On Wed, 2003-02-19 at 17:34, Juan Pablo wrote:
> Hi
> I have been folowing the thread, and all i can do is post what i know,
> wich is not the last word, of course...
>
> There is two kind of "visuals", truecolor and indexed (indexed are out
> of the scope)
>
> BPP R G B A
>
> 15 5 5 5 0
> 16 5 6 5 0
> 24 8 8 8 0
> 32 8 8 8 8
>
> Actualy 32bpp is not a visual, alpha channel is used into the program or
> if the hardware have acceleration for overlay, etc... but there is no
> alpha structure in X.
> Human eye can distinguish more luma(grayscale) than croma(colors), that
> is why, when the light is very low we start to see in grayscale.
> Because we have more luma receptors than croma (don know the name in
rods and cones.
iirc cones are the receptors for colour, and rods are for b&w
Of course.. that could be the opposite way around.. but I got a 50%
chance of being right ;)
> english :). This fact was used long ago by the people who invent the TV
> signal. NTSC for example, it uses YUV or YIQ (is is the same for this
> pourpose) half brandwidght for Y (luma) and the other half for UV.
> MPEG and its derivates like JPEG uses the same principle (see
> subsampling 2:1:1, 4:2:2 etc)
> So i do not recomend using a jpg to store cientific data, instead BMP is
> so basic that there is no way to loose data (tiff can be saved as jpeg:)
>
> So i recommend you to use all the bits you want per pixel in an internal
> data structure (array) and then shift to 8bpp to display it :)
>
> chau
>
> Juan Pablo
>
> http://xjuan.gq.nu
>
>
> _______________________________________________
> gtk-app-devel-list mailing list
> gtk-app-devel-list@gnome.org
> http://mail.gnome.org/mailman/listinfo/gtk-app-devel-list
--
Azrael
("\''/").___..--'''"-._
`0_ O ) `-. ( ).`-.__.`)
(_Y_.)' ._ ) `._ `. ``-..-'
_..`--'_..-_/ /--'_.' .'
((i).-'' ((i).' (((.-'
Of all God's creatures there is only one that cannot be made the slave
of the lash. That one is the cat. If man could be crossed with a cat it
would improve man, but it would deteriorate the cat.
ICQ#52944566
Registered Linux User: 269002
This is a digitally signed message part