Subject:
|
Re: Color Definition for Dull Silver
|
Newsgroups:
|
lugnet.cad.dev, lugnet.cad.ray
|
Date:
|
Wed, 24 Sep 2003 15:41:06 GMT
|
Viewed:
|
349 times
|
| |
|
|
In lugnet.cad.dev, Steve Bliss wrote:
> In lugnet.cad.dev, Don Heyse wrote:
> > I agree with Cary and Steve that dithering should go.
> Actually, I want to keep dithering.
Ok, I was agreeing with your proposal to make it optional and relegate
it to the end of the line with the MATERIAL properties. And hopefully
it'll never be used, except possibly to reproduce a texture that's
stippled in real life like the sharky crater baseplate.
> > The renderer should be able to convert from true color to dithered
> > colors if the display is underpowered. If you must have it for some
> > special effect like pearlescent sprinkles, then add it to the material
> > section. So colors should be 0xAARRGGBB because that's the way
> > I like to parse them. AA should be somewhere between FF and 00 for
> > transparent colors. (perhaps 80?)
>
> I'm OK with AARRGGBB, if everyone agrees with that.
>
> I'd recommend that programs recognize a variety of semi-standard
> formats, specifically:
> #AARRGGBB
> 0xAARRGGBB
> AARRGGBB
> INDEX(nnn) // LDraw color index
>
> If AA is not specified, it should be assumed to be 255 (ooo, tough
> one - have to differentiate #00FFFFFF from #FFFFFF).
>
> For the sake of the ldconfig.ldr, I recommend we pick one format to
> 'recommend' -- probably #AARRGGBB, since it's the shortest and more
> people (I suspect) are familiar with #RRGGBB from HTML.
I still prefer the "one true format" 0xAARRGGBB that's already in use
by ldraw programs. Who wants to write 3 or 4 different parsing
routines? Ldraw is not HTML or XML so I don't buy that argument.
> > I think the true color edge colors need the high bits to distinguish
> > them from color indices, so I vote for 0x2RRGGBB for that.
>
> I'd rather default to straight ARGB values, and have a special
> notation (ie, INDEX()) to flag when an LDraw code is used instead of
> a color value.
I'm not sure I like the INDEX() thing. I don't want to define
anything but edge colors with an index, and unless you want
transparent edge colors, you should be able to use the AA bits to
distinguish a truecolor edge from an index. If the index is over
255 you should be using a truecolor. Besides, we're already
doing it this way.
> > Multiple material properties are separated by the word MATERIAL.
> >
> > Contrived examples with completely bogus colors.
> >
> > 0 // Index ColorName EdgeColor Color ARGB Materials
> >
> > 0 COLOURDEF 33 TransBlue 0x2002266 0x800020A0 MATERIAL transparent
>
> Other than the see-throughness, does transparent have different
> material properties than standard opaque?
Probably not, but I suppose if they used different sorts of plastic
you might have different types of "transparent" with different indices
of refraction or other optional properties
> > 0 COLOURDEF 256 RubberBlack 0x2000000 0xFF212121 MATERIAL rubber
> > 0 COLOURDEF 334 ChromeGold 15 0xFFC40026 MATERIAL metallic SHININESS
> > 15
> > 0 COLOURDEF nnn PearlyTan 0x2808080 0xFFC40026 MATERIAL pearly 0x80123456
>
> In general, I agree with this. We'll want to be careful with
> MATERIAL values; 'metallic' and 'pearly' can be confused.
I was thinking pearly meant a translucent color with embedded bits of
some metallic color, hence the second ARGB color. But yeah, the basic
material properties need to be sorted out.
> And, I want an optional DITHER key:
> 0 COLOURDEF 64 OddLilac index(8) #EEFFFFFF DITHER #EEFF00FF
>
> Does it serve a purpose? Not really. Does it hurt to support it? Not really.
> If you'd rather propose a STIPPLE attribute, I'm open to that.
I don't mind *allowing* an optional dither key, but I would hope it
wouldn't be used too much. It would make the shark craterplate easier
to model though.
Don
|
|
Message has 1 Reply:
 | | Re: Color Definition for Dull Silver
|
| How about dropping the AA ? So far, IIRC, no-one has ever used it. If another value than the default (0x80?) is wanted, it can be put into the MATERIAL section: transparent=25% The we will use the familiar LDraw color codes for both EdgeColor and (...) (21 years ago, 25-Sep-03, to lugnet.cad.dev, lugnet.cad.ray)
|
Message is in Reply To:
| | Re: Color Definition for Dull Silver
|
| (...) Actually, I want to keep dithering. (...) Shrug. I dunno. (...) I'm OK with AARRGGBB, if everyone agrees with that. I'd recommend that programs recognize a variety of semi-standard formats, specifically: #AARRGGBB 0xAARRGGBB AARRGGBB (...) (21 years ago, 24-Sep-03, to lugnet.cad.dev, lugnet.cad.ray)
|
26 Messages in This Thread:
 
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
This Message and its Replies on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
