or Connect
AVS › AVS Forum › Display Devices › Display Calibration › HCFR - Open source projector and display calibration software
New Posts  All Forums:Forum Nav:

HCFR - Open source projector and display calibration software - Page 60

post #1771 of 3436
Thanks Vega,

I have the P58V10 plasma. From what your response indicates I should be using the refresh setting and APL patterns. W/B adjustments with regards to light areas are referring to the 50%-100%, with dark representing the 0-50%range. I have all enhancements turned off except for those where it's not an option to turn off.

With W/B based on your information, if I'm checking 80%iRE with HCFR and red is less than 100% should adjust W/B high R to get as close to 100%, and similarly for the blue? I will recheck tonight and see what I can dial in. Would it be inappropriate to post graphs for suggestions or comments on what I might do to further correct the calibration?

Thanks again...Your comments are greatly appreciated.
post #1772 of 3436
Quote:
Originally Posted by SnkEyez95 View Post

Thanks Vega,

I have the P58V10 plasma. From what your response indicates I should be using the refresh setting and APL patterns. W/B adjustments with regards to light areas are referring to the 50%-100%, with dark representing the 0-50%range. I have all enhancements turned off except for those where it's not an option to turn off.

With W/B based on your information, if I'm checking 80%iRE with HCFR and red is less than 100% should adjust W/B high R to get as close to 100%, and similarly for the blue? I will recheck tonight and see what I can dial in. Would it be inappropriate to post graphs for suggestions or comments on what I might do to further correct the calibration?

Thanks again...Your comments are greatly appreciated.

I had limited success with the apl patterns on both my Panasonic and my Pioneer. Small windows, 4%, worked best for me. When adjusting the RGB, since you do not have any adjustment for green, move the red and blue to line up with the green line, both for rgb and for the gamma. Starting with the 80% and 20% gray points is just that. Look over the whole spectrum and find where the best fit will be as you only have 2 point w/b controls from what you have posted.

In the end, what is important more than anything is that the picture is pleasing to you in your viewing environment. I've calibrated a few sets that had great graphs and numbers, but less than optimal picture. Sometimes small changes can make a very big difference. Don't be afraid to experiment. If you are so inclined keep a log of changes you made, and the effects they had in HCFR, and what you see on the set. Its a learning process, and to me it seems you never stop learning.

most definitely post your graphs, and the raw data, it makes it a lot easier to see what you are seeing. each graph can be saved as a graphic file by right clicking on the graph, then save as. you can then attach it as a photo in your post.
post #1773 of 3436
Alright...Here is my first attempt. Played around for quite some time. Not sure if it's the nature of calibrating or my Spyder4 budget meter, but I was getting some inconsistent retest readings. This was my most consistent results. Any suggestions? Weak ares? Areas for improvement?





post #1774 of 3436

looks pretty good for the controls you have available. you may want to try moving red high down a notch. The gamma at 2.48 seems a little high, unless you are viewing the set in a dark room. This may also be an effect of the patterns you used, and the limiting circuit in the set. With the APL patterns I can get a gamma of 2.37 and a light output of 32fL. Without making any changes to the controls, using the 4% patterns, light output jumps to 38fL, and gamma reads at 2.23. Obviously the effect of the limiting circuits with the APL patterns. Which is correct? What ever looks best to you.

 

how does actual content look?

post #1775 of 3436
Quote:
Originally Posted by vega509 View Post

looks pretty good for the controls you have available. you may want to try moving red high down a notch. The gamma at 2.48 seems a little high, unless you are viewing the set in a dark room. This may also be an effect of the patterns you used, and the limiting circuit in the set. With the APL patterns I can get a gamma of 2.37 and a light output of 32fL. Without making any changes to the controls, using the 4% patterns, light output jumps to 38fL, and gamma reads at 2.23. Obviously the effect of the limiting circuits with the APL patterns. Which is correct? What ever looks best to you.

how does actual content look?

I did calibrate on my PS3 for blue rays. I traditionally watch at night in an unlit room. Is a higher gamma more appropriate for dark viewing environments?

Is there a magic or a target fL I should be keying towards?

The set does look considerably better. A nice natural look.

Next I'm looking to calibrate for a "day" mode on another hdmi input, is there a different approach for day vs night calibration? Higher contrast for higher fL? I know a lot of this stuff is subjective, just curious if there a general area of focus.
post #1776 of 3436
Quote:
Originally Posted by SnkEyez95 View Post

I did calibrate on my PS3 for blue rays. I traditionally watch at night in an unlit room. Is a higher gamma more appropriate for dark viewing environments?
Typically yes. I like 2.3-2.35 for a dark room. If your set has poor black levels (which I don't think is the case here), bt.1886 is probably better.
Quote:
Is there a magic or a target fL I should be keying towards?
Everyone's eyes are different. Typically you want to be in the 30-35fL range for a dark room.
Quote:
Next I'm looking to calibrate for a "day" mode on another hdmi input, is there a different approach for day vs night calibration? Higher contrast for higher fL? I know a lot of this stuff is subjective, just curious if there a general area of focus.
Not really... just target a higher brightness. Seeing as how you have a plasma, use the contrast control as you said. You'll probably want to target a lower gamma, too.
post #1777 of 3436
Did anybody manage to compile the code with Visual Studio 2012 under Windows 8 x64?
First I got this error:
Quote:
Error 1 error RC1015: cannot open include file 'l.esp\afxres.rc'. C:\Users\János\Documents\hcfr\CHCFR21_ESPANOL.rc 3163 1 CHCFR21_ESPANOL
And I got a countless number of type definition errors when I removed the entire ESPANOL project from the solution.

I just wanted to update the ArgyllCMS code. But I faild to build the vanilla code to start with. frown.gif
post #1778 of 3436
Thanks for the reply rahzel. I set my fL in the 39-40 range. It was jumping up and down between the two. Would lowering that to something closer to 35fL have a decreasing effect on gamma. I'm obviously still very new at this, and trying to figure out what settings or changes produce the greatest changes to gamma. Thanks everyone for you help.

Anything to try next? I also wasn't very certain on what your comment on bt.1886 referred to. Probably something to advanced for my current knowledge.

When tweaking W/B should I be targeting R and B to 100% or to match G?

Thanks again!
post #1779 of 3436
Quote:
Originally Posted by SnkEyez95 View Post

When tweaking W/B should I be targeting R and B to 100% or to match G?

The usual "rule of thumb" is: Do not touch green! = Always keep the green channel contrast at it's default neutral maximum setting (unlimited but unclipped) and balance the gray-scale with the red and blue components around green.
But the real rule in the digital world is to keep one channel contrast (the one which should be the brightest of them all) at it's neutral maximum setting (unlimited but unclipped) and lower the other two.
On many flat screens, it's usually still the green channel though. But, for example, red should be max out on Samsung PDPs when you start from the real neutral settings in the Service Menu (because of this, unfortunately there is no other way to calibrate those TVs objectively). And Panasonic PDPs have failsafe controls anyway (they scale everything back if you move the Green Gain in the user menu to get the G-DRV back to neutral max.)
post #1780 of 3436
Ya, generally it's recommended to leave green alone. But if you know what you're doing, it should be ok to change green. If your set has a 10pt white balance and no gamma gain control, then that's the only way to change the gamma... by grouping RGB up and down. But if you only have a 2pt W/B system or you have a gamma gain control, you generally don't need to change green.
post #1781 of 3436
Quote:
Originally Posted by janos666 View Post

The usual "rule of thumb" is: Do not touch green! = Always keep the green channel contrast at it's default neutral maximum setting (unlimited but unclipped) and balance the gray-scale with the red and blue components around green.
But the real rule in the digital world is to keep one channel contrast (the one which should be the brightest of them all) at it's neutral maximum setting (unlimited but unclipped) and lower the other two.
On many flat screens, it's usually still the green channel though. But, for example, red should be max out on Samsung PDPs when you start from the real neutral settings in the Service Menu (because of this, unfortunately there is no other way to calibrate those TVs objectively). And Panasonic PDPs have failsafe controls anyway (they scale everything back if you move the Green Gain in the user menu to get the G-DRV back to neutral max.)

Quote:
Originally Posted by rahzel View Post

Ya, generally it's recommended to leave green alone. But if you know what you're doing, it should be ok to change green. If your set has a 10pt white balance and no gamma gain control, then that's the only way to change the gamma... by grouping RGB up and down. But if you only have a 2pt W/B system or you have a gamma gain control, you generally don't need to change green.

Thanks guys. I don't have a green control. Just Red and Blue. I only have a high and low adjustment so unfortunately can't dial in the 10pt scale. So I was using 30/80 IRE to adjust the w/b R B high and low so their values were as close to 100%. Is this correct method? Should I be dialing in R and B to match G or get R and B close to 100% as possible?

My set does have gamma normal, mid, full 1, and full 2. Haven't really been able to figure out exactly what that does. I suppose more testing is in order to dial gamma in better.

I was using the AVS HD 709 calman windows test patterns for grayscale. If that makes any difference.
post #1782 of 3436
When you set red and blue, green will get balanced without altering it. Ideally you want all of them at 100% from 0-100, but you probably won't get it perfect without a 10pt system. Shoot for an average dE of 2-3 or less.
30/80 (or 20/80) are just starting points... minimize overall errors by checking the entire grayscale.

Your gamma control sounds very limited... it probably doesn't allow you to adjust the gamma at different points... it's just a slider so to speak that sets the overall gamma. Set it to whatever is closest to your desired target, then calibrate the grayscale afterwards because it will probably throw off your RGB balance if you change it afterwards. When you have a gamma control like that, sometimes if you set it too high or too low, it will give you a bad gamma curve (ie dips or peaks in luminance at different levels) so you might want to settle for one that has a flatter gamma curve rather than one that is closer to your desired target.

The AVS window patterns are fine. You might also want to try the APL or window patterns on the GCD disc. Or there are others that are making their own calibration discs on these boards.
post #1783 of 3436
Quote:
Originally Posted by rahzel View Post

When you set red and blue, green will get balanced without altering it. Ideally you want all of them at 100% from 0-100, but you probably won't get it perfect without a 10pt system. Shoot for an average dE of 2-3 or less.
30/80 (or 20/80) are just starting points... minimize overall errors by checking the entire grayscale.

Your gamma control sounds very limited... it probably doesn't allow you to adjust the gamma at different points... it's just a slider so to speak that sets the overall gamma. Set it to whatever is closest to your desired target, then calibrate the grayscale afterwards because it will probably throw off your RGB balance if you change it afterwards. When you have a gamma control like that, sometimes if you set it too high or too low, it will give you a bad gamma curve (ie dips or peaks in luminance at different levels) so you might want to settle for one that has a flatter gamma curve rather than one that is closer to your desired target.

The AVS window patterns are fine. You might also want to try the APL or window patterns on the GCD disc. Or there are others that are making their own calibration discs on these boards.

Thanks again rahzel and the entire community. I was able to get my gamma down a bit to 2.15 with RGB pretty flat except for IRE 0 and 10. Delta E less than 3 except for IRE of 0 and 10. Seems my sensor has a hard time measuring that low, or the display just doesn't perform very well that low. I'm very happy with picture at this time.

My TVs gamma settings at full 1 (2.45) and gamma setting at mid (2.15) with nothing in between. I'm assuming this is the best I can get with my experience and equipment. Only thing I saw that may be an issue was at 100 my fL was close to 45.
post #1784 of 3436
@ Zoyd, Since you just dl masciors calibration disc from reading his thread,do you recommend we check the avs709,dve box or still leave it uncheck like GCD?

Also I'am looking hard at the ST60 and it has limited CMS (primaries only) control but offers 10pt wb and gamma adjustments.

How does one measure saturation,hue and luminance for the primaries using HCFR?
post #1785 of 3436
Quote:
Originally Posted by zoyd View Post

HCFR updates it 's reference white every time you measure it in the primaries/secondaries scan, so yes, if you don't want it to update check "none".
unchecked with GCD

HCFR seems to update reference white from AVS HD Contrast measurement to Primaries and Secondaries no matter what options you change from Preferences.
I gone back to use only AVS HD disk with APL patterns.
Using different disk´s with different APL´s is too confusing for me to use with same calibration.

btw, I wonder how someone managed to measure 60-70cd/m2 from Samsung F8500 3D IRE100?
Today I calibrated my´n PS64E8000 3D (Movie and also Game mode), measured 150cd/m2 from IRE100 but when put Samsung SSG-P41002 to front of i1 Display Pro it measured something like 15-30cd/m2 depends how far glasses were from meter lens.
Edited by Make73 - 4/13/13 at 11:50am
post #1786 of 3436
Thread Starter 
Quote:
Originally Posted by PeterLewis View Post

@ Zoyd, Since you just dl masciors calibration disc from reading his thread,do you recommend we check the avs709,dve box or still leave it uncheck like GCD?

unchecked
Quote:
Also I'am looking hard at the ST60 and it has limited CMS (primaries only) control but offers 10pt wb and gamma adjustments.

How does one measure saturation,hue and luminance for the primaries using HCFR?

You can't directly. HCFR reports deltaxy, deltaE, and delta Luminance (HCFR calls is luma), you have to use those along with the visual CIE chart alignments to calibrate whatever CMS you are working on.
post #1787 of 3436
Thread Starter 
Quote:
Originally Posted by Make73 View Post

HCFR seems to update reference white from AVS HD Contrast measurement to Primaries and Secondaries no matter what options you change from Preferences.
I gone back to use only AVS HD disk with APL patterns.
Using different disk´s with different APL´s is too confusing for me to use with same calibration.
There is currently no way to turn off the white update if you measure contrast. Stick to one disk at a time
post #1788 of 3436
Quote:
Originally Posted by SnkEyez95 View Post

Thanks again rahzel and the entire community. I was able to get my gamma down a bit to 2.15 with RGB pretty flat except for IRE 0 and 10. Delta E less than 3 except for IRE of 0 and 10. Seems my sensor has a hard time measuring that low, or the display just doesn't perform very well that low. I'm very happy with picture at this time.

My TVs gamma settings at full 1 (2.45) and gamma setting at mid (2.15) with nothing in between. I'm assuming this is the best I can get with my experience and equipment. Only thing I saw that may be an issue was at 100 my fL was close to 45.
A lot of sensors can't measure accurately below 10% (depending on the black performance of the set) so don't worry too much. The latest build of HCFR shouldn't even report errors below 10%.

As for your gamma, it's not just you... no one would really be able to do much more to the gamma. It is what it is with your limited controls.

Are you calibrating for day or night? 45fL is fine for day, but might be too bright for night/dark viewing. But then again, everyone's eyes are different.
post #1789 of 3436
Quote:
Originally Posted by rahzel View Post

A lot of sensors can't measure accurately below 10% (depending on the black performance of the set) so don't worry too much. The latest build of HCFR shouldn't even report errors below 10%.

As for your gamma, it's not just you... no one would really be able to do much more to the gamma. It is what it is with your limited controls.

Are you calibrating for day or night? 45fL is fine for day, but might be too bright for night/dark viewing. But then again, everyone's eyes are different.

Calibrating off my Blu Ray player which is 90% night time. Next I need to calibrate a different input for day time media player and directiv.

What one setting impacts 100% fL the most? I assume that change will compromise gamma?
post #1790 of 3436
Contrast directly controls 100% white; it is essentially the white control. If set properly, it shouldn't affect the gamma much, if at all. Brightness sets black, Contrast sets white, gamma sets in-between. If everything works as it should, they shouldn't really affect each other, but that is sometimes not the case... you should re-check everything if any changes are made.
Edited by rahzel - 4/13/13 at 2:05pm
post #1791 of 3436
Quote:
Originally Posted by rahzel View Post

I like 2.3-2.35 for a dark room. If your set has poor black levels (which I don't think is the case here), bt.1886 is probably better.

So only when you have poor MLL then you should use ITU-R BT.1886 ?
I found that it also may be better case because of shadow details.
Here´s interesting thread:
http://www.avsforum.com/t/1409045/how-power-law-gamma-calibration-can-lead-to-crushed-blacks
post #1792 of 3436
Thread Starter 
Quote:
Originally Posted by janos666 View Post

Did anybody manage to compile the code with Visual Studio 2012 under Windows 8 x64?
First I got this error:
And I got a countless number of type definition errors when I removed the entire ESPANOL project from the solution.

I just wanted to update the ArgyllCMS code. But I faild to build the vanilla code to start with. frown.gif

Which code base did you start with John's or my fork?
post #1793 of 3436
Just curious, but is there a way to set a custom white point in HCFR (something like 6000K)? I calibrated my grayscale and gamma last night to the Rec. 709 and BT. 1886 targets and got most of it below a dE of 1 (except around 30 IRE, which is around 1.4). I still haven't messed with the primaries and secondaries. But, the default D65 white point is a little too cold for my liking and I think I'd like to try warming it up a bit.

Also, my 42LK450 is mainly used as a PC monitor now (mainly for gaming/TV/movies) and I have a full RGB signal being displayed. Should I be using the Rec./BT 709 color space settings in HCFR and on the LG, or what provides the widest color space? Should I just go by the Rec 709 settings and turn on some level of "Dynamic Color" after calibrating if I think it needs more "pop"? Is there some way to do a proper calibration using the "Wide" (native) color space setting on the TV? Sorry if this isn't the right place to ask, I've just been doing this with HCFR so I'm trying to figure out a workflow around it.
post #1794 of 3436
Thread Starter 
there is a D55 white point option but not D60. For that and a wider color gamut you can open a dummy reference file (simulated probe or whatever you want) and then enter your targets manually in the primaries/secondaries page. Check the box "reference measure" and then open a real probe file and calibrate to your reference sheet.
post #1795 of 3436
Quote:
Originally Posted by Make73 View Post



So only when you have poor MLL then you should use ITU-R BT.1886 ?
I found that it also may be better case because of shadow details.
Here´s interesting thread:
http://www.avsforum.com/t/1409045/how-power-law-gamma-calibration-can-lead-to-crushed-blacks
If your set has good black levels, the bt.1886 gamma curve will be pretty flat anyway. It benefits sets with poor black levels more IMO. But if you choose to use it for all displays, that's fine.
post #1796 of 3436
Quote:
Originally Posted by zoyd View Post

there is a D55 white point option but not D60. For that and a wider color gamut you can open a dummy reference file (simulated probe or whatever you want) and then enter your targets manually in the primaries/secondaries page. Check the box "reference measure" and then open a real probe file and calibrate to your reference sheet.
Sorry, I'm completely new to hardware calibration and to using HCFR. I'm guessing I just tick the "editable data" box and enter the various xyY values into the cells for each desired gray on the gray scale page, and each colour for the primaries and secondaries. I found a table with the standard illuminant coordinates so I can try 6000K and D60 on the grayscale.

As for the colour, do you just set the TV to its native colour space and that's it? Or do you still have to figure out proper colour coordinates? If so, how would I go about that? Like, how do I figure out how wide of a colour space the display can output and determine the appropriate colour values to assign? Is there any documentation or a tutorial I can read on how to do this?

edit: I guess I could just calibrate to the BT 709 or sRGB standard for all of the primaries and secondaries, and then switch the colour space to "Wide" when I'm done. That might work?
Edited by Some dillweed - 4/14/13 at 7:42pm
post #1797 of 3436
Oookay, hopefully my last questions for a while. I noticed people mentioning zoyd's and kjgarrison's correction matrices for the i1d3 devices. Since those were apparently built on plasmas, should I not use them on my LCD? Should I be using something like the generic LCD correction file, or should I not be using a correction file at all unless I get a spectro to profile against?
post #1798 of 3436
Thread Starter 
Quote:
Originally Posted by Some dillweed View Post

Sorry, I'm completely new to hardware calibration and to using HCFR. I'm guessing I just tick the "editable data" box and enter the various xyY values into the cells for each desired gray on the gray scale page, and each colour for the primaries and secondaries. I found a table with the standard illuminant coordinates so I can try 6000K and D60 on the grayscale.

yes, you can cut and paste from a spreadsheet as well.
Quote:
As for the colour, do you just set the TV to its native colour space and that's it? Or do you still have to figure out proper colour coordinates? If so, how would I go about that? Like, how do I figure out how wide of a colour space the display can output and determine the appropriate colour values to assign? Is there any documentation or a tutorial I can read on how to do this?

edit: I guess I could just calibrate to the BT 709 or sRGB standard for all of the primaries and secondaries, and then switch the colour space to "Wide" when I'm done. That might work?

It depends on what you want to do with a gamut wider than BT709 as there is not currently any commercially broadcast or DVD/BD mastered material using anything wider.
Quote:
Oookay, hopefully my last questions for a while. I noticed people mentioning zoyd's and kjgarrison's correction matrices for the i1d3 devices. Since those were apparently built on plasmas, should I not use them on my LCD? Should I be using something like the generic LCD correction file, or should I not be using a correction file at all unless I get a spectro to profile against?

I would recommend starting with one of the LCD choices under the spectral sampling selection pulldown during meter set-up until you get a spectrometer to profile against.
post #1799 of 3436
Quote:
Originally Posted by zoyd View Post

It depends on what you want to do with a gamut wider than BT709 as there is not currently any commercially broadcast or DVD/BD mastered material using anything wider.
Yeah, I don't know why but I thought that the native colour space might be the best option to use for Windows. I'm sure most people just stick to Rec. 709 or sRGB for their calibration purposes, though. I was basically just thinking in terms of adding some extra saturation if I felt the picture needed some extra vividness, but I guess that's what settings like "Dynamic Color" are for.
Quote:
I would recommend starting with one of the LCD choices under the spectral sampling selection pulldown during meter set-up until you get a spectrometer to profile against.
You mean in the meter setup page, not the correction file pulldown with the meter selection just before that, right? Just want to make sure I'm doing the right thing. Yeah, I don't quite have the cash flow for a spectro at this point, so that'll have to wait.

Thanks for all of the help, zoyd, and thanks for all of your work on the software.
Edited by Some dillweed - 4/15/13 at 11:45am
post #1800 of 3436
Zoyd,

Any chance you'll release the source to your fork? I love what you've improved on so far. However, using 3.0.4.2, I notice the program is quite unstable for simultaneous meters. For example, if you cancel a read in the middle of a simultaneous read by pressing ESC, often the program will just crash.

Thanks.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Display Calibration
AVS › AVS Forum › Display Devices › Display Calibration › HCFR - Open source projector and display calibration software