Re: A quick draft of a preliminary proposal for a possible version of the crossfire protocol

["Carl Edman" <cedman@cedman.remote.Princeton.EDU>]

From: Philip Brown <philb@soda.berkeley.edu>:
>>>>[From Carl Edman]
>     OK, let me propose a compromise.  The server will not know or
>     care what   kind of color hardware the client has.  However the
>     client can REQUEST   not just PIX and SND types, but can request
>     the types BW1, BW2, BW3, ..   BW8 to receive black and white XPMs
>     of one to eight bit planes and CO1   to CO8 to receive color XPMs
>     of one to eight bit planes per color.
> 

> To use one of your arguments back at you.. do you _really_ want to
> "support" 8 different kinds of display for each and every item face?
> that's a lot of (anoying and unneccessary) work.

No, as you may have read in the next sentence (which you so judiciously  
cut) the server can have as many or as few actual representations  
stored as it wants.  When images are requested in a color resolution  
which is not available on store the server will reduce the color  
resolution from the next better resolution available.  That is trivial.   
My argument was about supporting different IMAGE FORMATS (which I still  
oppose), not COLOR RESOLUTIONS.

> 8-bit greyscale systems have default dithering from color to what
> they have got. less capable systems will have to deal with
> black-and-white. It's their fault for buying inadequate hardware.

Yes, and how do you get a useable black and white from a true color  
image ?  Kjetil claims that it is almost impossible and I believe him  
on this point.  


	Carl Edman