AVS Forum banner
941 - 960 of 2065 Posts

·
Registered
Joined
·
13,050 Posts

Quote:
Originally Posted by alluringreality /forum/post/15284755


I was told the secondary location is based on geometry. Draw a line from one primaray through white, and then draw a second line between the other two primaries. The secondary appears where the lines intersect.

Tom - This may drive you crazy to hear but with my RS20 I may use the CMS to improve the red and green over saturation somewhat, but purposely leave them a bit over saturated.


So with the above quote in mind - when it comes to then adjusting the primaries, am I right to assume I should draw the line between the other primaries and move yellow and cyan accordingly? Or would it be the case that even with the over saturated red and green I would still be better of put cyan and yellow at their Rec 709 points? Thanks.
 

·
Registered
Joined
·
121 Posts

Quote:
Originally Posted by alluringreality /forum/post/15284755


I was told the secondary location is based on geometry. Draw a line from one primaray through white, and then draw a second line between the other two primaries. The secondary appears where the lines intersect.


I've never ran across how to calculate Y from xy, which seems more important, so I just use the Accupel calculator.

Thanks.


Thanks.


I know that it is a trigonometric formula...but It's very long to derive...is it somwhere on the web?
 

·
Registered
Joined
·
7,996 Posts
Discussion Starter · #945 ·

Quote:
Originally Posted by lovingdvd /forum/post/15286661


Tom - This may drive you crazy to hear but with my RS20 I may use the CMS to improve the red and green over saturation somewhat, but purposely leave them a bit over saturated.


So with the above quote in mind - when it comes to then adjusting the primaries, am I right to assume I should draw the line between the other primaries and move yellow and cyan accordingly? Or would it be the case that even with the over saturated red and green I would still be better of put cyan and yellow at their Rec 709 points? Thanks.

You are correct. This does drive me crazy. But that aside, if you are going to intentionally engineer inaccurate primaries, then you should calibrate the secondaries to their calculated targets, rather than to the gamut's specs.
 

·
Registered
Joined
·
13,050 Posts

Quote:
Originally Posted by TomHuffman /forum/post/15290998


You are correct. This does drive me crazy. But that aside, if you are going to intentionally engineer inaccurate primaries, then you should calibrate the secondaries to their calculated targets, rather than to the gamut's specs.

By "calculated targets" you mean base the secondaries relative to the primaries right? So for instance draw a line from my somewhat-corrected red, through white, and follow that up to the left hand side to see where it lands to determine what to put cyan?
 

·
Registered
Joined
·
440 Posts
Learning what I could here I have been playing with my TV settings and taking readings... I have a rather lenghty post here...

http://www.avsforum.com/avs-vb/showt...5#post15292295


If anyone cares to comment on it. I post this because using ColorHCFR I am running analysis on a PZ85U Panasonic Plasma in Cinema Mode and using Rec 709 as the guideline/settings in the software. I think there is a problem with regards to trying to evaluate this particular TV with software that talks in 709 mode versus what the TV actually puts out.


It's there for those that care to see what I mean.


C.
 

·
Registered
Joined
·
440 Posts
As an aside, Getgray and DVE HD Blueray present different results when measuring greyscale. Noticeable not only on the DeltaE but also on the Gamma line page of HCFR. They are close but Getgrey's DeltaE's are lower in the lower IRE's and Higher in the higher IRE's.


That said AVS 709 is almost bang on to the DVE HD disk. But there is one problem. At 30% grey it is drastically different at 30% for DeltaE.


Kind of dissapointing. Which one to use for evaluating greyscale?


C.
 

·
Registered
Joined
·
7,996 Posts
Discussion Starter · #950 ·
I don't have access to DVE Blu-ray, but I compared GetGray with an Accupel signal generator, and the differences were negligible.


BTW, there's no mystery about the Panasonic. Its red and green color points are oversaturated, presumably by design.

 

GetGray_Accupel.pdf 183.6630859375k . file
 

Attachments

·
Registered
Joined
·
7,996 Posts
Discussion Starter · #951 ·

Quote:
Originally Posted by lovingdvd /forum/post/15292367


By "calculated targets" you mean base the secondaries relative to the primaries right? So for instance draw a line from my somewhat-corrected red, through white, and follow that up to the left hand side to see where it lands to determine what to put cyan?

Or, calculate it mathematically.
 

·
Registered
Joined
·
440 Posts

Quote:
Originally Posted by TomHuffman /forum/post/15293219


I don't have access to DVE Blu-ray, but I compared GetGray with an Accupel signal generator, and the differences were negligible.


BTW, there's no mystery about the Panasonic. Its red and green color points are oversaturated, presumably by design.

Thanks for the comparison on Getgray and Accupel. I am comming to understand where there are some tolerances in differences or slight changes from one reading to another and one generator or another. If Getgray is good enough for you to recommend to a non-pro, then it is certainly good enough for me.


As for the Panasonic Display, you showed that DCI is another standard set of xy's like 709 or 601 and this particular TV is hitting those coordinates almost bang on. Does that not mean this unit uses DCI gamut? Just got me to thinking that, if your software had a DCI mode along with SD, HD and PAL to choose from, would you be testing in that mode to evaluate DeltaE? When discussing dE, if the TV uses DCI gamut but we compare it against the 709 mathamatically for error evaluation, then might as well use pal or 601? In the end, the DeltaE from 701 is not really accurate against what the TV is pushing? That is, if in fact the PZ85U for instance, is pushing DCI Gamut. On the flipside, do the Bluerays we get as of today use DCI? If not then that's kind of crappy. Just my 2 cents. No big deal really.


That's not for this thread. Sorry about that. Now that I have tried calibration myself, I have a deeper respect for calibrators now. There are so many variables. You guys have to deal with alot of different angles out there. A person cannot truely appreciate calibration until they try it for themselves.


C.
 

·
Registered
Joined
·
29 Posts
I'm trying to calibrate my Samsung LCD and most of the guides I've seen with regards to calibrating primary and secondary colors refers to changing hue, saturation and lightness. However the colour managment system on my Samsung only allows you to change the red, green and blue values for each of the primary and secondary colors.


If I use these controls to get the primary and secondary colors close to their standard values, is this correct?
 

·
Registered
Joined
·
13,050 Posts

Quote:
Originally Posted by TomHuffman /forum/post/15293224


Or, calculate it mathematically.

Good point, mathematically would be much better. What is the formula I would use? For instance, given xyY for Red and xyY for green, how would I calculate xyY values for Yellow?
 

·
Registered
Joined
·
7,996 Posts
Discussion Starter · #955 ·

Quote:
Originally Posted by mparter /forum/post/15295928


I'm trying to calibrate my Samsung LCD and most of the guides I've seen with regards to calibrating primary and secondary colors refers to changing hue, saturation and lightness. However the colour managment system on my Samsung only allows you to change the red, green and blue values for each of the primary and secondary colors.


If I use these controls to get the primary and secondary colors close to their standard values, is this correct?

The Samsungs use a color mixing strategy that works very well.
  • To desaturate a primary, add equal amounts of both of the other 2 primaries to it. To desaturate a secondary, add the opposing primary to it.
  • To change the hue of a primary add unequal amounts of the other 2 primaries. To change the hue of a secondary, add unequal amounts of the contributing primaries.
  • To decrease the lightness of a primary, reduce the amount of the primary itself. To decrease the lightness of a secondary, decrease equally the amount of the contributing primaries.

You cannot increase saturation or lightness.
 

·
Registered
Joined
·
29 Posts

Quote:
Originally Posted by TomHuffman /forum/post/15296174


The Samsungs use a color mixing strategy that works very well.
  • To desaturate a primary, add equal amounts of both of the other 2 primaries to it. To desaturate a secondary, add the opposing primary to it.
  • To change the hue of a primary add unequal amounts of the other 2 primaries. To change the hue of a secondary, add unequal amounts of the contributing primaries.
  • To decrease the lightness of a primary, reduce the amount of the primary itself. To decrease the lightness of a secondary, decrease equally the amount of the contributing primaries.

You cannot increase saturation or lightness.

Thanks.


I'll just check I'm understanding this right;
  • If I want to desaturate red, I add equal amounts of blue and green. To desaturate yellow, I add blue.
  • To change the hue of red, add unequal amounts of green and blue. To change the hue of yellow, add unequal amounts of green and red.
  • To decrease the lightness of red, reduce the amount of red. To decrease the lightness of yellow, reduce the amount of green and red in equal quantities.
 

·
Registered
Joined
·
121 Posts
Hi !


I have a question :


If you convert CIE xyY color coordinates to sRGB coordinates and it gives you negative numbers or numbers greater than 255 (outside the color gamut?), what do you do with them ? Do you normalize them ?


For example CIE xyY (0.2170;0.3310;0.7924) gives sRGB (-19;262;258). What does that means -19?


Thanks in advance.
 

·
Registered
Joined
·
7,996 Posts
Discussion Starter · #959 ·
Yes, it is outside of the gamut. This is an oversaturated cyan.
 

·
Registered
Joined
·
13,050 Posts

Quote:
Originally Posted by lovingdvd /forum/post/15296158


Good point, mathematically would be much better. What is the formula I would use? For instance, given xyY for Red and xyY for green, how would I calculate xyY values for Yellow?

Tom - no rush on this, just wanted to make sure you saw the above question since I think you had posted just as I wrote it. Thanks.
 
941 - 960 of 2065 Posts
Top