MD_ERASE / S_OR_D in vrx_cpyfm

Johan Klockars johan at
Tue Apr 22 13:10:19 CEST 2003

> S_OR_D is not working properly in the following example (it's only a 
> quick & dirty example, it doesn't matter about clipping etc...)
> The first MD_ERASE isn't done correctly. Normally the result should 
> be a black box. S_OR_D does nothing.

Unfortunately, I can't test your program at the moment.

What do you get instead of the black box?

The code in the native part of the ARAnyM driver looks correct to me,
but my references disagree on what the reverse transparent mode is
really supposed to do.  ;-)
One book says very clearly that the 0's from the source will cause the
corresponding destination pixels to be set to the foreground colour
(the first in the provided array), one says equally clearly that it
should be the background colour (the second one in the array), and
one is somewhat vague but leaning towards the second interpretation.
An online version of the Compendium appears to agree with the latter
two references. I usually trust the first one, and that's what the
current code implements...

And no, I have not tested this with the original VDI, but can't
recally seeing (or hearing about) any problem before. Of course,
reverse transparent mode is not likely to be used much.

If you are certain that setting to the background colour is the 
correct behaviour in this case, the ARAnyM fVDI driver (and probably
the rest of them) should be updated.

You don't say what you get with S_OR_D, but if the vrt_cpyfm caused
a white box, vro_cpyfm with S_OR_D should indeed have no effect.

                        | Why are these |  e-mail:   johan at
                        |  .signatures  |
                        | so hard to do |  WWW:
                        |     well?     |            (fVDI, MGIFv5, QLem)
Aranym-user mailing list
Aranym-user at

More information about the cz-bobek-lists-aranym-user mailing list