• Please review our updated Terms and Rules here

Creating Physical TRS-80 Floppies From Disk Images

Arek,
I've looked at your image and it isn't quite what I expected. There seems to be a lot of jitter in the
image versus what I see for l3d5136.dmk. I'm using the Linux Version of HxC which is at Revision
2601, and I'm currently using Revision 2595. A few weeks ago when I checked I was using the
latest Version.

1. What Revision HxC Software are you using? Is it a Windows Version, or Linux?

The Linux version is at this URL:
svn checkout svn://svn.code.sf.net/p/hxcfloppyemu/code/hxcfloppyemu-code

I've attached a photo of your image with the jitter problem, an also an image that was on my TRS80GP
Emulator, for comparison.

I did get you image to boot, but if I do a DIR command after the second page scrolls, the Emulator locks up.
If you do a DIR and list two pages does your computer also lock up?

Larry
Dear Larry

I am using macOS HxC floppy emu version 2.5.6.6.

Doing tests I also noticed that there is something different with the graphical floppy image.

About disk image that I sent to you - at my place works ok on a real TRS80. Now that I have a little more knowledge, I think I understand the errors in my thinking. Converting floppy files itself is quite a difficult task, because Kryoflux can only record from IPF or KF stream files. Here's my problem - either I don't understand the Kryoflux principle/method of writing to floppy (which is highly likely), or I can't find the recording parameters for DTC in the documentation. In my head, this is how I put it together.

1. I have a DSK file
2. I convert to KF stream (I get 40 RAW files)
3. I run DTC with -w parameter, number of tracks, etc.
4. the DTC records a floppy disk.
5. I read the floppy disk through Kryoflux to IMG
6. I try to convert to HFE - and here is the stop. I get an error reading IMG. Neither the --format=ibm_mfm_dd parameters nor any magic spells I know help.

Basic question - is my thinking or my idea itself wrong? In theory (my theory) I should come full circle. From DSK --> converters --> Kryoflux write floppy --> Kryoflux read floppy --> IMG --> DSK(HFE).

In the end, I should get exactly the same DSK(HFE) file I used at the beginning of the process. Do I understand this correctly?


all the best
Arek
 
Arek,
After your point #2 you have saved a preservation image of the Floppy with the Kryoflux.
Assuming the oxide didn't come off the platter and gum up the floppy drive head(s), you
should make your conversion from the last know good read of the floppy. Keeping it safe
just in case you find that you really need just one more GOOD Read.

Your DTC command should have been something like this:
dtc -d0 -r5 -fMM001/ -i0 -z1 -n+18 -fMM001.img -i4 -l8 -s0 -e39 > yourlog.log

Reading Drive 0 which is 256 bytes per sector, and 18 sectors per track. Sector starts with 1
and tracks are numbered 0 thru 39. You can find all this information by looking at the tracks
in HxC and cursor over the sectors at tracks 0, 1, 2, and 3. REF: Attached Photo's.

The tutorial on the Kryoflux forum points you towards looking at the log file to see how your reads
are progressing.

https://forum.kryoflux.com/viewtopic.php?t=853
REF PART #2.

The proper Command Line should specify the Sector Size, and Sectors per Track:
Code: Select all
dtc -d0 -r5 -fMM001/ -i0 -z1 -n+18 -fMM001.img -i4 -s0 -e39
The Terminal now displays:
00.0 : frev: 39518, drift: 0.092 us, tfer: 187355 B/s, rpm: 297.044
00.0 : band: 3.959 us, 5.942 us, 8.001 us
00.0 : base: 1.981 us [99.443%], band: 3.959 us, 5.942 us, 8.001 us
00.0 : MFM: OK*, trk: 000, sec: 18, *H +12
00.1 : frev: 35157, drift: 4.120 us, tfer: 185962 B/s, rpm: 296.996
00.1 : band: 1.869 us, 3.084 us, 3.974 us
00.1 : base: 1.028 us [97.317%], band: 1.869 us, 3.084 us, 3.974 us
00.1 : MFM: <unformatted>
01.0 : frev: 40180, drift: 0.075 us, tfer: 190441 B/s, rpm: 296.985
01.0 : band: 3.973 us, 5.941 us, 8.007 us
01.0 : base: 1.980 us [99.485%], band: 3.973 us, 5.941 us, 8.007 us
01.0 : MFM: OK*, trk: 001, sec: 18, *H +14
01.1 : frev: 35073, drift: 4.378 us, tfer: 183815 B/s, rpm: 296.976
01.1 : band: 1.885 us, 3.105 us, 3.983 us
01.1 : base: 1.035 us [97.377%], band: 1.885 us, 3.105 us, 3.983 us

To write the stream files which were saved by the Kryoflux DTC command you need to decide if
you are going to use .IMG (which is a .RAW Sector Dump) type image. That is the ONLY
option that I see, unless there has been updates in the Software. (I haven't kept up with the
current version, so I may be incorrect on the current choices.) Earlier versions of Disk-analyse
was able to create the .IPF file, but not now. It could be because there are 18 256 byte sectors
on the LDOS531 image.

So, it looks as if the .IMG file is the only way to get a floppy written by using Kryoflux.
If you want to test with the IMG file I created I can post it on PCloud, and you can download.


Larry
 

Attachments

  • Track0-Side0-Sector1.png
    Track0-Side0-Sector1.png
    237.8 KB · Views: 2
  • Track0-Side0-Sector18.png
    Track0-Side0-Sector18.png
    243.3 KB · Views: 2
Last edited:
Dear Larry

I am very impressed with your knowledge and excellent ability to share knowledge. I read everything you write with great respect and gratitude.

I can see that I was quite close, but unfortunately "quite close" is too far from "exactly the right place" there too. One of the missing parameters was -n+18. I was completely at a loss as to how to apply it. I didn't link it to what I noticed when I analysed the images. My readout images had 9 sectors per track. The original had 18. I erroneously assumed that an inadequate conversion was to blame.

In addition, I did not use -r5 and -I8

I have learned a great deal, a lot has been made clearer to me. Let me now process everything you have written with my mind, I want to practice step by step on a real TRS80 machine having a laptop and Kryoflux kit next to it. I'm very excited to see what comes out of this. It's going to take me some time, between work and home duties, and my museum :)

Again, accept my gratitude for your very detailed descriptions and respect for your knowledge.

I will report back on the successes (hopefully there will be no failures :)

PS. I would be very happy to download your image as a template. It's always better to have something to refer to and compare.

all the best
Arek
 
Back
Top