A PET storage option

[MikeS]

Great stuff, looking forward to it; thanks!

My problem saving files turned out to be the cards themselves; I didn't notice that out of four cards, the three that couldn't save were all the same make (Strontium) and of course I never tried to save to the fourth. No problem using them in a PC and they load fine into the PET so they're still useful as master library disks, and all my other cards load & save fine.

I must say that what I'd really like to see, more than DLOAD and wildcards even, is support for saving long file names. I realize that most folks will be using these cards to load software that originates on a PC into the PET, but my main interest is to archive software currently on dozens of 8050 and 4040 diskettes, i.e. writing files rather than reading them.

And then there's the matter of sequential files, like the data files for adventure games, etc. etc....

Did you know what you were getting into? ;-)

 

[bitfixer]

Hey Mike,

Great stuff, looking forward to it; thanks!

Thanks, coming right up later this week.

My problem saving files turned out to be the cards themselves; I didn't notice that out of four cards, the three that couldn't save were all the same make (Strontium) and of course I never tried to save to the fourth. No problem using them in a PC and they load fine into the PET so they're still useful as master library disks, and all my other cards load & save fine.

Glad that saving is working on one of the cards - I might try to get hold of a Strontium card to see if I can nail down why things aren't working on them. In the meantime I'm putting together a list of cards that do and don't work so at least that information is out there.

I must say that what I'd really like to see, more than DLOAD and wildcards even, is support for saving long file names.

Absolutely, this will be coming up right after this update is out there. Just need to write a little code to create long file name entries, and read up a bit on how FAT32 uses them.

And then there's the matter of sequential files, like the data files for adventure games, etc. etc....

absolutely, there was also a Pascal interpreter (or compiler? not sure) that I was trying to run that seemed to require them. Back to the documentation..

Did you know what you were getting into?

ha ha, definitely not. But I'm having fun with it, and definitely want to bring as much functionality as possible to the PETdisk.

 

[MikeS]

Something else to think about: I was trying a couple of copy programs to copy from the 8050 to the PETdisk and apparently they failed when trying to get status/error info from the PETdisk; probably not too hard to work around but as I said, something to think about...

 

[bitfixer]

Interesting, sounds like it wouldn't be a difficult fix. When you get a chance maybe you can send me those copy programs and I can try them out.. although I will have to track down a real CBM drive for testing Or repair my semi-broken SFD-1001.

 

[MikeS]

Interesting, sounds like it wouldn't be a difficult fix. When you get a chance maybe you can send me those copy programs and I can try them out.. although I will have to track down a real CBM drive for testing Or repair my semi-broken SFD-1001.

Let me play with a few different ones when I have some time and see what I can glean, and I'll get back to ya.

 

[bitfixer]

quick warning about one of the PET games

quick warning about one of the PET games

Hi all,

I was just playing around with some of the games in that large collection Steve posted earlier. Thanks again for that Steve.
Just wanted to give a quick word of warning about one of the games in there -
I noticed that the game "andromeda" has the infamous killer poke right on the first line:
POKE 59458,62

without checking I ran this on my 8032, and then noticed that one line of text was especially bright. I turned the 8032 off right away, and thankfully nothing fried. I may do the rest of my testing on a non CRTC PET!
In any case, just a reminder to do a quick listing of the programs before running, to be safe.

 

[dave_m]

I will continue to nail down the firmware coding, and will try to wrap up tonight or tomorrow. Aiming to send out the chips Friday or Saturday. Thanks again for your patience and happy PETdisk'ing.

gub,

My new chip works fine. On the 8032 business keyboard, lower case must be used in the dload "file" on u9 command. In other words U9 will not work. I would guess with the graphic keyboards, upper case is what works. One minor issue on shipping was that even though you had the part in a plastic case, the pins got a little squished. A little repair with needle nose pliers fixed them up. I'm not sure I have ever seen a 28 pin skinny DIP. Were the sockets hard to find?

Do you want the old part back for some other application?
-Dave

 

[bitfixer]

Hi Dave,

Good to hear the new chip is working for you. I can check into the issue with the dload command. Fortunately now you can just load the software update from the SD card! I'll be posting the current firmware file as soon as I get a chance, along with the source code.
Sorry about the pin squishing, I repurposed some SD card holders as chip protectors but I guess they were a little too small.
The sockets for the 28 pin skinny dip were not hard to find, fortunately. I live just down the road from Jameco electronics so I got most of the parts there, and they had them. Unfortunately they seem to have run out of the vertical USB connectors recently, so I had to order my next batch from digi-key instead.

If it's convenient to send back the old part, I would make use of it eventually. Thanks for offering..

 

[Eudimorphodon]

Just thought I'd chime in and say the upgrade chip seems to be working fine. It seems to have fixed some directory anomalies I'd seen with the old chip, and better yet that conflict between the PETdisk and my SD-1 seems to have been resolved. (I moved several programs from the PETdisk to real floppies and didn't encounter those hanging/"lost drive" symptoms like I'd been seeing before.)

The only "bad" thing I saw was when using the disk with a 16k-upgraded 2001-8 with BASIC 2.0. (I was given back the calculator-keyboard 2001 I picked up from the Albany PET dump. My friend decided he had too many hobbies.) On that machine running a "LOAD "$",9" would hang returning to the READY prompt. I believe the problem is simply that I have too many files on the SD card and with only 16k I'm running out of RAM, but I haven't had the time to delete some and find out yet. LOAD-ing programs I know the name of works fine.

 

[sjgray]

Just thought I'd chime in and say the upgrade chip seems to be working fine. It seems to have fixed some directory anomalies I'd seen with the old chip, and better yet that conflict between the PETdisk and my SD-1 seems to have been resolved. (I moved several programs from the PETdisk to real floppies and didn't encounter those hanging/"lost drive" symptoms like I'd been seeing before.)

The only "bad" thing I saw was when using the disk with a 16k-upgraded 2001-8 with BASIC 2.0. (I was given back the calculator-keyboard 2001 I picked up from the Albany PET dump. My friend decided he had too many hobbies.) On that machine running a "LOAD "$",9" would hang returning to the READY prompt. I believe the problem is simply that I have too many files on the SD card and with only 16k I'm running out of RAM, but I haven't had the time to delete some and find out yet. LOAD-ing programs I know the name of works fine.

Each directory entry takes about 60 bytes when you use LOAD"$",x. The 300+ zip of pet games I put together takes about 17K ram to load the directory. This is why we need support for the CATALOG command, which just displays the directory without loading...

Steve

 

[MikeS]

Each directory entry takes about 60 bytes when you use LOAD"$",x. The 300+ zip of pet games I put together takes about 17K ram to load the directory. This is why we need support for the CATALOG command, which just displays the directory without loading...

Steve

Nah, we need that RAM/ROM board so everybody can upgrade to 32K ;-)

 

[sjgray]

Nah, we need that RAM/ROM board so everybody can upgrade to 32K ;-)

Why stop at 32K? LOL

Steve

 

[Eudimorphodon]

Why stop at 32K? LOL

Heh. I'll have to take some pictures of the delicate brain surgery that was done to that 2001-8. Technically it has 64k of RAM, just only 16k of it are visible. ;^) (I've been pondering how hard it would be to get access to another 16k of it for a full 32k.)

(This 2001-8 is the odd-bird one with a possibly stock "Dynamic PET" motherboard. Does anyone have any information about how PET serial numbers might relate to manufacture date? I haven't found much in the way of PET serial numbers online, but the SN of this 2001-8, which is something over 43000, seems to be very late for a calculator keyboard model.)

In another random question, anyone know of a good (downloadable from somewhere) PET-compatible disk utility to facilitate things like copying files from one disk unit to another? LOAD followed by SAVE works, but...

 

[bitfixer]

Just thought I'd chime in and say the upgrade chip seems to be working fine. It seems to have fixed some directory anomalies I'd seen with the old chip, and better yet that conflict between the PETdisk and my SD-1 seems to have been resolved. (I moved several programs from the PETdisk to real floppies and didn't encounter those hanging/"lost drive" symptoms like I'd been seeing before.)

Great! I'd like to say that the SD-1 now working is a result of me finding and fixing something.. but unfortunately it must be a good unintended side effect of the other changes I made. I did fix an issue with capital letters in filenames, so perhaps that fixed the directory anomalies.

The only "bad" thing I saw was when using the disk with a 16k-upgraded 2001-8 with BASIC 2.0. (I was given back the calculator-keyboard 2001 I picked up from the Albany PET dump. My friend decided he had too many hobbies.) On that machine running a "LOAD "$",9" would hang returning to the READY prompt. I believe the problem is simply that I have too many files on the SD card and with only 16k I'm running out of RAM, but I haven't had the time to delete some and find out yet. LOAD-ing programs I know the name of works fine.

It does seem like this is the result of running out of RAM. Will try to get the CATALOG command working soon..

 

[bitfixer]

Hi there,

Trying to prioritize my time over the next couple of weeks.. things are getting busy at work but I'm trying to keep moving forward with PETdisk development.

A few requested features for the PETdisk have been:
- support for long filenames when saving a file (currently not supported)
- directories
- sequential/data file support
- CATALOG and other BASIC 4.0 commands

Any opinions on which of these should be first? Or any suggestions for features not listed here?
Also, I welcome anyone with AVR programming experience or interest to try their hand at hacking on the firmware code for the PETdisk. I will be posting the 1.2 code on the bitfixer site shortly, so feel free to take a look or hack away at it. It would be great to evolve the firmware code into a community project.
Thanks again!
- Mike

 

[tezza]

My own ranking would be the following.

1. Sequential/data file support
2. CATALOG and other BASIC 4.0 commands
3. Long filenames
4. Directories

Tez

 

[dave_m]

1. Sequential/data file support
2. CATALOG and other BASIC 4.0 commands
3. Long filenames
4. Directories

Tez,
I agree with the order; I do not think subdirectories were supported by the 4040 drives, but with PETDisk having so much capacity, it would be nice. A pathname like subdirectory\filename will just be a long filename on the 4040 root directory but perhaps gub may be able to parse it properly and stick the file in the correct subdirectory of PETDisk.

Of course reading the contents of a subdirectory from the PET with the DIRECTORY or CATALOG command will probable not be possible (PET syntax error) but perhaps the old way of (LOAD"$\sub", 8 ) may be able to work?

Us old time PET guys have a soft spot for the Commodore 'Universal Wedge' shortcut commands (DOS Support program) where >$ will perform a CATALOG, and /filename will load a program from device 8, but that may be going too far afield.
-Dave

 

[MikeS]

Numero uno on my wish list is long file name support, followed by sequential and (ultimately if feasible) relative file support; without that I can't do what I really want to do with a device like this, namely conveniently transfer and archive my collection of disks. If it could play nice with utilities like Unit to Unit, Copyall etc. that would be icing on the cake.

CATALOG would be nice, but I can live without BASIC 4 support and directories.

But in the meantime it's convenient (and energy-efficient) for playing games that are already in PC format without firing up the 8050/4040 and hunting for the right diskette...

I'm still hoping Seth finishes his version; I'm hoping that between the two we'll have the best of both worlds.

 

[bitfixer]

Thanks for the feedback.. among these I think the simplest to implement would be support for long files when saving, and it seems like sequential file support is highly requested. For now I'll prioritize these two. I'll try to have at least the long file support ready by sometime next week, with sequential file support soon to follow.

 

[Nama]

The Pet in a suitcase now has a PETdisk.
I was having trouble getting the cassette side USB plug to fit due to limited room in the suitcase. I eventually found a USB cable with a shorter plug design, and it all ended up fitting very snugly.
I actually forgot to reformat my card as FAT32, so for about 30 mins I thought I had screwed something up in the building process and I could not get any directory listing. After reformatting the card correctly it all works great.

I have a second kit for my blue screen Japanese Pet 2001 which I'll build sometime soon.

Thanks gubbish.

Phil