The Sinclair QL Forum

Derek_Stewart wrote: Sun May 05, 2024 3:39 pm Hi,

The SMSQ/E writers seem to just change and us poor users have figure out what they have done.

They did an amazing job with SMSQE.
The stipple issue is a minor bug that affects almost no one, and that is why it took so long to surface.

Actually, both QDOS and Minerva have stipple patterns 2 and 3 reversed compared to the figures shown in the QL Manual (and QPC Concepts guide) under 'colour'.
In stipples 1 and 2, the white pixels are background colour and grey pixels are contrast colour. But in stipples 2 and 3 they are reversed.
Under SMSQ/E, the colours are only reversed in stipple 2, not in 3 (at least using QPC2)...

janbredenbeek wrote: Mon May 06, 2024 3:54 pm Actually, both QDOS and Minerva have stipple patterns 2 and 3 reversed compared to the figures shown in the QL Manual (and QPC Concepts guide) under 'colour'.
In stipples 1 and 2, the white pixels are background colour and grey pixels are contrast colour. But in stipples 2 and 3 they are reversed.
Under SMSQ/E, the colours are only reversed in stipple 2, not in 3 (at least using QPC2)...

You are perfectly right!
But the issue still remains: we should have the same stipples in QDOS, Minerva and SMSQE.

janbredenbeek wrote: Mon May 06, 2024 3:54 pm ...
Under SMSQ/E, the colours are only reversed in stipple 2, not in 3 (at least using QPC2)...

The Jan Jones book (Quanta issue) does not fully agree with your picture.

Stipple 0: Top Right is contrast
Stipple 1: Bottom two are contrast
Stipple 2: two Right-hand are contrast
Stipple 3: Top Right & Bottom Left are contrast

BSJR

BSJR wrote: Tue May 07, 2024 10:59 am The Jan Jones book (Quanta issue) does not fully agree with your picture.

Stipple 0: Top Right is contrast
Stipple 1: Bottom two are contrast
Stipple 2: two Right-hand are contrast
Stipple 3: Top Right & Bottom Left are contrast

So Jan Jones is right and the QL manual is wrong, at least with QDOS and Minerva...

janbredenbeek wrote: Tue May 07, 2024 10:30 pm

BSJR wrote: Tue May 07, 2024 10:59 am The Jan Jones book (Quanta issue) does not fully agree with your picture.

Stipple 0: Top Right is contrast
Stipple 1: Bottom two are contrast
Stipple 2: two Right-hand are contrast
Stipple 3: Top Right & Bottom Left are contrast

So Jan Jones is right and the QL manual is wrong, at least with QDOS and Minerva...

Or maybe there was a change in the screen driver after Jan published the book.

Suerly the GD2 colour drivers written by Tony Tebby is correct, or is God wrong as well.

An argument could be made that only for Stipple 0 does the order of colours matter. The others will all look the same from a distance! So neither OS might be wrong/right!

Derek_Stewart wrote: Wed May 08, 2024 7:26 am

janbredenbeek wrote: Tue May 07, 2024 10:30 pm

BSJR wrote: Tue May 07, 2024 10:59 am The Jan Jones book (Quanta issue) does not fully agree with your picture.

Stipple 0: Top Right is contrast
Stipple 1: Bottom two are contrast
Stipple 2: two Right-hand are contrast
Stipple 3: Top Right & Bottom Left are contrast

So Jan Jones is right and the QL manual is wrong, at least with QDOS and Minerva...

Or maybe there was a change in the screen driver after Jan published the book.

Suerly the GD2 colour drivers written by Tony Tebby is correct, or is God wrong as well.

The plot thickens.
The Jan Jones way has a certain logic in that Top Left is always the Main colour.
It's described in text and doesn't rely on a picture, and it's backed up by JS/QDOS & Minerva practice.

The mode 4 Stipple colour Tool from Oliver Fink has 0 to 2 the QDOS way but 3 the SMSQ way.
(Example from my QPC2 but identical on Aurora/JS.)
TT & God work in mysterious ways.

BSJR

I have done some more testing on the Stipples with QDOS, QPC2 & SMSQmulator .
There were a number of inconsistencies until I realised that SMSQ defaults with borders around the standard windows.
As the Stipple block should start at an even position, this may mess up the starting pixels with an odd border width.

Once these were eliminated only Stipple 3 remained that reversed the colours.
Assuming the Jan Jones Book is correct, then the Oliver Fink Tool also has the S3 example reversed but strangely the preview window is correct. Doing a PAPER 244 on window #1 (no border) has reversed the colours as expected (SMSQmulator).
(Apologies to anyone who has trouble distinguishing Main Green & Contrast Red.)

BSJR