New QLiberator v3.37

[RalfR]

Not tested but great to read this. Thanks for your effort

[NormanDunbar]

After 27 years we now have an update to Q_Liberator. The last version was v3.36.

Brilliant work, well done everyone involved.

Might I suggest that the source code etc be uploaded to the SinclairQL github repository for future use? https://github.com/SinclairQL

I'm happy to do this for you if you need a hand.


Cheers,
Norm.

[EmmBee]

After 27 years we now have an update to Q_Liberator. The last version was v3.36.

Brilliant work, well done everyone involved.

Might I suggest that the source code etc be uploaded to the SinclairQL github repository for future use? https://github.com/SinclairQL

I'm happy to do this for you if you need a hand.


Cheers,
Norm.

Hi Norm,

Yes, that sounds like a very good idea to me. I haven't used the github repository yet. I welcome your offer to help, so yes, please go ahead and upload the files. If there are any other files you need, please let me know.

Cheers,
Michael

[JonS]

This is really good work.

I've loaded it and given it a quick try and noticed (no criticism implied!):
a) Loading takes several seconds (about 5), the previous version was pretty much instant. Is that expected?
b) You can move beyond original screen, but the saved position is still within the original screen area.

Going to compile a number of my programs to see how it goes, especially ones with $ hex values in them.

Once again, great to see the compiler can be fixed, improved, etc. Well done.

QPC on i3 processor.

[RalfR]

You can move beyond original screen, but the saved position is still within the original screen area.

This seems to be a problem with the data block of QLibs own Pointer extensions (QLib6_bin). See also line

350 QLIB_area = ALCHP(64) : POKE_L (224 + RA6 ), QLIB_area

[EmmBee]

a) Loading takes several seconds (about 5), the previous version was pretty much instant. Is that expected?

I have noticed this long loading time myself. It seems that the larger screen sizes take a proportionately longer time for it to start up.
I could be wrong, but my guess is that it's got something to do with the detect_screen_limits routine - that's the one that finds SCRXLIM and SCRYLIM.

[RalfR]

It seems that the larger screen sizes take a proportionately longer time for it to start up.

Hmm, 3.36 starts immidiately on a 1024*768 QPC2...?!

[EmmBee]

It seems that the larger screen sizes take a proportionately longer time for it to start up.

Hmm, 3.36 starts immidiately on a 1024*768 QPC2...?!

Try this: add in ... 32529 IF VER$="HBA": SCRXLIM = SCR_XLIM(#c): SCRYLIM = SCR_YLIM(#c): CLOSE#c: RETurn

[NormanDunbar]

The zip file previously attached to this thread, version 3.37, has now been given a home - it's on GitHub under the Sinclair QL pages, and lives at https://github.com/SinclairQL/QLiberator.

Anyone thinking of working on this should fork the repository, do the necessary, then submit a pull request. If you wish that is!

Cheers,
Norm.

[RalfR]

Try this: add in ... 32529 IF VER$="HBA": SCRXLIM = SCR_XLIM(#c): SCRYLIM = SCR_YLIM(#c): CLOSE#c: RETurn

Ah, I see, you open #c as a dummy test channel, but that isn't necessary, if you are with SMSQ/E, so you can read SCR_XLIM and SCR_YLIM directly. You can also use Dilwyn's scrxlim tool as this is able to be used with standard PE, eg QL and of course SMSQ/E:

http://www.dilwyn.me.uk/tk/scrxlim.zip

The 5 sec mentioned seem to be the time of the loop detecting the screen size.