I had a .D64 disk image "open" via LOAD"$:DISK.D64",8 when I saved a BASIC program via SAVE"TEST",8 and this seems to have corrupted the disk image. See attached file — it used to have more than a dozen games and now only a handful of them are seen when doing a directory listing.

(Nothing valuable lost since this was just a downloaded file I copied onto the card in the first place but thought it might be worth reporting…)

TEHGAMES-brokt.D64.zip

One of the kids and I are trying to play with our own custom webserver for the PETdisk. The easiest way to do this is on a machine or VM on the LAN using a high-numbered (unprivileged) port like 8080.

I noticed however that the config parsing doesn't look for a : separator and instead port 80 is hardcoded in the HTTP client.

We were able to work around this with a port forward on our router, and there's probably other tricks available using ssh or nginx proxies or similar, but it would be convenient if a custom port number could be configured.

(Just browsing it in GitHub without the aid of an IDE, I couldn't 100% trace through how the host thing is handled with the offset but I think it might just be a matter of adding a port field around setUrlData assuming there's some sort of atoi conversion routine handy? Maybe I'll try a stab at it but always takes me a bit to set up each different toolchain and such.)

This is a minor, but I frequently forget to add the '.D64' extension when mounting disk images.

For example, I'll type LOAD "$:FOO",8 instead of LOAD "$:FOO.D64",8.

Rather than return a ?FILE NOT FOUND ERROR, the PETDisk-Max acts as if the mount succeeded but shows an empty directory.

LOAD "$:FOO",8           <-- Ooops.  I meant 'FOO.D64'
SEARCHING FOR $:FOO
LOADING
READY.
LIST

0 "PETDISK MAX V1.0"     <-- Appears to mount an empty disk/directory
1   BUILD 41090BA
65535 BLOCKS FREE.

Once in this state, "$:.." does not appear to unmount the directory.

LOAD "$:..",8            <-- Attempt to unmount
SEARCHING FOR $:FOO
LOADING
READY.
LIST

0 "PETDISK MAX V1.0"     <-- Unmount failed
1   BUILD 41090BA
65535 BLOCKS FREE.

The only way I've found to recover is to power cycle.

PS - Great project, BTW!

Hi,

As you may know, the availability of ATMega1284p in SMD is pretty much non-existing before next year. I noticed though, that Digikey has around 4000 of those in DIP size packages.

So I solved the problem of not getting PETDisk Max myself by designing a DIP version. Some components are SMD and they are all placed underside the board (so you can still order SMD-populated board from JLC for example). I also solved the problem of getting power to the board by designing passthrough to cassette port, where you can take power. Also the PETDisk Max itself has passthrough in the PCB.

I tested this (without ESP-module as I don't have a use for it) and it seems to work :)

If there's any interest, I could drop PCB files & Gerbers here? (I'm still using Eagle, unfortunately haven't had time to really master Kicad yet).