CRT gamma calibration question - AVS Forum
Forum Jump: 
 
Thread Tools
post #1 of 17 Old 05-19-2013, 10:00 PM - Thread Starter
Advanced Member
 
spacediver's Avatar
 
Join Date: May 2013
Location: Toronto
Posts: 839
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 70 Post(s)
Liked: 69
Hi there,

I've been using Kal's awesome greyscale & color calibration guide to calibrate my CRT monitor (FW900).

Using a second hand DTP94 and HCFR, I've been able to achieve what appears to be excellent white point balance (delta E's under 2 for 50% and above).

However, my luminance curve needs a bit of a boost to reach the 2.2 gamma standard.

I love the fact that the guide emphasizes hardware controls (brightness, contrast, RGB gain and bias), however, as far as I can tell, I'll need to employ the use of a modified LUT to adjust the midtones independently, as I have no ability to make the necessary adjustments through the on screen controls.

So I have two questions:

1: Does HCFR have the ability to calculate and generate a modified LUT so that my gamma is boosted to fit the curve?

2: If so, would that LUT be implented through an ICC profile that I just load using display properties/advanced/color management (I'm using XP)?

I'd be very grateful for any guidance here. I have a feeling I'm on the right track, but not sure.
spacediver is online now  
Sponsored Links
Advertisement
 
post #2 of 17 Old 05-20-2013, 10:25 AM
AVS Special Member
 
benareeno's Avatar
 
Join Date: Oct 2003
Location: Ottawa, Ontario
Posts: 2,029
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 2 Post(s)
Liked: 14
what is your contrast setting or light output?

And FWIW, there is a gamma program that you may be able to search out in the htpc forum...I'm guessing you already looked. I forget the name, but it would allow you to adjust all 3 colors gamma independently.
benareeno is offline  
post #3 of 17 Old 05-20-2013, 10:36 AM - Thread Starter
Advanced Member
 
spacediver's Avatar
 
Join Date: May 2013
Location: Toronto
Posts: 839
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 70 Post(s)
Liked: 69
contrast set to 100% on OSD. Haven't touched anything in my nvidia control panel (gamma, contrast, brightness, digital vibrance, etc all set to neutral defaults).

Measured luminance ranges from 0.01 to 77.73 cd/m2.

Given that my white balance looks great, I'm guessing that I can get away with using the same values in the LUT for R G and B respectively.

No I wasn't aware of this program - I'll try searching it out.

In the mean time, I suppose I can use trial and error with the nvidia control panel's gamma control, but it would be nice to be able to create a custom ICC profile that had all that stuff taken care of (assuming such a thing is possible: I'm not really clear on what information ICC profiles contain).
spacediver is online now  
post #4 of 17 Old 05-20-2013, 10:52 AM
AVS Special Member
 
sotti's Avatar
 
Join Date: Aug 2004
Location: Seattle, WA
Posts: 6,621
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 28 Post(s)
Liked: 167
The FW900 really doesn't have a gamma control.

In windows you can adjust gamma with the graphics card hardware LUT, but you'll need some software to write to it.

Joel Barsotti
SpectraCal
CalMAN Lead Developer
sotti is offline  
post #5 of 17 Old 05-20-2013, 11:08 AM
AVS Special Member
 
zoyd's Avatar
 
Join Date: Sep 2006
Location: Planet Dog
Posts: 4,514
Mentioned: 18 Post(s)
Tagged: 0 Thread(s)
Quoted: 118 Post(s)
Liked: 324
Quote:
Originally Posted by spacediver View Post

So I have two questions:

1: Does HCFR have the ability to calculate and generate a modified LUT so that my gamma is boosted to fit the curve?

2: If so, would that LUT be implented through an ICC profile that I just load using display properties/advanced/color management (I'm using XP)?

I'd be very grateful for any guidance here. I have a feeling I'm on the right track, but not sure.

HCFR does not calculate 1-d luts for gamma manipulation. You can use ArgyllCMS (command line) or ArgyllCMS+DispcalGUI (GUI front-end) to do this for you and load the corrections to your graphics card.

-----

To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.
zoyd is offline  
post #6 of 17 Old 05-20-2013, 12:57 PM - Thread Starter
Advanced Member
 
spacediver's Avatar
 
Join Date: May 2013
Location: Toronto
Posts: 839
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 70 Post(s)
Liked: 69
thanks for the replies - i avoided dispcalgui because I wasn't able to get it to detect the DTP94, however I got it working once I loaded the correct argyllcms drivers. Found a decent youtube tutorial so I think I'm in good shape.
spacediver is online now  
post #7 of 17 Old 05-20-2013, 06:51 PM
Member
 
cyberbeing's Avatar
 
Join Date: Apr 2006
Posts: 61
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 12
Quote:
The FW900 really doesn't have a gamma control.

If the GDM-FW900 is anything like my GDM-F520, increasing Brightness/Bias will lower gamma at the expense of black level. You can get various types of curves by tweaking the ratios between the individual Red/Green/Blue Brightness/Bias & and Master Brightness/Bias. For example, last week I successfully got my GDM-F520 to reproduce something very close to a BT.1886 curve naturally. When using HCFR with a CRT like this, you should also remember to enable Red, Green, Blue display on the Gamma tab to balance each color guns' Brightness/Bias (low-end) & Contrast/Gain (high-end) at the same time you do grayscale white balance.

These Sony CRTs also have the WinDAS service tool available (requires direct TTL Serial cable connection to the back of the CRT) to perform white balance adjustment (which by default targets ~2.2 gamma), and gives access to a lot of other advanced geometry and convergence functionality not exposed in the OSD, as well as voltage adjustments.
cyberbeing is offline  
post #8 of 17 Old 05-20-2013, 08:07 PM - Thread Starter
Advanced Member
 
spacediver's Avatar
 
Join Date: May 2013
Location: Toronto
Posts: 839
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 70 Post(s)
Liked: 69
yep I have the winDAS cable and once i get a spare cheap laptop I'll experiment with it on one of my backup fw900s.

When I did the white balance, I used a high luminance white pattern (RGB 220) and adjusted the RGB gain until the RGB levels were each at 100%, and then used a low luminance white pattern (RGB 50) and adjusted the RGB bias. A couple back and forths and both patterns showed 100% each within 1%.

Is this what you meant by "enable Red, Green, Blue display on the Gamma tab"? I don't see a Gamma tab, but I do have something called RGB levels that shows realtime RGB levels during a measurement. And I also have a RGB levels graph that shows individual curves for each color as a function of stimulus intensity.
spacediver is online now  
post #9 of 17 Old 05-20-2013, 08:30 PM
Member
 
cyberbeing's Avatar
 
Join Date: Apr 2006
Posts: 61
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 12
Quote:
Originally Posted by spacediver View Post

Is this what you meant by "enable Red, Green, Blue display on the Gamma tab"? I don't see a Gamma tab, but I do have something called RGB levels that shows realtime RGB levels during a measurement. And I also have a RGB levels graph that shows individual curves for each color as a function of stimulus intensity.

RGB levels is more of representation of deviation from your white point reference (i.e. D65), and not what I'm referring to.

I'm talking about the "Gamma graph" (Graphs->Gamma Graph or toolbar button) which can also be seen in the Information "Gamma" drop-down from the main ColorHCFR window. By default it only shows a Yellow luminance line. You need to right click on the Gamma Graph display and enabled the individual Red, Green, and Blue gamma plots. It will be populated when you take grayscale measurements

First balance the Red, Green, Blue plots on the Gamma graph so they lined up as much as possible on both the low-end and high-end.
Second do a rough correction of RGB Levels to balance the Red, Green, Blue plots on both the low-end and high-end.
At that point, it's a balancing act between the "Gamma graph" and the "RGB Levels". Switch back and forth between the two as you fine-tune with 5% grayscale measurements.
Overall "RGB Levels" (white balance) takes priority, but using the Expert Color controls from the OSD, you should be able to achieve close to matching Red/Green/Blue Gamma plots as well.

I've found this to be important with these CRTs, since otherwise you could be fooled by perfect color temperature & luminance gamma plots, when your Red & Blue gamma plots are way off target.
___
cyberbeing is offline  
post #10 of 17 Old 05-20-2013, 10:47 PM - Thread Starter
Advanced Member
 
spacediver's Avatar
 
Join Date: May 2013
Location: Toronto
Posts: 839
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 70 Post(s)
Liked: 69
thanks for the clarification. I've noticed that my white point balance is off at the very low luminance levels (delta E of ~80 at 0%, and then rapidly improves so that by 20% it's below 3). This is reflected by a wide divergence of the RGB levels at low luminance levels.


The luminance graph shows a nice tight curve for all three channels that matches the reference curve extremely well.

The gamma graph shows a pretty good fit, the yellow luminance line averages around 2.2, and the individual channels track the luminance levels well, but they're all a bit off at the high and low luminances by around 0.1 gamma.

And finally, in the CIE diagram, the blue target is met well, but the green and red targets are noticeably off.

This was done after calibrating and profiling using dispcalgiu.

cyberbeing, I will attempt to use your suggestions to balance things using OSD exclusively - the info you've provided is very useful.
spacediver is online now  
post #11 of 17 Old 05-21-2013, 02:59 AM
AVS Special Member
 
Elix's Avatar
 
Join Date: Jun 2011
Location: Dungeon, Pillar of Eyes
Posts: 1,212
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 29 Post(s)
Liked: 24
I also have Sony FW900 and I set black level just so I can't see it's ON when the room is completely black. Of course with such low brightness (G2 voltage) you get crushed blacks. Severely crushed. So I use some tricks to get both stellar blacks and shadow detail... Gamma (what gives you shadow detail - that, and native contrast) can be calibrated to near-perfection with video card's LUT. It can be calibrated with AgryllCMS to create an .icm profile (a LUT curve basically) in 8-bit and then you may convert it to 16-bit with this: http://www.avsforum.com/t/1042160/videoequalizer-2-0-powerful-gamma-correction-for-your-pc
But I actually prefer another method which is quite a bit easier. I use a little neat program called Atrise Lutcurve. There you set target gamma, pick a correction point at 1 and move it up until you're able to distinctly see shadows up to 1 (make sure to untick the correction point at 127). After that you adjust overall gamma at the gamma tab until you won't see any color bars. Works like a charm.
Elix is offline  
post #12 of 17 Old 05-21-2013, 07:44 AM - Thread Starter
Advanced Member
 
spacediver's Avatar
 
Join Date: May 2013
Location: Toronto
Posts: 839
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 70 Post(s)
Liked: 69
thanks Elix - I'm curious - when you adjust your master brightness to get inky blacks, what is the luminance output for white stimuli?
spacediver is online now  
post #13 of 17 Old 05-21-2013, 06:59 PM - Thread Starter
Advanced Member
 
spacediver's Avatar
 
Join Date: May 2013
Location: Toronto
Posts: 839
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 70 Post(s)
Liked: 69
cyberbeing, I'm having a bit of conceptual difficulty with your distinction between the RGB gamma levels, and the RGB levels.

if the RGB levels are kept balanced as a function of stimulus intensity, won't everything else fall into place in terms of accurate color reproduction? Yes, this doesn't guarantee the correct luminance curve, but if RGB levels are balanced, then any deviations from gamma 2.2/2.4 will be equally distributed among the three channels, no?

Also, I'm having great difficulty balancing my RGB levels at 30% intensity or below. I'm hoping that this reflects the limitations of my DTP94 - from 30% to 100% my levels are great as indicated by the colorimeter. Any insights here?
spacediver is online now  
post #14 of 17 Old 05-21-2013, 07:02 PM
AVS Special Member
 
benareeno's Avatar
 
Join Date: Oct 2003
Location: Ottawa, Ontario
Posts: 2,029
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 2 Post(s)
Liked: 14
I'm thinking 100% contrast is not optimal...it almost can't be...try calibrating to a lower light output and recheck your gamma.
benareeno is offline  
post #15 of 17 Old 05-21-2013, 07:31 PM - Thread Starter
Advanced Member
 
spacediver's Avatar
 
Join Date: May 2013
Location: Toronto
Posts: 839
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 70 Post(s)
Liked: 69
Here are the results of my calibration using OSD. I also boosted my gamma in the nvidia control panel.

Do these results look ok? In particular, the CIE diagram shows a couple primaries a bit off, but not sure how serious it is.









spacediver is online now  
post #16 of 17 Old 05-21-2013, 11:43 PM
AVS Special Member
 
Elix's Avatar
 
Join Date: Jun 2011
Location: Dungeon, Pillar of Eyes
Posts: 1,212
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 29 Post(s)
Liked: 24
Quote:
Originally Posted by spacediver View Post

thanks Elix - I'm curious - when you adjust your master brightness to get inky blacks, what is the luminance output for white stimuli?
Between 90-100 cd/m2. G2 voltage is set via WinDAS @ 140, iirc.
Elix is offline  
post #17 of 17 Old 05-22-2013, 01:56 PM - Thread Starter
Advanced Member
 
spacediver's Avatar
 
Join Date: May 2013
Location: Toronto
Posts: 839
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 70 Post(s)
Liked: 69
interesting. I'd like to be able to get really good black levels with 90-100 cd/m2. Perhaps I'll try raising the infamous G2, although I'm also curious to see how the contrast changes when I remove the antiglare.

Does anyone have any comments on my calibration results? I'd like to hit those primaries a bit more tightly, but not sure whether it's feasible. I'm also surprised that they're off given my good delta Es on the grayscale level.
spacediver is online now  
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