8043316.u9
-rwxr-xr-x@ 1 pski staff 2048 Dec 24 1996 8043216.u11
gp2000
Experienced Member
I think the -bios option will do it. For instance, -bios u54.u11
Failing that, there's always the glorious hack of copying the ROM image over top of the one it is currently using. It'll complain about the checksums not matching which won't matter. That might be necessary because I think it only allows you to choose a ROM from a fixed list of names.