[Search for users] [Overall Top Noters] [List of all Conferences] [Download this site]

Conference bulova::decw_jan-89_to_nov-90

Title:DECWINDOWS 26-JAN-89 to 29-NOV-90
Notice:See 1639.0 for VMS V5.3 kit; 2043.0 for 5.4 IFT kit
Moderator:STAR::VATNE
Created:Mon Oct 30 1989
Last Modified:Mon Dec 31 1990
Last Successful Update:Fri Jun 06 1997
Number of topics:3726
Total number of notes:19516

1666.0. "Color customization on dual-head Color/BW system" by ORAN::ORAN (Dave Oran) Fri Nov 03 1989 09:46

Has anybody gotten proper customization to happen under DECW V2 on a dual
head system where one head is color and the other is black and white.

Even though I put #ifdef COLOR and #ifndef COLOR directives in my
resource files, I can get only ONE of these to take.

I heard from a friend that XRM is either broken or missing some necessary
feature in the color/BW environment.

The usual symptom is that XRM thinks my BW screen is color, even though
I have things set up properly (I think). Here is (part of) my DECW$server_0 
logical name table:

  "DECW$BITONAL" = "F"
        = "T"
  "DECW$COLOR" = "T"
        = "F"
  "DECW$DEFAULT_FONT" = "FIXED"
  "DECW$DEFAULT_KEYBOARD_MAP" = "NORTH_AMERICAN_LK201LA"
  "DECW$FONT" = "DECW$SYSCOMMON:[SYSFONT.DECW.USER_CURSOR16]"
        = "DECW$SYSCOMMON:[SYSFONT.DECW.CURSOR16]"
        = "DECW$SYSCOMMON:[SYSFONT.DECW.USER_75DPI]"
        = "DECW$SYSCOMMON:[SYSFONT.DECW.75DPI]"
        = "DECW$SYSCOMMON:[SYSFONT.DECW.USER_100DPI]"
        = "DECW$SYSCOMMON:[SYSFONT.DECW.100DPI]"
  "DECW$KEYBOARD" = "GAA1:"
  "DECW$KEYMAP" = "DECW$SYSCOMMON:[SYS$KEYMAP.DECW.USER]"
        = "DECW$SYSCOMMON:[SYS$KEYMAP.DECW.SYSTEM]"
  "DECW$MONITOR_DENSITY" = "75"
  "DECW$POINTER" = "GAA2:"
  "DECW$SERVER_DISABLE_CH" = "N"
  "DECW$SERVER_SCREENS" = "GAA0"
        = "GCA0"
  "DECW$SERVER_TRANSPORTS" = "DECNET"
        = "LOCAL"
        = "LAT"
  "DECW$SYSCOMMON" = "SYS$COMMON"
  "DECW$XSIZE_IN_PIXELS" = "1024"
  "DECW$YSIZE_IN_PIXELS" = "864"
T.RTitleUserPersonal
Name		DateLines
1666.1DECWIN::FISHERBurns Fisher 381-1466, ZKO3-4/W23Fri Nov 03 1989 09:514
I don't believe there is a way to tell the resource manager to do something
different for b/w vs color.

Burns
1666.2QAR material?ORAN::ORANDave OranFri Nov 03 1989 12:214
>I don't believe there is a way to tell the resource manager to do something
>different for b/w vs color.

Then what are those #ifdef COLOR directives I see sprinkled around?
1666.3cppMELTIN::dickGvriel::SchoellerFri Nov 03 1989 12:466
I don't remember what the name of it is, but the ULTRIX equivalent to the
RELOAD hack on VMS will allow you to run the c pre-processor on .Xdefaults
before they are processed.  Because of this you can force evaluation of
#ifdefs.

Dick
1666.4The problem seems to be that you need both...IO::MCCARTNEYJames T. McCartney III - DTN 381-2244 ZK02-2/N24Fri Nov 03 1989 13:0510
What we've seen is that the color resources get loaded for screen 0 but the 
mono resources aren't loaded for screen 1. It seems to use color for both. The
problem is (i'm guessing) that the window manager (or one if his friends) isn't
setting the second head correctly. Is there a hack so that we can force the 
resources for screen 1 only?

I don't think that the C pre-processor will help in this case.

James
1666.5The symptoms are in concert with .-1's analysisORAN::ORANDave OranFri Nov 03 1989 14:298
Can someone familiar with the code (and who has access to it) look into
this?

Given the kind of hardware we ship (PVax, PMax, etc.) with a color option
plus integral MFB, this kind of configuration will most likely be pretty
common.

We really ought to make it work with color and BW on the same system.
1666.6PRAVDA::JACKSONKing CynicWed Nov 08 1989 11:1925
From a non-developers point of view, the ONLY application that I've 
had trouble with due to (I think) the application thinking it's on 
the color display (display0) when it's running on the mono display (display
1) is DECwrite.

The DECwrite folks have found their problem, and have a fix "in the next
version" (which I'm patiently waiting for).



My DECW$PRIVATE_SERVER_SETUP.COM is below, it sets up the mono system
as screen 1 and color at screen 0.  (I never tried it the other way)



-bill

$ if f$getdvi("GCA0:","EXISTS") then goto gc_present
$ run sys$system:sysgen 
connect gca0/adap=0/csr=%X5000/vect=%o400/numv=01/driver=gcbdriver
$gc_present:
$ DECW$DEVICE == "GAA0,GCA0"
$ DECW$COLOR == "T,F"
$ decw$server_density == 75
$exit
1666.7enter a QARSTAR::BROUILLETTEMon Nov 13 1989 13:3127
    
    The session manager will read different files to set the resource
    properly on each head.  For monochrome it reads and merges
    	decw$system_defaults:decw$sm_bw.dat
    	decw$system_defaults:decw$sm_general.dat
    	sys$login:decw$sm_bw.dat
    	sys$login:decw$sm_general.dat
    	sys$login:decw$xdefaults.dat
    
    For color it reads and merges
    	decw$system_defaults:decw$sm_color.dat
    	decw$system_defaults:decw$sm_general.dat
    	sys$login:decw$sm_color.dat
    	sys$login:decw$sm_general.dat
    	sys$login:decw$xdefaults.dat
    
    The X resource manager does not understand #ifdef.  In fact, any line
    beginning with # is considered a comment.  So, I doubt that those lines
    are doing any good.  In fact, if you have a resource listed twice in
    the file, the second one is probably the one which gets used all of the
    time.
    
    We have a dual head machine with one monochrome and one color head for
    testing purposes and the resources seem to be working correctly on each
    head.    If you are still having a problem, please enter a QAR with a
    copy of the resource files and I will look into it.
1666.8I was hoping to use the application-dependent fileORAN::ORANDave OranMon Nov 13 1989 15:4310
Putting all of the color customizations in the root window property is
kind of a pain.

I currently have all of the application-specific color customizations
in the application resource file (e.g. DECW$Notes.Dat).

This is why I wanted some way to break these out. Maybe the breakdown
used by the seesion manaer could be extended to ALL resource files.

I notice that Mail has done something akin to this for their V3.
1666.9Work aroundMELTIN::dickGvriel::SchoellerMon Nov 13 1989 16:048
Dave,

A short term work around could be to have 2 different startup .com files
which point DECW$USER_DEFAULTS to different places depending on whether
the mono or color copy is desired.  I know it sucks but it could tide
you over until a color architecture is defined and implemented.

Dick
1666.10ThanksORAN::ORANDave OranMon Nov 13 1989 21:249
I actually do this for the window manager so that I load different MWM
resources on the two screens.

It also makes some sense for those applications where the only/main
customizations are for color/appearance. It's a tradeoff to have
to duplicate the other stuff and worry about consistency.

I'm beginning to think given the current limitations the easiest compromise
is to put all color costomizations in the DECW$sm_color and bw files.
1666.11sysgen config for dual heads?HIBOB::KRANTZNext window please.Mon Mar 12 1990 16:2810
I am having problems making a dual headed configuration work.

Would someone who has a dual headed system working (preferable one
with both color and B&W heads) post or mail me the outputs from
MCR SYSGEN
SHOW/CONFIG
SHOW/UNIBUS

please?
			Joe
1666.12STAR::MFOLEYJammin with Bill and TedMon Mar 12 1990 16:3624
RE: .-1

	You mean for a 3100??

	Here ya go...

Axel> mcr sysgen
SYSGEN>  SHOW/CONFIG

        System CSR and Vectors on 12-MAR-1990 16:33:21.16

 Name: OPA  Units: 1  Nexus:0    (410)
 Name: PKA  Units: 1  Nexus:0    (410)
 Name: TTA  Units: 4  Nexus:0    (410)
 Name: DUA  Units: 1  Nexus:0    (410)
 Name: ESA  Units: 3  Nexus:0    (410)
 Name: GAA  Units: 1  Nexus:0    (410)
 Name: GCA  Units: 1  Nexus:0    (410)
SYSGEN>  SHOW/UNIBUS
SYSGEN>
                                                                                


								mike
1666.13HIBOB::KRANTZNext window please.Mon Mar 12 1990 17:328
>RE: .-1
>
>	You mean for a 3100??
>
	thanks for the quick response, but no I am trying to make it work
on a 3900 and/or a VAXstation II...

		Joe
1666.14Info for VSIIAIRBAG::SWATKOElectrons are cheap. Trees are not.Tue Mar 13 1990 10:0887
Here's the requested info from a 2 headed VAXstation II...

SYSGEN>  SHO/CONF

        System CSR and Vectors on 13-MAR-1990 10:02:47.57

 Name: XQA  Units: 3  Nexus:0    (UBA) CSR: 774440  Vector1: 774  Vector2: 000
 Name: PUA  Units: 1  Nexus:0    (UBA) CSR: 772150  Vector1: 154  Vector2: 000
 Name: PTA  Units: 1  Nexus:0    (UBA) CSR: 774500  Vector1: 260  Vector2: 000
 Name: PUB  Units: 1  Nexus:0    (UBA) CSR: 760354  Vector1: 310  Vector2: 000
 Name: TTA  Units: 4  Nexus:0    (UBA) CSR: 760100  Vector1: 300  Vector2: 304
 Name: GAA  Units: 3  Nexus:0    (UBA) CSR: 777400  Vector1: 320  Vector2: 324
 Name: LDA  Units: 1  Nexus:0    (UBA) CSR: 760620  Vector1: 360  Vector2: 000
 Name: GCA  Units: 3  Nexus:0    (UBA) CSR: 777300  Vector1: 130  Vector2: 134
SYSGEN>  SHO/UNI


 ** UNIBUS map for nexus #0 on 13-MAR-1990 10:02:51.27 **

 Address 760100 (8016A040) responds with value 4060 (hex)
 Address 760102 (8016A042) responds with value 0000 (hex)
 Address 760104 (8016A044) responds with value 0000 (hex)
 Address 760106 (8016A046) responds with value 0000 (hex)

 Address 760354 (8016A0EC) responds with value 0000 (hex)
 Address 760356 (8016A0EE) responds with value 0000 (hex)

 Address 760620 (8016A190) responds with value 020C (hex)
 Address 760622 (8016A192) responds with value 003A (hex)
 Address 760624 (8016A194) responds with value FFFF (hex)
 Address 760626 (8016A196) responds with value FFFF (hex)
 Address 760630 (8016A198) responds with value FFFF (hex)
 Address 760632 (8016A19A) responds with value FFFF (hex)
 Address 760634 (8016A19C) responds with value 00F0 (hex)
 Address 760636 (8016A19E) responds with value 2030 (hex)

 Address 772150 (8016B468) responds with value 0001 (hex)
 Address 772152 (8016B46A) responds with value 006C (hex)

 Address 774440 (8016B920) responds with value FF08 (hex)
 Address 774442 (8016B922) responds with value FF00 (hex)
 Address 774444 (8016B924) responds with value FF2B (hex)
 Address 774446 (8016B926) responds with value FF06 (hex)
 Address 774450 (8016B928) responds with value FFEC (hex)
 Address 774452 (8016B92A) responds with value FFD1 (hex)
 Address 774454 (8016B92C) responds with value 01FC (hex)
 Address 774456 (8016B92E) responds with value 1151 (hex)

 Address 774500 (8016B940) responds with value 0000 (hex)
 Address 774502 (8016B942) responds with value 0000 (hex)

 Address 777300 (8016BEC0) responds with value F747 (hex)
 Address 777302 (8016BEC2) responds with value FFFF (hex)
 Address 777304 (8016BEC4) responds with value 1213 (hex)
 Address 777306 (8016BEC6) responds with value FFFF (hex)
 Address 777310 (8016BEC8) responds with value FF00 (hex)
 Address 777312 (8016BECA) responds with value FF1F (hex)
 Address 777314 (8016BECC) responds with value FF00 (hex)
 Address 777316 (8016BECE) responds with value FFCF (hex)
 Address 777320 (8016BED0) responds with value FFFF (hex)
 Address 777322 (8016BED2) responds with value FFFF (hex)
 Address 777324 (8016BED4) responds with value FFFF (hex)
 Address 777326 (8016BED6) responds with value FFFF (hex)
 Address 777330 (8016BED8) responds with value FFFF (hex)
 Address 777332 (8016BEDA) responds with value FFFF (hex)
 Address 777334 (8016BEDC) responds with value FFFF (hex)
 Address 777336 (8016BEDE) responds with value FFFF (hex)
 Address 777340 (8016BEE0) responds with value FF07 (hex)
 Address 777342 (8016BEE2) responds with value FF00 (hex)
 Address 777344 (8016BEE4) responds with value FFFF (hex)
 Address 777346 (8016BEE6) responds with value FFBD (hex)
 Address 777350 (8016BEE8) responds with value FF0F (hex)
 Address 777352 (8016BEEA) responds with value FF00 (hex)
 Address 777354 (8016BEEC) responds with value FFF8 (hex)
 Address 777356 (8016BEEE) responds with value FF90 (hex)
 Address 777360 (8016BEF0) responds with value FF07 (hex)
 Address 777362 (8016BEF2) responds with value FF00 (hex)
 Address 777364 (8016BEF4) responds with value FFFF (hex)
 Address 777366 (8016BEF6) responds with value FF00 (hex)
 Address 777370 (8016BEF8) responds with value FF0F (hex)
 Address 777372 (8016BEFA) responds with value FFFF (hex)
 Address 777374 (8016BEFC) responds with value FFFF (hex)
 Address 777376 (8016BEFE) responds with value FFFF (hex)
 Address 777400 (8016BF00) responds with value FF94 (hex)

 Address 777500 (8016BF40) responds with value 0020 (hex)
SYSGEN>