[AD] patches for color.c and graphics.c (was: Re: 4.2.1)

[Andrei Ellman <ae-a-alleg@xxxxxxxxxx>]

Elias Pschernig wrote:
> On Thu, 2006-11-23 at 23:58 +1100, Peter Wang wrote:
> > > Also, I notice that the patches I submitted on 25-01-2006 ( 
> > > http://sourceforge.net/mailarchive/message.php?msg_id=14585796 ) for 
> > > color.c and graphics.c were not included in either 4.2 or 4.3. Was it 
> > > decided not to include them, or has someone just forgotten to apply them?
> > I seem to have a vague recollection that Elias looked into the color.c
> > patch?  I'm not considering the patches for 4.2.1.
>
> Ah, yes, I remember now. Looking at the thread
> http://sourceforge.net/mailarchive/forum.php?thread_id=9484132&forum_id=34599 it seems my conclusion was there's no point, after making a test image ( http://allefant.sourceforge.net/pics/allegro/palette.png - left/right halfes of each gradient are before/after).. and I can only even see the separation line when zooming in 10 times and looking at my monitor from the side for a while.


Even though there is no perceptible difference in the PNG between the 
two methods of converting 6-bit to 8-bit, I still think my patches 
should be applied for the following reasons:

+ With the shifting method, the progression of values will be linear. 
Using an integer-divide creates a progression of values with a slight 
curvature.
+ create_blender_table() will work faster - especially on older 
processors where multiplies and divides are expensive.
+ Using shifts will make the code consistent. _set_colorconv_palette() 
currently uses shifts, which means that there is currently an 
inconsistency in the way that Allegro converts 6-bit values to 8-bit vales.

I think that the last point is particularly important, as for a given 
6-bit input, the 8-bit result should always be the same.


AE.