• Please review our updated Terms and Rules here

Apple II disks / Greaseweazle / Fluxengine

E

ethump

Experienced Member
Joined
Jan 29, 2022
Messages
68
Location
Sheffield, UK
Hi -

I've picked up another Apple II Europlus. If you look back at my posts on here you'll see me battling another Europlus. I still have that one and I intend to use the new one to try find out what's wrong with the old one. Before I do that though I want to have a good play with this new one since I've never really had (a working) one before.

My question here is to do with disks and what I can expect from a Greaseweazle/Fluxengine setup. I was under the impression the Apple II used GCR encoding. When I download disk images like this AppleDOS one, and open it in the Floppy Disk Emulator, it decodes as FM. Seems to work fine in an Apple II emulator though. I seem to be able to write the DSK just fine (using a 40 track 5.25" drive) but it doesn't work when I try boot the Apple II from it.

I can do a "fluxengine read apple2 shugart_drive" from a known working physical Apple DOS 3.3 disk and then write it out to a blank just fine. That .dsk also decodes as an FM disk. If I force the Apple II decode, I see nothing sensical. The .dsk I created from my known working disk and the .dsk from that archive.org file look similar in terms of file size, and their representation in FloppyDiskEmulator both seem pretty similar. I've tried to write out a few other .dsk files too, none work.

So, I'm not sure what's going on. Is this consistent? If so, why? or am I doing something goofy?

cheers!
e
 
Hi, time has passed but I've also just tried reading Apple II system disks with a Greaseweazle v4 and Fujitsu Copal 360k 5.25" floppy drive. Using the greaseweazle apple2.appledos.140 format the read process only recognizes tracks from T20 to T34; with scattered sector remnants beforehand. eg 1 of 16 sectors, 4 of 16 sectors, 7 of 16 sectors etc before T20.
This is the case for no format SCP reads and appledos formatted .do reads and img.
Using fluxengine is see the same result with the greaseweazle setup. Displayed more graphically on the read graph created during reading.

The same setup has no trouble reading a TRS-80 system disk 35T SSSD, so I am curious if the Apple GCR standard is not compatible with this Fujitsu drive? Has anyone had a similar issue?
 

Attachments

  • Capture.PNG
    Capture.PNG
    24.6 KB · Views: 6
  • Capture1.PNG
    Capture1.PNG
    27.7 KB · Views: 6
Update: After discussions with Keir Fraser, creator of the Greaseweazle, I've done a Fluxengine driveresponse test on the Fujitsu Copal drive. I found it only responds effectively to 10us while Apple II GCR disks require a response sensitivity of up to 12us. This failure to actuate bit data on the early tracks of the read Apple2.appledos.140 GCR disks caused corruption on tracks <T20 which made the image unworkable.
 
Hi @richarjo , how did you do this drive response test? I've got a Greaseweazle and need to go through and/or archive a moving box worth of Apple II floppies.

I'm not sure what would be better, hook up a complete Apple II with all it needs or try to find a Greaseweazle setup which works
 
powerlot said:
Hi @richarjo , how did you do this drive response test? I've got a Greaseweazle and need to go through and/or archive a moving box worth of Apple II floppies.

I'm not sure what would be better, hook up a complete Apple II with all it needs or try to find a Greaseweazle setup which works
Click to expand...
Hi powerlot. Thanks for the question. I did the driver response test on the Teac 360k FD-55BR 102-U fdd and it reported <2.5us to 15us read timings which is brilliant.
The drive and greaseweazle read single sided Apple II disks with appledos.140 format and .do disk type perfectly.

The only caveat is with flippy disks. I didn't realize at the time but the Teac 5.25" fdd (and most other 5.25" DSDD fdd's) are offset by 4 tracks on side 1. This meant the drive reads side 1 T4 as T0, missing the first 4 tracks on the side of the flippy disk.
The Teac it requires the index sensor to function and flipping a disk gives no sync because the index hole is on the wrong side. The "flippy offset" option in GWGUI is only for modded drives.

The greaseweazle is ideal for standard disks but the Apple II set up with something like Diskcopy Plus would be needed for flippy's utilizing the Apple Disk II drive with no index sensor - unless you ran a modded drive, cut another index hole in the disk or used Applesauce hardware.
 
Last edited:
richarjo said:
Hi powerlot. Thanks for the question. I did the driver response test on the Teac 360k FD-55BR 102-U fdd and it reported <2.5us to 15us read timings which is brilliant.
The drive and greaseweazle read single sided Apple II disks with appledos.140 format and .do disk type perfectly.
Click to expand...
Thanks for the reply, I would like to go through my drives and check if they are good for this task. Which command did you run to obtain the read timings?
 
powerlot said:
Thanks for the reply, I would like to go through my drives and check if they are good for this task. Which command did you run to obtain the read timings?
Click to expand...
The following command is on the fluxengine site. It didn't run initially so the author recommended specifying a drive which I did. The step interval is normally .1 and I get a buffer overflow from the greaseweazle so you may need to play around with it; .2 or .5 are easier.
fluxengine analyse driveresponse --cylinder 0 --min-interval-us=0 --max-interval-us=15 --interval-step-us=.1 --write-img=output.png -d drive:0
 
Last edited:
Thanks. Which build of fluxengine is working for you? It seems to be bugged in my case. gw.exe works fine but the latest version of fluxengine doesn't spin the drive or complains about tracks or indexes not found or just hangs.

EDIT: Here it also talsk about a custom connector for attaching an Apple II drive... https://github.com/davidgiven/fluxengine/blob/master/doc/disk-apple2_drive.md
Any ideas what that could be? I couldn't find anything else about it
 
Last edited:
Thanks for the link. I don't currently have fluxengine installed though the last time I used it it was the development build from September 3.
 
Double posting apologies.
powerlot said:
Thanks. Which build of fluxengine is working for you? It seems to be bugged in my case. gw.exe works fine but the latest version of fluxengine doesn't spin the drive or complains about tracks or indexes not found or just hangs.

EDIT: Here it also talsk about a custom connector for attaching an Apple II drive... https://github.com/davidgiven/fluxengine/blob/master/doc/disk-apple2_drive.md
Any ideas what that could be? I couldn't find anything else about it
Click to expand...
The apple II Disk II compatible command is already in the fluxengine command set afaik. The custom connector isn't a thing according to Keir Fraser (GW developer) and would need to be made up following the wiring config of the drive and GW.

Attached is the earlier September dev release of fluxengine for Windows if that is suitable to test your setup.
 

Attachments

Thanks again. After downloading your version I still had the same problem. When going through the open issues I found this one:
github.com

Addressing drive ID 0 (NOT PC drive A:) · Issue #764 · davidgiven/fluxengine

Using FluxEngine + Greaseweazle on MacOS: I have the Greaseweasle working with gw with a DD disk drive from an Atari ST (drive ID 0 and straight non-twisted cable) and using a command like gw read ...
github.com github.com

Apparently fluxengine doesn't handle this properly without specifying shugart_drive in the configuration or command, whereas gw makes a clear distinction between drive 0, 1, A and B and because I had a straight cable and DS0 this was a problem in fluxengine because it assumes your drive is hooked up with a twisted cable
github.com

fluxengine/doc/drives.md at master · davidgiven/fluxengine

PSOC5 floppy disk imaging interface. Contribute to davidgiven/fluxengine development by creating an account on GitHub.
github.com github.com

.\fluxengine.exe analyse driveresponse shugart_drive --cylinder 0 --min-interval-us=0 --max-interval-us=15 --interval-step-us=.1 --write-img=output.png -d drive:0

I could verify that my drive (Mitsubishi MF501C-312M) can handle 2us-15us.

richarjo said:
The custom connector isn't a thing according to Keir Fraser (GW developer) and would need to be made up following the wiring config of the drive and GW.
Click to expand...

Is this documented somewhere?

en.wikipedia.org

Disk II - Wikipedia

en.wikipedia.org en.wikipedia.org

As far as I know the Shugart floppy connector doesn't have separate phase signals for the stepper motor.
 
powerlot said:
I could verify that my drive (Mitsubishi MF501C-312M) can handle 2us-15us.
Click to expand...
Congratulations on a good result. The command line produces a graph image as well that can show effective read/write performance and drive chatter. IDK where it is saved, last time I did the test I couldn't find the image where previously it was in the fluxengine folder. I think the stored file location must vary between the stand-alone fluxengine package and the installed version...

powerlot said:
Is this documented somewhere?
Click to expand...
The comment was part of a post by Keir in the Greaseweazle Facebook group.
 
You must log in or register to reply here.
Back
Top