Secondary color locations - AVS Forum
Forum Jump: 
 
Thread Tools
post #1 of 18 Old 10-17-2007, 10:36 AM - Thread Starter
AVS Special Member
 
alluringreality's Avatar
 
Join Date: Jul 2006
Posts: 3,120
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 6 Post(s)
Liked: 33
Does anyone have an online source for giving secondary xy locations based on primary locations? I've read that the secondary should lie at the intersection of two lines. The first line being between two primaries. The second line being between the third primary and white (extending past white to intersect the first line).

Based on that description I can use geometry to come up with secondary locations based on primary locations, but I was wondering if there was any other source I might be able to use with less work. The way I'm thinking about locating a point, the geometry way might involve a number of angles and lengths to locate all three secondaries. It's probably not any worse than statics force calculations, but I just thought there might be a simpler way someone here might be familiar with.
alluringreality is offline  
Sponsored Links
Advertisement
 
post #2 of 18 Old 10-17-2007, 12:01 PM
Advanced Member
 
Jim Burns's Avatar
 
Join Date: Jun 2001
Location: Middle of nowhere NY State
Posts: 735
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Here are the NTSC numbers

White (D65) = 0.3127, 0.3290
Red = 0.630, 0.340
Green = 0.310, 0.595
Blue = 0.155, 0.070
Yellow = 0.421, 0.507
Cyan = 0.231, 0.326
Magenta = 0.314, 0.161

Jim Burns

cheap does not mean value
Jim Burns is offline  
post #3 of 18 Old 10-17-2007, 12:03 PM
Advanced Member
 
Jim Burns's Avatar
 
Join Date: Jun 2001
Location: Middle of nowhere NY State
Posts: 735
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
This should help with links to other areas too.

http://en.wikipedia.org/wiki/User:PAR/Work3

Jim Burns

cheap does not mean value
Jim Burns is offline  
post #4 of 18 Old 10-17-2007, 01:37 PM - Thread Starter
AVS Special Member
 
alluringreality's Avatar
 
Join Date: Jul 2006
Posts: 3,120
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 6 Post(s)
Liked: 33
I can't change my primary color locations. I've read that the gregr luminance calculator can be used to figure luminance levels based on primary locations. Now I just want to do the same for secondary locations. I wouldn't mind seeing the math on the luminance calculator also, but the Denver library doesn't seem to have many books on color where I might be able to get the equations to figure secondaries and luminance based on primaries and white point location.
alluringreality is offline  
post #5 of 18 Old 10-17-2007, 01:45 PM
Advanced Member
 
Jim Burns's Avatar
 
Join Date: Jun 2001
Location: Middle of nowhere NY State
Posts: 735
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
My mistake, I now see what you are looking for, let me look in some of my books, I know I have the answer..... just where to find it?

Jim Burns

cheap does not mean value
Jim Burns is offline  
post #6 of 18 Old 10-17-2007, 01:52 PM
AVS Addicted Member
 
krasmuzik's Avatar
 
Join Date: Oct 1999
Location: NewPort, VA
Posts: 11,270
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
That is easy - simply add the two primaries luminance to make the secondary.
krasmuzik is offline  
post #7 of 18 Old 10-17-2007, 02:03 PM - Thread Starter
AVS Special Member
 
alluringreality's Avatar
 
Join Date: Jul 2006
Posts: 3,120
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 6 Post(s)
Liked: 33
I know adding primary Y values gives me secondary Y, but I don't see how adding Y gives me secondary xy. I'm just looking to be able to find secondary xy points for the primaries that I have. HCFR doesn't seem to have any way to set secondaries aside from looking at the CIE charts. So all I could think of to plug into a spreadsheet to easily tell how close my secondaries are from intended xy was to use the geometry you talked about in the dummies chart thread.
alluringreality is offline  
post #8 of 18 Old 10-17-2007, 03:49 PM
AVS Addicted Member
 
krasmuzik's Avatar
 
Join Date: Oct 1999
Location: NewPort, VA
Posts: 11,270
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
You have to decide if you want to align your secondary xy with your primaries xy and D65 geometrically - for a theoretically perfect display for that display gamut - or decide if you want to hit D65 and CMY on REC709 with as little error as possible and sacrifice your RGB gamut to your displays limitations.

I prefer the latter - simply because if you were to plot the colors existing in most source material - you would find a high probability that it is contained within the secondary gamut - going out to primary gamut does not happen as often.
krasmuzik is offline  
post #9 of 18 Old 10-17-2007, 04:25 PM - Thread Starter
AVS Special Member
 
alluringreality's Avatar
 
Join Date: Jul 2006
Posts: 3,120
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 6 Post(s)
Liked: 33
Quote:
Originally Posted by krasmuzik View Post

You have to decide if you want to align your secondary xy with your primaries xy and D65 geometrically - for a theoretically perfect display for that display gamut

I guess that seems to be in line with how I took your comment in the other thread, of drawing from the primary through white to locate the secondary.

Just to be very specific, I have an SXRD TV where I can't seem to change primaries and they're all outside the 709 gamut. I seem to be able to move secondaries quite far along the outside of the gamut, but I'm not quite sure where is the right place to set that secondary point in HCFR. Originally I was using RGB values figured off 709, but after reading the comment in the chart sticky I decided maybe the geometry you talked about there was the correct way to go about it.

Quote:


or decide if you want to hit D65 and CMY on REC709 with as little error as possible and sacrifice your RGB gamut to your displays limitations.

I really don't understand what this part means. Does that mean drawing a line from the D65 xy point through the Rec709 secondary? Would where that line intersects the outside of the gamut be where you're suggesting to set the secondary in accordance with the sort of picture you prefer?
alluringreality is offline  
post #10 of 18 Old 10-17-2007, 10:14 PM - Thread Starter
AVS Special Member
 
alluringreality's Avatar
 
Join Date: Jul 2006
Posts: 3,120
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 6 Post(s)
Liked: 33
Should I try to get the Y for the secondary to match the Y from the two related primaries?

Okay... maybe I get the first "add the two primaries luminance to make the secondary" comment now. I had always seen it that way in the charts, but I didn't get the relevance. Looking at a couple measurements I had taken of tint settings I can see that my secondaries didn't add to the primaries. In the secondaries involving green on one side of the 709 secondary point my measurements show that the secondary Y is higher than the two primaries added and on the other side it shows that the secondary Y is lower than the primary Y values added. With magenta, both sides of the 709 value shows that Y on the secondary is low. So if I aim to get the secondary Y to equal the two relevant primaries added together, is that the recommended location for the secondary?
alluringreality is offline  
post #11 of 18 Old 10-18-2007, 12:04 PM
AVS Addicted Member
 
krasmuzik's Avatar
 
Join Date: Oct 1999
Location: NewPort, VA
Posts: 11,270
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
My comment to hit the secondary gamut - I mean adjust your set so that CMYW is hitting the specified xyY values for REC709/D65.

Then do your best to get the primaries to their specified values - giving most priority to at least getting hue and brightness correct.

I don't believe one should try to tweak their decoding to be theoretically perfect for ones existing primaries.
krasmuzik is offline  
post #12 of 18 Old 10-18-2007, 01:22 PM
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 4,434
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 81 Post(s)
Liked: 303
Quote:
Originally Posted by krasmuzik View Post

My comment to hit the secondary gamut - I mean adjust your set so that CMYW is hitting the specified xyY values for REC709/D65.

I think he's stuck working in the expanded(non-709) gamut. So setting Y can be accomplished using the luminance calculator and Ym=Yb+Yr for example. But how does he calculate xy of the secondaries from xy of the primaries+white instead of drawing a line on his computer screen.
zoyd is offline  
post #13 of 18 Old 10-18-2007, 05:12 PM - Thread Starter
AVS Special Member
 
alluringreality's Avatar
 
Join Date: Jul 2006
Posts: 3,120
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 6 Post(s)
Liked: 33
I can't really move primaries. Red moves a bit with a lot of tint, but other primaries' xy seems fixed. No matter what, my primaries and secondaries are going to generally look somewhat like the attached picture, I was just looking for if there was some target to shoot for when primaries are off.

The best sense I could make of kras's last comments were the first three sentences here about getting dE close (http://www.avsforum.com/avs-vb/showt...9#post10414549). I certainly cannot match all three values of xyY from the 709 reference for my secondaries. I also don't see how it's possible to change my primaries' hue (I thought tint had to do with rotation).

Quote:


So setting Y can be accomplished using the luminance calculator and Ym=Yb+Yr for example.

The way I read this is that setting the secondaries to match the given Y would be an acceptable way to set the secondaries' position if I use the luminance calculator. (I think it would also be equal RGB, if I knew how to calculate RGB related to the primaries)

Quote:


But how does he calculate xy of the secondaries from xy of the primaries+white instead of drawing a line on his computer screen.

If I knew what I was supposed to be doing with the dotted line in the picture, I could figure a point in 2d space.

TomHuffman's guide made setting primary Y levels very clear, but for secondaries I'm just not sure if there's some accepted way of something to shoot for when the primaries are off. Here are the ways that at one point or other have seemed like they might be a way to set secondaries, but I still have little idea which one might be the way to go:

- Equal RGB bars (HCFR calculates RGB based on colorspace setting)
- Geometry
- DeltaE against 709 reference
- Setting Y secondary = Y primary1 + Y primary2 (Similar to setting primary Y levels)
LL
alluringreality is offline  
post #14 of 18 Old 10-19-2007, 02:18 PM
AVS Addicted Member
 
krasmuzik's Avatar
 
Join Date: Oct 1999
Location: NewPort, VA
Posts: 11,270
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Quote:
Originally Posted by zoyd View Post

I think he's stuck working in the expanded(non-709) gamut. So setting Y can be accomplished using the luminance calculator and Ym=Yb+Yr for example. But how does he calculate xy of the secondaries from xy of the primaries+white instead of drawing a line on his computer screen.

Standard geometrical math - I did line intersections in high school. I am not about to attempt doing it umpteen years (I mean decades ) later. But it looks like the HCFR chart draws the line intersections for you - so what is the issue? Just click on it's CMY and see what xy it came up with?

I would rather try to hit REC709 xy on CMY and leave RGB xy where they are stuck. It's not that I hated geometry...
krasmuzik is offline  
post #15 of 18 Old 10-19-2007, 03:39 PM
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 4,434
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 81 Post(s)
Liked: 303
Quote:
Originally Posted by krasmuzik View Post

But it looks like the HCFR chart draws the line intersections for you - so what is the issue? Just click on it's CMY and see what xy it came up with?

Of course... HCFR already does it for you, another case of over-thinking a problem.
zoyd is offline  
post #16 of 18 Old 10-19-2007, 05:32 PM - Thread Starter
AVS Special Member
 
alluringreality's Avatar
 
Join Date: Jul 2006
Posts: 3,120
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 6 Post(s)
Liked: 33
I take it that the suggestion is to send the dotted line from the primary through the 709 secondary.
alluringreality is offline  
post #17 of 18 Old 10-19-2007, 06:30 PM
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 4,434
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 81 Post(s)
Liked: 303
Quote:
Originally Posted by alluringreality View Post

I take it that the suggestion is to send the dotted line from the primary through the 709 secondary.

If you want to find the theoretical secondary xy locations for your gamut you can do that by aligning the line that HCFR draws between the primary and secondary through the white point. If you do that and then mouse over the secondary location it will tell you the xy location. Perceptually though, it's recommended you move the secondary as close to the rec709 (or 601 depending on source) point as possible and not worry about intersecting the white point.
zoyd is offline  
post #18 of 18 Old 11-19-2007, 09:47 AM
AVS Special Member
 
richlo's Avatar
 
Join Date: Aug 2004
Posts: 1,463
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Back to this..

In setting Primaries to Ref on the Gamut (or close to), my secondary’s are OFF-LINE from the line drawn between all the primaries. I have no saturation control in my CMS, just for brightness and hue. Is it important that I move my primaries so that it falls in line with my secondary’s so long as CMY is going through their ref points and then through white, next setting my primaries from where my secondary’s are established, through white..

Is this correct?? Hope Im not confusing anyone..

Rich L

Go Chicago Bears

richlo is offline  
Reply Display Calibration

User Tag List

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off