Toolkit 3 and its initialization

Anything QL Software or Programming Related.
Post Reply
User avatar
RalfR
QL Wafer Drive
Posts: 1140
Joined: Fri Jun 15, 2018 8:58 pm

Toolkit 3 and its initialization

Post by RalfR »

Toolkit 3 was written by Martin Berndt (Ultrasoft) and was intended as an extension to TK2.

I wasn't entirely happy with the name at the time, but that was Martin's decision.

In my opinion, Toolkit 3 contains some very good extensions, the one that is probably the most problematic today is the handling of all directory devices when "TK3_EXT" is entered (which also initialize TK2).

It even went so far that TT prevented initialization in SMSQ/E. Not a nice move, actually, because the source code has been public for a long time and it could have been changed without any problem.

Now to the topic: In principle, you can blame me. The fact that the device drivers are handled at all is largely my fault.

I started saving my programs and games in the "pseudo" subdirectories of TK2 early on. At that time there was no talk of Level2 drivers or "MAKE_DIR".

This is what it looked like on flp1_:

game1_boot
game1_loading_scr
game1_code
game2_boot
game2_loading_scr
game2_code

Of course, you couldn't start a program this way, because they didn't know anything about the "directories".

So I had the idea that you could simply enter "DDOWN game1_" using TK2 and then start "LRUN flp1_BOOT" and "game1" would find its files, because the device driver from flp1_ would insert the "directory" into it without the programs noticing.

And that worked (in SuperBASIC and machine code), or still works under standard QDOS/Minerva with Toolkit 3. This meant that you could also use the pseudo "directories" from TK2. (as long as your device just has Level1 drivers, but you can remove any device at will in the source code if you want).

Martin specifically left out the "WIN" device because he correctly assumed that hard disk drivers should bring their own things to subdirectories, which was always the case in the early days.

Thanks to the source code of Toolkit 3, I myself assembled a version a long time ago that does not use this initialization of the device drivers (I would have to blame TT and SMSQ/E for that).

TT should have simply asked once instead of simply banning it.
Last edited by RalfR on Thu Feb 20, 2025 9:40 am, edited 3 times in total.


7000 4E75
Derek_Stewart
Font of All Knowledge
Posts: 4610
Joined: Mon Dec 20, 2010 11:40 am
Location: Sunny Runcorn, Cheshire, UK

Re: Toolkit 3 and its initialization

Post by Derek_Stewart »

Hi Ralf,

I only commented on this in the other thread, as someone else brought the subject up, and I was just adding my experience with Ultrasoft Toolkit 3.

I bought Toolkit 3 from Digital Precision about the same time Miracle Systems released he 40Mb Winnie Hard Drive, this was an upgrade to the existing 20Mb system.

Up to then no Hard drives were commercially available and it fitted into the Trump Card setup with Minerva operating system, I was running, whether the £450 price tag was economically viable is questionable. But as a QL user I bought the system.

Toolkit 3 seemed a good update to the Toolkit 2 in the Trump Card and may work with Miracle Hard Drive, but I did not really understand the device validation in Toolkit 3, so I disassembled the Toolkit 3 code, and produced a disassembly that was able to be reassembled with QMAC.

I knew enough to make Toolkit 3 initialise without the Device validation, which suited my own personally purposes, the disassembly, was never released anywhere and still on a Floppy disk and QLWA back file.

I suppose that is why I disassemble everything now, whether I understand or not, but I try.


Regards,

Derek
Post Reply