On Fri, 14 Jan 2000, graydon hoare wrote:

> >>>>> "JMT" == Jon M Taylor <[EMAIL PROTECTED]> writes:

["Physical display device dimensions should be standard in Berlin" snipped
- I agree now, although I still think it is a lot of work.]
>     JMT>      Sure, which is like what pantone color matching
>     JMT> requires you to do.  But Windows does not automatically
>     JMT> re-calibrate the color selections of your desktop theme based
>     JMT> on the pantone calibration settings in your printer driver
>     JMT> |->.
> wait, this is absurd. you're arguing _against_ the model GGI
> proposes. 
> when I run a GGI app, it specifies a color in linear 16
> space, then calls ggiMapPixel() and gets a ggi_pixel in device space
> which it can draw with. This is precisely what we're talking about,
> only in monitor-attribute-space rather than memory-layout-space. the
> fact that monitor-attribute-space is a little harder to get the
> details of does not reduce the importance of doing the interpolation.

        Fair enough.  I never considered the idea of using physical
display characteristics in ggiMapPixel().  That is pretty strange, but I
don't see why it wouldn't work, aside from the already mentioned problem
that even GGI does not usually have access to this info.

        If this is done, the physical display calibration will need to be
done at the GGI target level, which also implies that we will need
per-target config files (like etc/ggi/display/kgi.conf) which can store
persistent config data.  We do not currently have this capability in
LibGGI(?), but appearently we need it.

        I have now crossposted this discussion to the GGI mailing list.
I'm not familiar enough with this aspect of GGI to be able to help you
much more in this area, as you can see |->.  I'll bet Andy and Marcus will
be able to work this out just fine, though, and I'll jump in if I am
needed.  OK?


'Cloning and the reprogramming of DNA is the first serious step in 
becoming one with God.'
        - Scientist G. Richard Seed

Reply via email to