| T.R | Title | User | Personal
Name | Date | Lines |
|---|
| 1564.1 | This is a Session Mgr limitation, not X limitation | STAR::ORGOVAN | Vince Orgovan | Fri Oct 13 1989 18:45 | 17 |
| In X a color is represented by 16 bits of data for each of the
red, green and blue intensities in an RGB values, i.e. they are
scaled between 0 and 65535. The session manager presents this
range to the user with a scale that goes from 0 to 100%.
This means that the session manager only allows you to select
from 101 different red intensities, instead of offering you a
choice of 65536 red intensities.
The session manager does the conversion between scale value
and color intensity using:
color_intensity = (scale_value * 65535) / 100
Thus each scale percentage point changes the color intensity by
about 6.
|
| 1564.2 | VSII/GPX has less granularity also | DECWIN::FISHER | Burns Fisher 381-1466, ZKO3-4/W23 | Sun Oct 15 1989 23:13 | 13 |
| Vince has explained the problem you are seeing. Now for the problem
you might see on a different system: Note that as Vince says, the
protocol has a granularity of 65535 intensities per primary. When this
number gets to the device, it then must be scaled to what the h/w can
render. In the case of the VS2000/GPX 4-plane you correctly state that
this granularity should be 256 intensities per primary. However, if
you have a 4-plane VSII/GPX my recollection is that you now only have
16 intensities per primary. Thus, you will see a limitation greater
than that of the session manager in this particular hardware.
Burns
|
| 1564.3 | | PSW::WINALSKI | Careful with that VAX, Eugene | Sun Oct 15 1989 23:39 | 10 |
| RE: .2
Wait a minute. I thought that a GPX, regardless of the number of planes,
gave you 8 bits = 256 intensities per primary, for a total possibility of
16M different colors. The number of planes merely governs how many of these
can be simultaneously on the screen. The GPX implements hardware pseudo-color,
or so I thought.
--PSW
|
| 1564.4 | Some GPX's have 4 bits per channel | STAR::BMATTHEWS | | Mon Oct 16 1989 09:11 | 6 |
| Well it actually depends on which DAC is used in conjunction with the dragon
chip set(GPX). Dragon doesn't contain a DAC itself. On all but the 4 plane
vs2000/pvax there is 8 bits per channel. For that one board there is only
4 bits per channel.
Bill
|
| 1564.5 | | MU::PORTER | misuse a milk crate today! | Mon Oct 16 1989 12:00 | 16 |
| > Well it actually depends on which DAC is used in conjunction with the dragon
> chip set(GPX). Dragon doesn't contain a DAC itself. On all but the 4 plane
> vs2000/pvax there is 8 bits per channel. For that one board there is only
> 4 bits per channel.
Oooooops.
I was trying to figure this out with Dave Garrod and
another guy. I *insisted* that the DAC absolutely
definitely was the same on all systems, and therefore
there had to be another explanation.
A little knowledge is a dangerous thing, eh?
Sorry.
|
| 1564.6 | | DECWIN::FISHER | Burns Fisher 381-1466, ZKO3-4/W23 | Thu Oct 19 1989 20:45 | 7 |
| re .-1: You are right about a little knowledge. Note that I had the
hardware backwards. Sigh. Why don't I just shut up about hardware
and let Bill handle it?
Burns
|