Which emulator?

[Rhodanaj]

first explanation, i will work this out furder:

 

when you open the PHOENIX disk with my tool, you see that there are 720 TI-sectors and 720 P-code blocks on your disk.

my program is working for 720 TI-sectors and 360 P-code blocks

Your disk is singel density and double sited. you can see this with 'View block' in block 0 after the TI-disk name 02D0 = 720 sectors.

Also with 'View block' you can see in block 2 after the P-code disk name there is 02D0 or 720 blocks on the disk.

TI has sectors of 256 byte, P-code uses blocks of 512 byte.

When I press 'Save as' i see that the program counted 1440 sectors.

So somehow the diskinfo is not right.

How did you make this virtual disk ?

 

What went wrong:

When you open your PHOENIX disk and you click on Extented Disk info left under, you will see an empty space of 318 sectors between file 9 and 10.

When a text file is changed with Notepad, the program deletes the old file, crunch the disk and writes the changed file at the end. 

 

I have to see how the sectors are arranged in this kind of disk. I use the normal TI layout:

first side the sectors 0 to 359 = sector 0 at the start, sector 359 at the end

second side the sectors 719 to 360 = sector 719 at the start, sector 360 at the end 

This way when you have to write sector 360 after 359 the writing head does not have to go back to the beginning 

 

When the second side on your (virtual) disk begins with 360 then it will explain the error. I'll study your disc more closely.

With the tools 'View Block' , 'Extend Disk info' and 'showing contend of' you can examen the disk.

The disk image that you have opened with my program does not change until you press ReSave or when you Overwrite it with 'Save Disk'.

 

Edited August 18, 2020 by Rhodanaj

[Rhodanaj]

The problem is not in the arrangemend of the disk, that i good news  

My program looks for the disksize to the amount of sectors in block 0. For the disk made by Fred Kaal it says 720 sectors, even when it are 1440 sectors.

In the 'Save as' commando, i count the real amount of sectors.

The 'Crunch' command does not go beyond the disksize, so the sectors above 720 do not move, but the addresses to those files in the FDR have changed.

When i use the real amount in the 'Crunch' command, everything was working again.  

I will make a new version of this program (V3.0) and i will change the disksize for new P-code disks to 1440 sectors or 720 blocks. 

[+Vorticon]

Glad this got sorted out! Nice work.

[apersson850]

I've just started playing with the emulator (Classic 99) and the p-code tool. I've only edited a text in Windows (Notepad++), removed the old file from the p-system's disk, created a new file from the Windows text with the same name as the old one, changed that name by adding .TEXT and do a ReSave.

That works. At least as I don't forget to save it from Notepad++ after doing a change...

 

The p-system should be able to handle even larger disks than the 360 Kbytes the TI supports, provided the sector access DSR can handle requests for larger than normal sector numbers. In a high-fidelity emulation, it will not work with a floppy, but making a RAM-disk could work.

[Rhodanaj]

until V3.0 is ready, you can use this version V2.2

every imported textfile gets now a '.TEXT' extention.

The program checks if the name already exist and put '01' behinde the name when it does.

This version can only use 360 block disks (TI 720 sectors). When you open a 720 block disk there is no warning. 

Version V3.0 will be able to use 720 blocks.

 

The commands 'Get P-code file' and 'Get Data-file' are not checking if there is already a file with the same name.

When you import those files, I assume you will need the same name.

 

p-code-toolV2.2.exe

Edited August 20, 2020 by Rhodanaj
about 'Get P-code file' and 'Get Data-file'

[apersson850]

If you aren't too fond of TI's peculiar big and small upper case font, you can place the SYSTEM.CHARAC file on this disk on your root volume (#4) and re-initialize. When this file resides on the root volume (prefix *), it will be used in place of the OS:SYSTEM.CHARAC, in spite of the latter still being there.

Pascalchar.dsk

[+Vorticon]

Nice font indeed. Thanks!

[Rhodanaj]

Version 3.0 of the p-code-tool is ready. I needed this program for myself, but everyone who thinks he can use it, can download it for free.

You can now handle 1440 sector disks (720 blocks in Pascal) with it. Also you can resize a 360 sector, or 720 sector disk to a 1440 sector disk (sector or track image file).

I also made a short manual for it.

p-code-tool-V3.0.exe manual.pdf

[JasonACT]

On 8/19/2020 at 5:10 AM, Rhodanaj said:

I use the normal TI layout:

first side the sectors 0 to 359 = sector 0 at the start, sector 359 at the end

second side the sectors 719 to 360 = sector 719 at the start, sector 360 at the end 

This way when you have to write sector 360 after 359 the writing head does not have to go back to the beginning

@Rhodanaj I think @Tursi has depreciated the option to use this format in Classic99 (you have to edit the ini file directly now) and has a comment it's being removed eventually.

 

I'm adding this format to my hardware project at the moment, since this tool seems the best way to create p-code disks, and >90KB disks were giving me trouble in my p-code implementation / disk DSR sector read.

 

I'll assume 1440 sector disks are arranged the same way, split and going in reverse at the mid point.

 

Thanks for your cool utility.

[Rhodanaj]

@JasonACT I have never had a real TI drive that could handle 1440, so I'm dependent of 1440 disks that I find on the internet. It is indeed logical to arrange the 1440 disk the same way, split and going in reverse at the mid point.

When I decided to pay attention to the virtual P-code again, some years ago, I immediately ran into a wall of annoyance because of the hassle of loading the editor and the compiler everytime when you make a small mistake. I had to look for a possibilitie of creating the text through another editor and transferring it somehow to the TI disk. But only after I found Peter Miller's site, with all the information about the structure of the P-code disk, things made progress. With that site Peter Miller will never be forgotten... https://petermiller.work/pmiller/

 

I have now completed version 4.1. A text file can now be opened immediately by a double click and when closed after a change, the disk is immediately saved. Still be careful if you work alternately in Pascal and Assembler. You have to reload the disk every time you have compiled or assembled a file, otherwise you will lose the exe file when linking (I speak from experience...). https://drive.google.com/drive/folders/1NmDA1LMIFFecp0aERg9jQR8ZLkaP9pDV?usp=sharing

[Rhodanaj]

Again a new version of Pcode tool....

Now you can get and save a datafile from and to Windows, without changing something in the file.

p-code-tool 5.0.exe manual v5.0.pdf

[apersson850]

On 11/4/2023 at 11:14 PM, Rhodanaj said:

...with all the information about the structure of the P-code disk...

There are several versions of the p-system internal architecture guide, which has such information.

Internal_Architecture.pdf