30. Make it possible to take the eyedropper outside the app, to sample
images on the desktop, other programs, etc.;
29. Support for Pantone colours (important for colour accuracy with
print houses);
28. A way to choose the colours in your palette if you are reducing
colours in the Gif Optimizer... or at least to weight the colours prior
to reducing. Perhaps just a hook to the manual Colour Reduction dialog;
27. How about a floating toolbar for Recently Used Colours? Make it
expandable up to a full 256-colour .pal file and it would be a treat
(and yes, I know Porter's trick - this would just be built in);
26. And how about a double selection for the Clone tool, allowing you to
constrain both the "Clone from" and "Clone to" areas so you don't stray
outside the area you want by accident?;
25. A particle type generator - you could make particle "seeds" the
same way you make tubes, but they would be a single image. The particle
engine would take care of variations in size, hue, spacing, rotation,
opacity etc. and distribute them over a selection or the entire image -
a quick way to add sparkles, bubbles, fireworks, etc. to an image. Add a
random seed so the patterns aren't predictable, and you have a winner;
24. How about a "force seamless" option on some of the filters? I
noticed EyeCandy 2000 has this, and it makes tileable effects _much_
easier;
23. A simple way to match colours between two images - perhaps an
extension of the Manual Colour Correction feature;
22. An "Autostitch" Wizard for joining photos; and
21. Colour halftones.
Too much? Honestly, these really are the result of brainstorming. Of
all of them, the only one that I am firmly committed to is a macro
function mentioned in the first list - the rest range from "Gee, that
would make things easier" to ""Hmm, that would be cool".
Porter - you've created a monster.
Jackie
--
< o \"/ Don't play cat and mouse with me! (
---© ) ()-()
< o /"\ Jackie Laderoute jflad...@home.com (o o)
************************************************************/\o/\
Yup, I have someone rubbing two sticks together already. Should
be able to light the tinder by Friday if it stays dry :) Seriously
though, keep the ideas coming.
> 30. Make it possible to take the eyedropper outside the app, to sample
> images on the desktop, other programs, etc.;
>
> 29. Support for Pantone colours (important for colour accuracy with
> print houses);
Expensive because proprietary.
> 28. A way to choose the colours in your palette if you are reducing
> colours in the Gif Optimizer... or at least to weight the colours prior
> to reducing. Perhaps just a hook to the manual Colour Reduction dialog;
>
> 27. How about a floating toolbar for Recently Used Colours? Make it
> expandable up to a full 256-colour .pal file and it would be a treat
> (and yes, I know Porter's trick - this would just be built in);
>
> 26. And how about a double selection for the Clone tool, allowing you to
> constrain both the "Clone from" and "Clone to" areas so you don't stray
> outside the area you want by accident?;
>
> 25. A particle type generator - you could make particle "seeds" the
> same way you make tubes, but they would be a single image. The particle
> engine would take care of variations in size, hue, spacing, rotation,
> opacity etc. and distribute them over a selection or the entire image -
> a quick way to add sparkles, bubbles, fireworks, etc. to an image. Add a
> random seed so the patterns aren't predictable, and you have a winner;
Isn't this more an animation thing rather than still image deal?
> 24. How about a "force seamless" option on some of the filters? I
> noticed EyeCandy 2000 has this, and it makes tileable effects _much_
> easier;
>
> 23. A simple way to match colours between two images - perhaps an
> extension of the Manual Colour Correction feature;
As far as I'm concerned, you can do it now. What's wrong with
the current arrangements? (Note: run MCC on Image 1, move cursor
off dialog and see it turn into an eye dropper, pick color from
Image 2. Voila - color match. Make a selection on Image 1 first
if you want the correction to apply only to a region of the image.)
> 22. An "Autostitch" Wizard for joining photos; and
>
> 21. Colour halftones.
More detail please. The word "halftones" means different things
to different people. Perhaps an example or two?
>
> Too much? Honestly, these really are the result of brainstorming. Of
> all of them, the only one that I am firmly committed to is a macro
> function mentioned in the first list - the rest range from "Gee, that
> would make things easier" to ""Hmm, that would be cool".
>
> Porter - you've created a monster.
Perhaps, simply, a previously invisible monster has become
visible :)
>
> Jackie
>
> --
> < o \"/ Don't play cat and mouse with me! (
> ---© ) ()-()
> < o /"\ Jackie Laderoute jflad...@home.com (o o)
> ************************************************************/\o/\
Kris
----------------------------------------------------------------------
Kris Zaklika Jasc Software, Inc.
Phone: (952) 934-8888 ext 544 e-mail: kzak...@jasc.com
----------------------------------------------------------------------
>29. Support for Pantone colours (important for colour accuracy with
>print houses);
Good idea. Follow Adobe and their emphasis on print (least, to date).
>26. And how about a double selection for the Clone tool, allowing you to
>constrain both the "Clone from" and "Clone to" areas so you don't stray
>outside the area you want by accident?;
They'd prob. implement it, though, with only that on the menu, making
the default behavior now more inaccessible to a newbie, and boosting
the learning curve even more? PSP didn't used to be known for a
'learning curve' - Adobe did.
>24. How about a "force seamless" option on some of the filters? I
>noticed EyeCandy 2000 has this, and it makes tileable effects _much_
>easier;
Wrap should be a check box on every such menu, I would think. Agree
with you. I've been suggesting this for a long while. But then I tend
to seamless tiles, myself - www.scenic-route.com
One problem, say for tiles, is that I believe the tiles are 'randomly
seeded', or whatever. It would take more programming effort to make
them seamless. Flood an area with a seamless tile, now, apply a tile
filter, and if you cut the same size anywhere from the result, it
won't nec. be seamless. Still, certain filters might lend themselves
more easily to seamlessness, and Jasc could start with those?
>21. Colour halftones.
>Too much?
You didn't post 1-20. But I would suggest a few obv.
Of course - macros.
Also, a selection selector, just like an object selector. Say you went
with the wand, and your 'marching ants' rectangle is just two pixels
to the left, too thin, and you want to 'march' that right side only
two pixels to the right; expand horiz. And so on.
Batch processing of frames from Animation Shop. Amazingly, no ability
to do this, now. But it could be solved, indirectly, by allowing gang
changes to every layer in a group - unless this is now possible,
somehow.
Crash protection. Cool Edit Pro, for ex., is able to try and restore
the workspace if the program crashes out. You restart, and whatever
was dynamically saved is restored. Now, I realize PSP makes EXTENSIVE
use of temp memory (I once saw over 1000 little files there after
using the retouch tool for about 10 minutes; unless that was not
normal and rather the result of some PSP error). Perhaps all that
would have to be saved are not the undo, or command history, new
unsaved settings for filters, but just the images open in the
workspace - and that's it?
Background removal filter. These seem to be getting increasingly
popular, and very expensive. And perhaps Jasc could take a shot at one
of their own for PSP?
Check the post 15 minutes after yours.
--
Bill - PSP 7, PSP 6, and Media Center Plus Private Beta Tester
PSP Terrorist - D'Lanok de Caresk chapter - Anti-Troll Unit 235
"If you're not making waves, you're not underway!"
--------------------------------------------------------------
The new PSP 7 Style Palette:
http://www.frontiernet.net/~willshak/style_palette/
--------------------------------------------------------------
The USS Salem, CA-139. The World's only preserved Heavy Cruiser,
Visit at Quincy, MASS http://www.frontiernet.net/~willshak/salem/
I'm curious. Why is it good to follow Adobe? Why not do something
else, perhaps sufficiently different to have them follow us for
a change?
[snip]
>
> Batch processing of frames from Animation Shop. Amazingly, no ability
> to do this, now. But it could be solved, indirectly, by allowing gang
> changes to every layer in a group - unless this is now possible,
> somehow.
I believe you just make every frame a separate layer in a PSP
image.
[snip]
>
> Background removal filter. These seem to be getting increasingly
> popular, and very expensive. And perhaps Jasc could take a shot at one
> of their own for PSP?
Could you please explain what you mean by "background removal"?
Do you want the program to determine what in an image is of
interest to you and assign everything else to background? This
seems beyond the current state of the art in image understanding.
Also, would you please give specific examples of "popular and
expensive filters" for this?
--
Joe Cilinceon
http://newdawn.gzinc.com/
"Kris Zaklika" <kzak...@jasc.com> wrote in message
news:3A6371AB...@jasc.com...
Psssst! Don't turn around!
> Mark Johnson wrote:
> >
> > Jackie Laderoute <jflad...@home.com> wrote:
> >
> > >29. Support for Pantone colours (important for colour accuracy with
> > >print houses);
> >
> > Good idea. Follow Adobe and their emphasis on print (least, to date).
>
> I'm curious. Why is it good to follow Adobe? Why not do something
> else, perhaps sufficiently different to have them follow us for
> a change?
Color management would be nice, and would allow me to use PSP for
professionally printed graphics. However, I think that you stated the
reason this hasn't been done in an earlier post (and as I'd always
suspected). My guess is that Pantone Corp (or whatever they're called) want
a pretty penny or two for the license, and the price of PSP would soar.
> [snip]
> >
> > Batch processing of frames from Animation Shop. Amazingly, no ability
> > to do this, now. But it could be solved, indirectly, by allowing gang
> > changes to every layer in a group - unless this is now possible,
> > somehow.
>
> I believe you just make every frame a separate layer in a PSP
> image.
I once had to brighten the area within the same selection on every frame of
an animation. The only way I could think to do it was to Export to PSP as
individual images, select each image one at a time, load the selection from
disk, brighten, and save. I couldn't think of a way to do that to each of
the layers in a PSP file. I believe that the batch processing mentioned
above (hmmmm, sounds like applying a macro in batch processing), or applying
some changes to all layers in a group would allow me to do that.
> > Background removal filter. These seem to be getting increasingly
> > popular, and very expensive. And perhaps Jasc could take a shot at one
> > of their own for PSP?
>
> Could you please explain what you mean by "background removal"?
Is that like changing your identity? New birth certificate, Social Security
number, ...?
--
**********************************************
Chuck Anderson • Boulder, CO CycleTourist at
http://www.CycleTourist.com home.com
Tolerance is recognizing that other people have different ideals
and needs than you. Compromise is acting on that knowledge.
***********************************************************
Witness Protection Program?
.............a bunch of cool ideas snipped
> 27. How about a floating toolbar for Recently Used Colours? Make it
> expandable up to a full 256-colour .pal file and it would be a treat
> (and yes, I know Porter's trick - this would just be built in);
Your built-in is a better idea, since we could always still open fakes too.
I keep fantasizing about color swatching skins
http://people.we.mediaone.net/campratty/mcc.html available on some of the
larger dialogs like Manual Color Correction, Drop Shadow, and some of the
special effects boxes where we could color pick right off the front of those
dialog boxes without having to give up deskspace to fake palettes. Colors
swatching dialog box skins that could be turned off/on through .... (Start
clutching your heart Joe, she's about to say it) ... Preferences.
> Porter - you've created a monster.
Better wishful enthusiasm where people keep thinking,
rather than drifting along waiting for Page Curl Guy to speak for them.
Sure, we may not get all of it (or even any of it) but a few well thought
out
hopes & fantasies sure beats the heck out of boredom & resignation. ;)
Ideas I have - the ability to make them more than a reality, no.
>> 29. Support for Pantone colours (important for colour accuracy with
>> print houses);
>
>Expensive because proprietary.
How expensive, though? I know XaraX has had this support since version
1.5 (http://www.xara.com - the PSP of the vector world, in my opinion).
>> 25. A particle type generator - you could make particle "seeds" the
>> same way you make tubes, but they would be a single image. The particle
>> engine would take care of variations in size, hue, spacing, rotation,
>> opacity etc. and distribute them over a selection or the entire image -
>> a quick way to add sparkles, bubbles, fireworks, etc. to an image. Add a
>> random seed so the patterns aren't predictable, and you have a winner;
>
>Isn't this more an animation thing rather than still image deal?
Not necessarily - take a look at Ulead's Particle plugin
(http://www.webutilities.com/products/FXRazor/runme.htm). I'm thinking
of something like that, but with the ability to set the base image.
>> 23. A simple way to match colours between two images - perhaps an
>> extension of the Manual Colour Correction feature;
>
>As far as I'm concerned, you can do it now. What's wrong with
>the current arrangements? (Note: run MCC on Image 1, move cursor
>off dialog and see it turn into an eye dropper, pick color from
>Image 2. Voila - color match. Make a selection on Image 1 first
>if you want the correction to apply only to a region of the image.)
That's what I do now, but it would be nice to have the "before" and
"after" images replaced by "Image 1" and "Image 2" in this situation -
allow you to zoom in on various areas of each, see them side by each.
Not a huge thing, more of a usability thing (Hey - whatever happened to
Don?).
>> 21. Colour halftones.
>
>More detail please. The word "halftones" means different things
>to different people. Perhaps an example or two?
I'm not talking about true print halftoning here, I'm talking about the
coloured dot generators found in Photopaint and PS. Again, of limited
use - except when it's exactly what you need.
>> Porter - you've created a monster.
>
>Perhaps, simply, a previously invisible monster has become
>visible :)
Oh, sure - how quickly they turn on you <g>.
Jackie "monster" Laderoute
I'm thinking of it being in the tool options for the Clone tool; default
behaviour wouldn't change, and the option would be there for more
demanding users.
>>24. How about a "force seamless" option on some of the filters? I
>>noticed EyeCandy 2000 has this, and it makes tileable effects _much_
>>easier;
>
>Wrap should be a check box on every such menu, I would think. Agree
>with you. I've been suggesting this for a long while. But then I tend
>to seamless tiles, myself - www.scenic-route.com
Nice site.
>One problem, say for tiles, is that I believe the tiles are 'randomly
>seeded', or whatever. It would take more programming effort to make
>them seamless. Flood an area with a seamless tile, now, apply a tile
>filter, and if you cut the same size anywhere from the result, it
>won't nec. be seamless. Still, certain filters might lend themselves
>more easily to seamlessness, and Jasc could start with those?
Hey - I didn't say I could code them... all I can do is dream <g>.
>>21. Colour halftones.
>
>>Too much?
>
>
>You didn't post 1-20. But I would suggest a few obv.
Sure did - check out
http://www.deja.com/[ST_rn=ps]/getdoc.xp?AN=714811692&fmt=text
- 20 in all, now up to 30 - plus a couple of throwaways in the
selections thread.
(Don't worry, Kris - I'm back to work tomorrow.)
Jackie "monster" Laderoute
>Mark Johnson wrote:
>> Jackie Laderoute <jflad...@home.com> wrote:
>> Batch processing of frames from Animation Shop. Amazingly, no ability
>> to do this, now. But it could be solved, indirectly, by allowing gang
>> changes to every layer in a group - unless this is now possible,
>> somehow.
>I believe you just make every frame a separate layer in a PSP
>image.
Well . . but it's tedious to reapply an assortment of filters to each
linked frame, in turn - when it might be accomplished at once. But
then that might complicate an undo for that? I don't know.
>> Background removal filter. These seem to be getting increasingly
>> popular, and very expensive. And perhaps Jasc could take a shot at one
>> of their own for PSP?
>Could you please explain what you mean by "background removal"?
I don't know, here. Knockout - MaskPro - AutoMask - etc? Plus all the
techniques for Photoshop, PSP, PhotoPaint, etc. ? etc?
Maybe I don't understand your question.
>Do you want the program to determine what in an image is of
>interest to you and assign everything else to background? This
>seems beyond the current state of the art in image understanding.
>Also, would you please give specific examples of "popular and
>expensive filters" for this?
Well, a bunch of them. And they're expensive. It might be interesting
to see a Jasc stab at this, incl. with PSP.
-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----== Over 80,000 Newsgroups - 16 Different Servers! =-----
Don (Mr. Usability) Crenshaw is no longer with us. He left
shortly after the release of PSP 6.
[snip]
OK, I've got you now. You want something that will assist you
in cutting an object from an image, but you're not interested in
automatically determining what is the subject and what is not
(i.e. what is background). My confusion stemmed from the fact
that some of our users expect PSP to know what is background.
Tom
"Kris Zaklika" <kzak...@jasc.com> wrote in message
news:3A646D4D...@jasc.com...
You can already do this in PSP 7.0x.
Hover your cursor over a solid swatch on the color palette and press the
Ctrl key down. Keep the Control key down as you pass your cursor over your
desktop or other applications. Click when you have the color you want.
Voila!
--
Nancy Peterson
Quality Assurance, Jasc Software
"Jackie Laderoute" <jflad...@home.com> wrote in message
news:al366t8e0uv751bo3...@24.66.95.16...
>> 30. Make it possible to take the eyedropper outside the app, to sample
>> images on the desktop, other programs, etc.;
>
>You can already do this in PSP 7.0x.
>Hover your cursor over a solid swatch on the color palette and press the
>Ctrl key down. Keep the Control key down as you pass your cursor over your
>desktop or other applications. Click when you have the color you want.
>Voila!
Hmmm - mine only works within the PSP app window. I lose the eyedropper
as soon as I try to leave it... anyone else?
Same here. Nancy must be thinking of PSP 8. :-)
>
> Jackie
>
> --
> < o \"/ Don't play cat and mouse with me! (
> ---© ) ()-()
> < o /"\ Jackie Laderoute jflad...@home.com (o o)
> ************************************************************/\o/\
--
Well, I've seen Nancy do it on her computer. I failed on mine
and Nancy failed on mine, too. Perhaps mine is possessed by
demons. We will look into this, however.
>
>> 30. Make it possible to take the eyedropper outside the app, to sample
>> images on the desktop, other programs, etc.;
>
>You can already do this in PSP 7.0x.
>Hover your cursor over a solid swatch on the color palette and press the
>Ctrl key down. Keep the Control key down as you pass your cursor over your
>desktop or other applications. Click when you have the color you want.
>Voila!
>
Does not work for me.
Barb
Ditto over here Jackie.
Jo
Works like Nancy says for me. NT4 and unpatched psp7.
select dropper tool - move cursor onto color swatch or styles swatch
-push and hold down the ctrl key-move the cursor to other apps and pick
colors.
Chunk Kiesling
> On Thu, 18 Jan 2001 10:15:15 -0600, "Nancy \(Jasc Software\)"
> <npet...@jasc.com> wrote:
>
> >> 30. Make it possible to take the eyedropper outside the app, to sample
> >> images on the desktop, other programs, etc.;
> >
> >You can already do this in PSP 7.0x.
> >Hover your cursor over a solid swatch on the color palette and press the
> >Ctrl key down. Keep the Control key down as you pass your cursor over your
> >desktop or other applications. Click when you have the color you want.
> >Voila!
>
> Hmmm - mine only works within the PSP app window. I lose the eyedropper
> as soon as I try to leave it... anyone else?
I can't get it to work outside the PSP window either.
And if I pass the cursor over the current dropper color swatch, it leaves
trails of the little popup window that follows the cursor in the side bar and
in the browser window.
> >
> > 29. Support for Pantone colours (important for colour accuracy with
> > print houses);
>
> Expensive because proprietary.
How about CMYK support instead as CMYK equivalents to the Pantone colours
are easily obtainable but RGB equivalents are not.
I had this problem recently with our corporate logo which must be printed
using the design guidelines specified by the graphics consultants who
designed the corporate branding.
Like most print houses they specified Pantone colours together with their
CMYK equivalents.
Among other things I was set the task of updating this guide to include web
graphics which necessitated finding out RGB equivalents of the CMYK values.
The only way I could find the correct RGB values was to avail of someone
else with Photoshop that can use either CMYK or RGB.
Would it be possible to get PSP to toggle between either set of values for a
particular colour and display either CMYK or RGB?
Stephen
I agree. This is a good idea, and would bring PSP a step closer to commercial
printing.
>Perhaps it's to do with NT. Won't work on my Win98 machine on
>anything not PSP though it will grab colour from PSP palettes parked
>outside the PSP window.
That latter must have been what she meant. For ex., you can resize PSP
and leave one of the options windows outside on the 'desktop', and use
the dropper to pick a color there. But you can't do it with any other
windows. Have to be PSP windows.
> > Nancy of Jasc wrote:
> > >You can already do this in PSP 7.0x.
> > >Hover your cursor over a solid swatch on the color palette and press the
> > >Ctrl key down. Keep the Control key down as you pass your cursor over your
> > >desktop or other applications. Click when you have the color you want.
> > >Voila!
> Works like Nancy says for me. NT4 and unpatched psp7.
> select dropper tool - move cursor onto color swatch or styles swatch
> -push and hold down the ctrl key-move the cursor to other apps and pick
> colors.
AHA! Yes, it does somewhat. Win98 original and PSP 7.01
Like you said.
Pick Eyedropper Tool
Move over to the color palette (no need to click on anything)
when the cursor changes to the dropper on the color palette, press the
Ctrl button and then move all over the PSP screen, PSP toolbars, title
bar, etc. and watch the active color swatch change colors. Click a mouse
button to save the color.
But... This only works in the PSP window, though. It won't work if I
resize the PSP window and try to get a color from another source sitting
under the PSP window. Netscape, etc., which is what the original poster
wanted.
nor for me under win95 ...
--
mairi macdonald
Gary
>>> 30. Make it possible to take the eyedropper outside the app, to sample
>>> images on the desktop, other programs, etc.;
>>
>>You can already do this in PSP 7.0x.
snipped
Kris Zaklika wrote:
>
> Jackie Laderoute wrote:
> >
> >
> > Hmmm - mine only works within the PSP app window. I lose the eyedropper
> > as soon as I try to leave it... anyone else?
> >
> > Jackie
**********************************************/\o/\
>
> Well, I've seen Nancy do it on her computer. I failed on mine
> and Nancy failed on mine, too. Perhaps mine is possessed by
> demons. We will look into this, however.
--
Angela M. Cable
http://www.neocognition.com
PSP Tutorial Links:
http://www.geocities.com/Heartland/Plains/9871/PSPlinks.html
5th Street Studio, free graphics, websets and more:
http://www.fortunecity.com/westwood/alaia/354/
BladePro Visual Archive:
http://www.crosswinds.net/~angeal/
No, Nancy meant what she said. It works sometimes and not other
times on some machines and not others, I'm afraid.
>
> -----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
> http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
> -----== Over 80,000 Newsgroups - 16 Different Servers! =-----
--
> I can't get it to work outside the PSP window either.
>
> And if I pass the cursor over the current dropper color swatch, it
leaves
> trails of the little popup window that follows the cursor in the side
bar and
> in the browser window.
Mine did that too Chuck....leaving trails of that little RGB info popup
all over my Color bar as I moved the eye dropper around. The cursor
even got stuck for a few moments in dropper mode, not changing back to a
regular cursor, even when over the desktop like it should have.
Clicking the mouse buttons did nothing...it was just stuck. I waited
for a bit and it seemed to correct itself, so not sure what happened
there.
Using Win98 and PSP 7.01
Jo
Kris
>Are the two of you running the same OS? I seem to remember from the
>beta testing that it works under Win2K and WinNT, but not Win95 or
>Win98. I'm running Win98SE, does not work.
>
Running Win98SE and does not work.
Barb
> But which CMYK for which ink set assuming which dot growth?
I did a job for some friends a year or so ago, making business cards. I took it
to a service bureau with a Heidelberg, five ink, electronic press (no plates
necessary - I want one!). That really makes things easy as you can use process
colors (CMYK) to come up with about anything you want (or spot colors like
Pantone). The fifth "color" can be a lacquer (or a spot color).
Learning things the hard way, I first brought them a tiff file made in psp. No
good. They needed the colors in CMYK or Pantone. I looked at their pantone chips
(which also show the closest CMYK equivalent) and found a close match to the color
of the cards I was making. I came back to psp and . . . . . . . . . was at a
loss. How do I plug in the correct CMYK color?
After another lesson or two down the road (they needed a post script file to get
rid of jaggies on text), I gave up and bought Illustrator 8.0. I now have a very
expensive CMYK to RGB converter :-)
I don't know what ink set I used or which dot growth, but when I plugged in the
correct CMYK combination I got what I was looking for.
The addition of CMYK is probably not enough to get to the service bureau, though,
as you pretty much have to use the same program as they do, and bring a file in
it's native format. Still, I find myself occasionally opening Illustrator for no
other reason than to plug in a CMYK color and the see what the RGB values are.
That's the reason for my, "me too" for supporting CMYK color.
--
You may want one, but you can't afford it :)
[snip]
> The addition of CMYK is probably not enough to get to the service bureau, though,
> as you pretty much have to use the same program as they do, and bring a file in
> it's native format. Still, I find myself occasionally opening Illustrator for no
> other reason than to plug in a CMYK color and the see what the RGB values are.
I think this may be the most telling comment of all. It is very
hard to overcome that issue.
>
> That's the reason for my, "me too" for supporting CMYK color.
[snip]
> **********************************************
> Chuck Anderson • Boulder, CO CycleTourist at
> http://www.CycleTourist.com home.com
> Tolerance is recognizing that other people have different ideals
> and needs than you. Compromise is acting on that knowledge.
> ***********************************************************
Kris
Err.... you've got me there (or roughly translated "I haven't a clue what
you're talking about") :)
Do I take it you're saying it's more complicated than I assume?
I know there are difficulties in translating colours in that it's impossible
(or relatively so) to create pure ink colours. This means that the additive
process doesn't always produce the colour you'd think it would e.g. mix
equal quantities of cyan, magenta and yellow inks and you don't get black
you get brown (which BTW is one of the reasons fro the inclusion of the K or
black ink in the equation).
I was presuming the fact that it's available in other software meant that
this would be achievable in PSP or have I completely got the wrong end of
the stick?
BTW you've whetted my curiosity now, what's the significance of (and what
is) dot growth?
Is this related to colour bleeding on the final printed page?
Sorry for all the questions and thanks for your reply, but I'm not exactly
an expert on printing processes although I'm sure it's a bit of an arcane
art to most PSP users as well.
Stephen
I hesitate to say anything here because I'm rather ignorant of
the whole topic. There are, however, a number of issues.
Some color conversions, like RGB to YIQ (the NTSC television color
space), are just a simple matrix multiplication and you can easily
convert from one color space to another. Others, like RGB to HSL
are done according to certain rules and again are simply a one to
one mapping (except for black, white and grey, which have undefined
hue). When you convert from RGB to CIE L*a*b* you need to introduce
one more concept, that of the color of the light source or the white
point, but again it's basically a matter of matrices and rules.
The big problem is that in RGB to CMYK conversion you are going
from 3 colors to 4, not 3 to 3. Because you end up with more colors
than you start with, there is no unique way to map RGB onto CMYK.
The conversion problem is underconstrained and you need to get
some additional constraints from somewhere to solve this problem.
Usually, this additional information comes from empirical knowledge
about how printing presses behave.
The next issue to tackle is that RGB colors are emissive colors
(i.e. your monitor phosphors), while CMYK colors are inks. These
absorb light and modify the reflectivity of the paper they are on
to produce the effect of color, e.g. yellow ink absorbs blue and
red but allows the yellow component of the white light to bounce
off the paper and into your eye, producing a yellow sensation.
The particular yellow you see thus depends on the type of paper
and its reflectivity characteristics, not just on the ink color.
(It also, of course, depends on the light you are using to
illuminate the paper with, e.g. daylight, fluorescent or
incandescent light.) Further, the range of colors (also known
as gamut) that you can display in RGB is different than the
gamut of CMYK. There is a significant overlap of the gamuts but
some colors in one space simply cannot be reproduced in the other.
This requires you to somehow deform the gamut of one space onto
the other so every color in one color space can be represented
in the other. There is no unique way to perform this gamut
mapping. Depending on the intended perceptual result you may
choose to take one approach or another.
Saying C, M, Y or K ink is itself not a unique description.
Depending on the standards to which the inks conform, the
color of one standard cyan, for instance, might be different
from another. A further complication is that, in order to
reproduce some specific very important color (like Coca Cola
red or Kraft yellow) you may use separate spot colors, or
inks of exactly the right shade, in addition to or in place
of CMYK.
Another thing to consider is the printing plate itself. You
can think of it as a sort of grid of squares (called the line
screen). Each of these squares can contain a dot with different
surface properties from the rest of the square. Because of this
the dot picks up the ink and deposits it on the paper. When the
dot is very small (say 5% of the square) we have a pastel color,
mostly white paper with a speck of ink. When the dot is, say,
95% of the square, we have a deep, strongly saturated color.
There is a separate plate for each of the 4 colors, with the
grids running at angles to each other to avoid moire patterns.
However, the size of the ink dot on the plate is not the same
as the size of the ink dot on the paper. A variety of factors
contribute to this, but you can imagine ink spreading readily
when printed on coarse absorbent paper. Thus you can imagine a
dot growth process relative to the dots on the plate. In fact,
this growth is not a constant percentage of the dot size but
varies with the size of the dot. (It can get even more wild
with stochastic screening, which doesn't use a regular grid.)
So, what's the bottom line? I don't know exactly, but there is
a whole pile of complications. It's not possible to make a
single button labeled "Convert to CMYK" because the conversion
depends on what kind of printing will be done later. The user
has to understand that and take it into account. This is, perhaps,
one reason why printers are so wedded to a particular set of
programs that they know well and understand. (If you have a bit
of time on your hands, run that "expensive program" and take a
look at the choices under File > Color Settings > CMYK Setup,
then explain them to me :)
Actually the more I think about this the more I wonder.
If PSP can already produce CMYK separations from RGB then obviously JASC
have found an acceptable method of doing this.
So why can't it use the same method in reverse to translate CMYK to RGB?
Stephen
In PSP 7 you can convert from RGB to CMYK using Colors >
Split Channel > Split to CMYK. Also you can go from CMYK
to RGB using Colors > Combine Channel > Combine from CMYK.
So, albeit indirectly, you can translate to and from RGB
and CMYK. All you have to do is put in the correct numbers
for your service bureau's printing process in File >
Preferences > CMYK Conversion Preferences > Modify.
Doesn't have to be psp windows on my machine. (NT4)
Chunk
Yeah, prob. an NT thing. She probably just wasn't consciously aware of
the OS being used - so thought it just worked that way across OS and
platforms, in general.
We will definitely look into it, but it's probably not major enough to mess
with in an .02 update.
One clarification on my process is that PSP is not maximized when I first
click and hold the control key down on the color palette. Which as it turns
out, still may or may not work for you, since I've seen this feature fail on
Kris's machine, with the exact steps that work on mine.
--
Nancy Peterson
Quality Assurance, Jasc Software
"Mark Johnson" <1023...@compuserve.com > wrote in message
news:f4tm6t08usnhp144n...@4ax.com...
It's an eyedropper tool from Inertia.com
http://eyedropper.inetia.com/HTML/eng/default.asp
I like this because it's a small program - 32KB, and can be used to see what the
values of colors are on screen in RGB, CMYK, and HEX, Plus it's not stuck to the
PSP program, so you can use it on anything on the screen.
It does a few other wiz-bang things too.
Plus it's free. :)
Phoebe
Yes, it's a very interesting little tool. However, you should
be warned that it does not do proper RGB to CMYK conversion.
By proper I mean CMYK as it would appear on press. This must
be the case, as the dialog shown has no options for entering
ink types, or transfer characteristics, or whether to use UCR
or GCR, or anything about gamuts. There is a pretty trivial
algorithm for converting RGB to CMYK (which I suspect is
probably used here) but it would be very misleading to use it
for color matching to press.
Having said this, I don't mean to imply that this tool could
not be useful for other purposes.
Kris
Could you explain this a bit more please? How might this work
in practice? It seems to me that the Octree or Median Cut
algorithms already provide (indirectly) a choice of weighting.
What did you have in mind for applying weights?
[snip]
>Jackie Laderoute wrote:
>>
>[snip]
>> 28. A way to choose the colours in your palette if you are reducing
>> colours in the Gif Optimizer... or at least to weight the colours prior
>> to reducing. Perhaps just a hook to the manual Colour Reduction dialog;
>
>Could you explain this a bit more please? How might this work
>in practice? It seems to me that the Octree or Median Cut
>algorithms already provide (indirectly) a choice of weighting.
>What did you have in mind for applying weights?
I should have been more specific - perhaps a better way to explain this
would be a combination of the ability to pick colours for
inclusion/boosting from within the Export Dialog (I'm forever forgetting
to select the colour and have to cancel out of the dialog to choose it)
-and- the ability to see the resulting palette and perhaps editing the
entries manually before exporting.
For that matter, it would be nice to have a way to pick up a colour for
transparency from the "before" preview, a la Ulead's Smartsaver.
It's all about my control issues, I know.
>Well, these theories are all plausible...except that I am using Windows 98.
>We will definitely look into it, but it's probably not major enough to mess
>with in an .02 update.
Would be neat to nail it down, though. The ability to move to other
'process'/app windows and color pick would be something a lot of
people could use. Maybe it's an 'unexpected feature' based on some
accidental registry setting? But I'm guessing.
>One clarification on my process is that PSP is not maximized when I first
>click and hold the control key down on the color palette. Which as it turns
>out, still may or may not work for you, since I've seen this feature fail on
>Kris's machine, with the exact steps that work on mine.
I can only get the dropper on PSP options windows, and such, in that
test case. If it's Word or something else, can't.
> On Mon, 22 Jan 2001 12:14:35 -0600, Kris Zaklika <kzak...@jasc.com>
> wrote:
>
> >Jackie Laderoute wrote:
> >>
> >[snip]
> >> 28. A way to choose the colours in your palette if you are reducing
> >> colours in the Gif Optimizer... or at least to weight the colours prior
> >> to reducing. Perhaps just a hook to the manual Colour Reduction dialog;
> >
> >Could you explain this a bit more please? How might this work
> >in practice? It seems to me that the Octree or Median Cut
> >algorithms already provide (indirectly) a choice of weighting.
> >What did you have in mind for applying weights?
>
> I should have been more specific - perhaps a better way to explain this
> would be a combination of the ability to pick colours for
> inclusion/boosting from within the Export Dialog (I'm forever forgetting
> to select the colour and have to cancel out of the dialog to choose it)
> -and- the ability to see the resulting palette and perhaps editing the
> entries manually before exporting.
I second being able to see and edit the palette from the Export dialog. That
would make it a lot easier really grind those gif sizes down while maintaining
the colors you want. That's an extra step I usually don't bother with, I just
crank up the number of colors until I have what I want.
--
I use an old copy of LView that has (IMHO) a great background color
selection option. The palette is displayed along with a thumbnail of
the image and the image is masked (selectable with white or black) with
the currently selected background color. Select a new background color
from the palette, the thumbnail mask changes to show you exactly what
parts of the image will be transparent with the new background. Makes
transparent color selection a snap.
--
Rick Simon
rsi...@cris.com
Greetings,
Pantone is included in Painter 5.5, which given away free with
PC Format issue 112 (September 2000). I'm not enthused by the program.
but keep it installed purely for this reason.
Could Pantone exist as a separate plug-in type thing? That way those
of us that want it can go and buy it & those who don't, don't.
IMO PSP (any version) beats the opposition round the head with a
medium-sized mammal any day of the week, a few more steps in the DTP
direction may well prove fruitful.
Regards.
ps: Sava As/Export as *.ai with all the vector info is a real
wantwantwantneedmusthave (IMO).
I love requests made with that characteristic British
restraint :)
Both requests have been added to the pile, which is now
teetering dangerously. It's very tall and the mammals
keep brushing up against it.
I like Canadians too, even when they whine, eh. :)
>
> Ron
>
> *~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*
> Ron Lacey
> Murillo Ont.
> ron at ronstoons dot com
> http://ronstoons.com/
> *~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*~*
>Jackie Laderoute <jflad...@home.com> wrote:
>
>>29. Support for Pantone colours (important for colour accuracy with
>>print houses);
>
>Good idea. Follow Adobe and their emphasis on print (least, to date).
[...]
One of the problems I found with trying to render Pantone colors as
RGB is that their RGB equivalents are frequently fractional. For
example, the RGB values that Pantone gives for Mecca Orange are
expressed on a scale of 100; therefore, they must be converted to
Paint Shop Pro's scale of 0-255:
__Pantone >> PSP___
R: 58/100 >> 148.48
G: 52/100 >> 133.12
B: 00/100 >> 0
You need to round those values up to the nearest whole number (except
for white or black) and enter them into the PSP palette to get the
HTML value of #968500.
Unfortunately, you also get that same value if you enter R=150, G=133,
B=0.
It should be intuitively obvious that RGB: 150, 122, 0 is not the same
as RGB: 149, 134, 0. But, PSP is based on a 24-bit color system while
Pantone is based on a higher one, so the only way that Jasc can
provide Pantone (licensing requirements and costs aside) would be to
change their program to handle greated color depth.
NOTE to Jasc: since more and more scanners are providing higher color
output (30- to 48-bit, have you considered a corresponding change to
PSP?
co
>On Mon, 15 Jan 2001 12:15:12 -0800, Mark Johnson
>One of the problems I found with trying to render Pantone colors as
>RGB is that their RGB equivalents are frequently fractional. For
>example, the RGB values that Pantone gives for Mecca Orange are
>expressed on a scale of 100; therefore, they must be converted to
>Paint Shop Pro's scale of 0-255:
>
>__Pantone >> PSP___
>R: 58/100 >> 148.48
>G: 52/100 >> 133.12
>B: 00/100 >> 0
Strange, my conversion comes up
147.9
132.6
0
for RGB.
Rounding up to
148
133
0
>
>You need to round those values up to the nearest whole number (except
>for white or black) and enter them into the PSP palette to get the
>HTML value of #968500.
I get #948500
>
>Unfortunately, you also get that same value if you enter R=150, G=133,
>B=0.
You made an error in the previous calculation.
You can only have one set of RGB values for a given Hex value. You
cannot have two different hex values giving the same RGB value.
>
>It should be intuitively obvious that RGB: 150, 122, 0 is not the same
>as RGB: 149, 134, 0. But, PSP is based on a 24-bit color system while
>Pantone is based on a higher one
Eh? how is that again? You're representing more colors with a range
of 0 to 100 than 0 to 255? Neat trick! Is it done with mirrors?
>so the only way that Jasc can
>provide Pantone (licensing requirements and costs aside) would be to
>change their program to handle greated color depth.
I think you have it backwards. 24 bit color can already be close to
Pantone colors without modification, all you need to do is toss out
15.7 million colors from the 16.7 million.
>
>NOTE to Jasc: since more and more scanners are providing higher color
>output (30- to 48-bit, have you considered a corresponding change to
>PSP?
That extra byte is used to settle dithering on the LSB, not as an
addition to colors. Even the $800 high priced spread tosses out the
extra 8 bits for graphics editing.
Jaggiemeister Ron- PSP Terrorist - D'Lanok de Caresk chapter, Executive Officer.
Great Beginner info at http://www.jasc-canada-psp-edu.com
FAQs at http://www.alphageo.com/psp/faq.html
Very FAQ at http://members.aol.com/psptopten/topten.html
My tips at http://hometown.aol.com/ronaldlvick/index.html
OK, I'll admit that some math is required to compute 58 * 255 / 100,
though I'm not sure this qualifies as a "problem". Moreover, it
could be argued that the problem is with the software that insists
on giving you Pantone RGB values in an unconventional way, not
with PSP.
> You need to round those values up to the nearest whole number (except
> for white or black) and enter them into the PSP palette to get the
> HTML value of #968500.
Actually, not quite. You need to round values less than nnn.5
_down_ to nnn and values of nnn.5 or greater _up_ to nnn.
> Unfortunately, you also get that same value if you enter R=150, G=133,
> B=0.
True, but not unfortunate I think. Your monitor cannot display
color with better than one bit resolution (i.e. one part in 256),
nor can your printer. Further, your eye cannot see the difference
between:
R = 148.48, G = 133.12, B = 0.0 and
R = 148, G = 133, B = 0.
A difference that is not difference is not, in fact, a
difference.
> It should be intuitively obvious that RGB: 150, 122, 0 is not the same
> as RGB: 149, 134, 0.
It's not only intuitive. These are two completely different
colors, differing especially in green. They are different to
your eye and different in PSP. (However, I'm not clear where
you got these two sets of numbers and what I'm supposed to
conclude from them.)
> But, PSP is based on a 24-bit color system while
> Pantone is based on a higher one, so the only way that Jasc can
> provide Pantone (licensing requirements and costs aside) would be to
> change their program to handle greated color depth.
I disagree. For example, if Pantone decided to use a 2048-bit
description of color your reasoning would require PSP to have
2048-bit color channels instead of the current 8 bits. My
position would be that it is a complete waste of memory and
disk space storing color information that you cannot see. One
reason Pantone uses more than 8-bits is to avoid loss of
accuracy in complex and repeated integer arithmetic involved
in various manipulations used in color management for
printing (not because you or I can actually see e.g. 16-bits
of color information). This doesn't arise in PSP.
> NOTE to Jasc: since more and more scanners are providing higher color
> output (30- to 48-bit, have you considered a corresponding change to
> PSP?
This would again be an issue of limiting calculation errors,
though there could be some perceptual issues too. One could
certainly consider this, but I don't know what the brains
trust here would ultimately decide. I'm inclined to think that
few users would see any visual effect of this, other factors
such as sub-optimal scanner software settings being much more
important. However, if many users need this....
> co
>
> baseb...@hotmail.com (aka sabr...@my-deja.com)
Kris
Boy, I never thought to actually check his arithmetic. I guess
this math stuff is more of a problem for people than I thought :)
>On Sun, 28 Jan 2001 22:25:03 GMT, baseb...@hotmail.com wrote:
>
>>On Mon, 15 Jan 2001 12:15:12 -0800, Mark Johnson
>
>>One of the problems I found with trying to render Pantone colors as
>>RGB is that their RGB equivalents are frequently fractional. For
>>example, the RGB values that Pantone gives for Mecca Orange are
>>expressed on a scale of 100; therefore, they must be converted to
>>Paint Shop Pro's scale of 0-255:
>>
>>__Pantone >> PSP___
>>R: 58/100 >> 148.48
>>G: 52/100 >> 133.12
>>B: 00/100 >> 0
>
>Strange, my conversion comes up
>
>147.9
>132.6
>0
>for RGB.
>Rounding up to
>148
>133
>0
That will happen if you multiply the fraction 58/100 by 255, but I
though 0 to 255 emcompassed 256 numbers.
>>
>>You need to round those values up to the nearest whole number (except
>>for white or black) and enter them into the PSP palette to get the
>>HTML value of #968500.
>
>I get #948500
This, too, depends on the multipliplier.
>>
>>Unfortunately, you also get that same value if you enter R=150, G=133,
>>B=0.
>
>You made an error in the previous calculation.
>You can only have one set of RGB values for a given Hex value. You
>cannot have two different hex values giving the same RGB value.
While I'm not sure that I made an error in those calculations, I sure
misread the hex number I got. Of course, there has to be a one to one
correspondence between hex and decimal numbers in 24-bit color
schemes. That should have been intuitively obvious to me.
>>It should be intuitively obvious that RGB: 150, 122, 0 is not the same
>>as RGB: 149, 134, 0. But, PSP is based on a 24-bit color system while
>>Pantone is based on a higher one
>
>Eh? how is that again? You're representing more colors with a range
>of 0 to 100 than 0 to 255? Neat trick! Is it done with mirrors?
>
>>so the only way that Jasc can
>>provide Pantone (licensing requirements and costs aside) would be to
>>change their program to handle greated color depth.
>
>I think you have it backwards. 24 bit color can already be close to
>Pantone colors without modification, all you need to do is toss out
>15.7 million colors from the 16.7 million.
If Pantone bases their color scheme on, say, a 36-bit system, then
there will not be a one-to-one correspondence between all their colors
and ones that can be represented by 24-bit words.
>>NOTE to Jasc: since more and more scanners are providing higher color
>>output (30- to 48-bit, have you considered a corresponding change to
>>PSP?
>
>That extra byte is used to settle dithering on the LSB, not as an
>addition to colors. Even the $800 high priced spread tosses out the
>extra 8 bits for graphics editing.
If your graphics program won't handle all the colors.
My HP says it has "36 color output." I took that literally. One of the
Mustek scanners claims to have 48-bit depth with 281 trillion colors.
Picture Window Pro "can read and write 16-bit black and white images
in either TIFF or FITS format, and it can read and write 48-bit color
images in TIFF format." PNG also supports 16-bit color resolution for
each color value (RGB).
This may be a matter of semantics, but there are programs and file
structures that claim to handle higher color depth than PSP. If the
information is there, and increasing it seems to be, then I would like
to see PSP use it.
Thanks for pointing out my error in the hex-decimal RGB
representation.
co
Correct. Observe the following:
binary decimal
0 0
255 100
Therefore, to convert from a decimal range of 0 to 100 into a range of
0 to 255, you convert the decimal range to a percentage, then multiply
by 255. (Using a multiplier of 256 would call for a upper value of
256.)
Note that I'm using the ranges you supplied.
You cannot have a binary value less than 0 or more than 255 for an 8
bit number.
>
>If Pantone bases their color scheme on, say, a 36-bit system, then
>there will not be a one-to-one correspondence between all their colors
>and ones that can be represented by 24-bit words.
However, you made the statement that the Pantone RGB was in the range
of 0 to 100. Such a range can be expressed in binary using 7 bits,
and you'll still have 27 colors outside the pantone range.
>
>>>NOTE to Jasc: since more and more scanners are providing higher color
>>>output (30- to 48-bit, have you considered a corresponding change to
>>>PSP?
>>
>>That extra byte is used to settle dithering on the LSB, not as an
>>addition to colors. Even the $800 high priced spread tosses out the
>>extra 8 bits for graphics editing.
>
>If your graphics program won't handle all the colors.
>
>My HP says it has "36 color output." I took that literally. One of the
>Mustek scanners claims to have 48-bit depth with 281 trillion colors.
>
>Picture Window Pro "can read and write 16-bit black and white images
>in either TIFF or FITS format, and it can read and write 48-bit color
>images in TIFF format." PNG also supports 16-bit color resolution for
>each color value (RGB).
*Snort* Marketing ploys for those that don't know better. Tell me,
isn't that a waste when you cannot display it or print it?
> Boy, I never thought to actually check his arithmetic. I guess
> this math stuff is more of a problem for people than I thought :)
> --------------------
>
For some of us this math stuff is an enormous problem and any thing in
PSP that requires any math at all can be difficult to impossible
Peggy J
>
--
Peggy Jentoft, Energy work Trainer,metaphysics, Reiki manuals online
http://people.we.mediaone.net/skygreen/index.html
Soulstar: Huna, Angels, Crystals,fairies, practical metaphysics
http://www.geocities.com/solarraven/index.html
MysticSparrow free art http://members.xoom.com/Skygreen/index.html
Spirited emotion: http://members.tripod.com/~sunsparrow/index.html
I firmly believe that we should minimize the amount of mental
math a user has to do, if only as a simple convenience. However,
since only adding, subtracting, multiplication and division
are involved I'm genuinely curious why it is so difficult.
It's basically the same math as making change or splitting a
bill - no algebra, calculus or trigonometry here, thank heavens. :)
> >
>
> --
> Peggy Jentoft, Energy work Trainer,metaphysics, Reiki manuals online
> http://people.we.mediaone.net/skygreen/index.html
> Soulstar: Huna, Angels, Crystals,fairies, practical metaphysics
> http://www.geocities.com/solarraven/index.html
> MysticSparrow free art http://members.xoom.com/Skygreen/index.html
> Spirited emotion: http://members.tripod.com/~sunsparrow/index.html
--
----------------------------------------------------------------------
Kris Zaklika Jasc Software, Inc. The
Product Ideas: id...@jasc.com Power
Customer Service: customer...@jasc.com To
Technical Support: tec...@jasc.com Create
----------------------------------------------------------------------
>baseb...@hotmail.com wrote:
>>
>> On Mon, 15 Jan 2001 12:15:12 -0800, Mark Johnson
>> <1023...@compuserve.com > wrote:
>>
>> >Jackie Laderoute <jflad...@home.com> wrote:
>> >
>> >>29. Support for Pantone colours (important for colour accuracy with
>> >>print houses);
>> >
>> >Good idea. Follow Adobe and their emphasis on print (least, to date).
>> [...]
>>
>> One of the problems I found with trying to render Pantone colors as
>> RGB is that their RGB equivalents are frequently fractional. For
>> example, the RGB values that Pantone gives for Mecca Orange are
>> expressed on a scale of 100; therefore, they must be converted to
>> Paint Shop Pro's scale of 0-255:
>>
>> __Pantone >> PSP___
>> R: 58/100 >> 148.48
>> G: 52/100 >> 133.12
>> B: 00/100 >> 0
>
>OK, I'll admit that some math is required to compute 58 * 255 / 100,
>though I'm not sure this qualifies as a "problem". Moreover, it
>could be argued that the problem is with the software that insists
>on giving you Pantone RGB values in an unconventional way, not
>with PSP.
I'm not sure that I agree with you here--perhaps it is something that
I'm overlooking-- but as an example, the color red takes values from
00 to FF in hexadecimal, which translates to 0-255 decimal.
There are 256 numbers from 0 to 255, inclusive, so why isn't the
computation (58/100)*256?
>> You need to round those values up to the nearest whole number (except
>> for white or black) and enter them into the PSP palette to get the
>> HTML value of #968500.
>
>Actually, not quite. You need to round values less than nnn.5
>_down_ to nnn and values of nnn.5 or greater _up_ to nnn.
Yes, tht is the conventional method of rounding, but since I used a
multiplier of 256 in my calculations, the only way I could get their
RGB numbers was to round all fractions up.
This all comes down to which of us was using the correct multiplier.
>> Unfortunately, you also get that same value if you enter R=150, G=133,
>> B=0.
>
>True, but not unfortunate I think. Your monitor cannot display
>color with better than one bit resolution (i.e. one part in 256),
>nor can your printer. Further, your eye cannot see the difference
>between:
>R = 148.48, G = 133.12, B = 0.0 and
>R = 148, G = 133, B = 0.
>A difference that is not difference is not, in fact, a
>difference.
>
>> It should be intuitively obvious that RGB: 150, 122, 0 is not the same
>> as RGB: 149, 134, 0.
>
>It's not only intuitive. These are two completely different
>colors, differing especially in green. They are different to
>your eye and different in PSP. (However, I'm not clear where
>you got these two sets of numbers and what I'm supposed to
>conclude from them.)
[...]
I agree that the eye might not be able to perceive differences in
colors that close together, but I do wonder whether they might not
give a better perceived sharpness to the picture, especially in the
extreme ends of the gamut--more "depth" to the darkness; subtle
gradations in the lightness. Why would a slide scanned at 3600 dpi,
and printed out, produce a better result than if it were scanned at
2400 dpi? There is more information in the first scan, isn't there?
Thanks,
co
co
> >> One of the problems I found with trying to render Pantone colors as
> >> RGB is that their RGB equivalents are frequently fractional. For
> >> example, the RGB values that Pantone gives for Mecca Orange are
> >> expressed on a scale of 100; therefore, they must be converted to
> >> Paint Shop Pro's scale of 0-255:
> >>
> >> __Pantone >> PSP___
> >> R: 58/100 >> 148.48
> >> G: 52/100 >> 133.12
> >> B: 00/100 >> 0
> >
> >OK, I'll admit that some math is required to compute 58 * 255 / 100,
> >though I'm not sure this qualifies as a "problem". Moreover, it
> >could be argued that the problem is with the software that insists
> >on giving you Pantone RGB values in an unconventional way, not
> >with PSP.
>
> I'm not sure that I agree with you here--perhaps it is something that
> I'm overlooking-- but as an example, the color red takes values from
> 00 to FF in hexadecimal, which translates to 0-255 decimal.
>
> There are 256 numbers from 0 to 255, inclusive, so why isn't the
> computation (58/100)*256?
There are 256 numbers from 0 to 255, but there are only 255
intervals between these numbers and you divide by the number
of intervals. Think of this simple analogy: let's say there
are 2 people sharing lunch, one person having lunch, and no
people (i.e. zero people) having lunch. Two of you receive
the bill. There are three numbers from 0 to 2 (0, 1 and 2)
but you split the bill in two (not in three). One of you has
lunch and you divide the bill by what? One, even though there
are two numbers (0 and 1). OK, now none of you has lunch.
You would have to divide the bill by zero, which is an
undefined operation. Fortunately, however, you don't get a
lunch bill if none of you have lunch :)
> >> You need to round those values up to the nearest whole number (except
> >> for white or black) and enter them into the PSP palette to get the
> >> HTML value of #968500.
> >
> >Actually, not quite. You need to round values less than nnn.5
> >_down_ to nnn and values of nnn.5 or greater _up_ to nnn.
>
> Yes, tht is the conventional method of rounding, but since I used a
> multiplier of 256 in my calculations, the only way I could get their
> RGB numbers was to round all fractions up.
Yes, that is because you used the wrong multiplier.
>
> This all comes down to which of us was using the correct multiplier.
I did, I'm afraid. Another way of looking at the math is this
piece of simple algebra:
Y = m X + c
Y = color on a scale of 0 to 255
X = color on a scale of 1 to 100
m = slope
c = intercept
When X = 0, Y also = 0, so c must be zero and all you need is:
Y = m X
When Y = 255, X = 100, so m = 255 / 100 (and not not 256 / 100)
The eye cannot see differences between colors even further apart
as a matter of fact, especially in certain blue shades. The eye
also sees detail information primarily in brightness, not in
color. This is why the JPEG file format throws away four times
as much color information as brightness information. Roughly
speaking, the eye can see finer detail the higher the contrast
in the image. The ability to see spatial frequency as a function
of contrast is determined by the contrast sensitivity function.
You can do an experiment for yourself at this link (which might
have wrapped):
http://totoro.berkeley.edu/Demonstrations/VSOC/izumi/CSF/A_JG_RobsonCSFchart.html
You will see that the more closely spaced lines are, the more
contrast you need to be able to see them. So, your spatial
detail comes mostly from lightness or brightness variation.
Color is secondary, the contrast sensitivity function falling
off more rapidly at higher frequencies for color than for
lightness. The rest of the information in the image is color
difference information. The CIE L*a*b* color space is
approximately perceptually uniform when small color differences
are compared. If you compare colors in this space which differ
by one just noticeable difference (JND, also called Delta E) and
you convert both the colors to back RGB you may be astonished by
how big the difference in RGB coordinates actually is. I just
compared two visually indistinguishable blues, differing by much
less than one JND. One had R, G, B values of 101, 99, 246 and
the other had 103, 99, 245. These differences are much larger
than the fraction of a unit you are worried about. (And I didn't
specially pick these colors either :)
>
> Thanks,
> co
> co
>
> baseb...@hotmail.com (aka sabr...@my-deja.com)
You're welcome.
Kris
----------------------------------------------------------------------
Yeah, we *are* pretty cool, *huh*?
a
Actually I suspected that you might have some problem of
this kind but I didn't want to say so outright in case of
giving offense. As I said, it is our aim to relieve people
of unnecessary calculations. I too use a piece of paper and
would rather not have to.
>> I am reasonably intelligent but math is an extremely weak area for me
>> Peggy J
>Actually I suspected that you might have some problem of
>this kind but I didn't want to say so outright in case of
>giving offense. As I said, it is our aim to relieve people
>of unnecessary calculations. I too use a piece of paper and
>would rather not have to.
I think in light of making it easier for artists, so that whatever is
unnecessary can be calculated by machine, a number of things could be
improved in AS (speaking of which).
Basically, it IS tedious to have to modify each frame individually,
rather than in batch or by selecting all and applying. I think
stepping up the grouped layers to more than just a group move or hide
would be helpful. But, of course, a better solution would be to link
in filters from AS, itself, and be able to apply to all selected
frames.
And - AS doesn't have a movie/strip editor, even though animated gifs
can be every bit the film, with overlays and fades, etc. The more
intuitive way, I would think, would be to have a 'mixer' of various
strips which can be opacity faded and masked, slid around, etc. and
ultimately combined, flattened, into one. Hopefully, this is the next
step up for AS. Doesn't have to be Premier. But more would be nice.
>On Mon, 29 Jan 2001 00:11:26 GMT, baseb...@hotmail.com wrote:
>>My HP says it has "36 color output." I took that literally. One of the
>>Mustek scanners claims to have 48-bit depth with 281 trillion colors.
>>
>>Picture Window Pro "can read and write 16-bit black and white images
>>in either TIFF or FITS format, and it can read and write 48-bit color
>>images in TIFF format." PNG also supports 16-bit color resolution for
>>each color value (RGB).
>*Snort* Marketing ploys for those that don't know better. Tell me,
>isn't that a waste when you cannot display it or print it?
Not necessarily. Besides its obvious usefulness in the case of printing
to film, if the scanner does color and gamma correction on the raw image
using ICC profiles--particularly if the profiles are user-selectable--
then extra precision will help prevent visible banding.
If the printer additionally makes the deep pixels available to software,
then the benefits are increased tenfold for image-manipulation tools that
are used to perform additional color adjustments or image transformations
(e.g., RGB -> YUV -> RGB or RGB -> CMYK), perhaps repeatedly. See the
Photoshop newsgroup for a number of recent postings on this topic, or
check out film.gimp.org.
--
Greg Roelofs n_e_w_t_(at)_p_o_b_o_x_(period)_c_o_m
Newtware, PNG Group, Info-ZIP, Philips Research, ...
Thank you for letting me know. :) It is a useful little tool, especially when
you want to see what colors have been used on a webpage, or graphic.
Phoebe
>Ronald Vick <ronal...@mindspring.com> wrote:
>
>>On Mon, 29 Jan 2001 00:11:26 GMT, baseb...@hotmail.com wrote:
>>>My HP says it has "36 color output." I took that literally. One of the
>>>Mustek scanners claims to have 48-bit depth with 281 trillion colors.
A pity the human eye can only distinguish 10 million colors.
;)
--
Owen F. Ransen
http://www.ransen.com/
Home of Gliftic & Repligator Image Generators