[jdbimmer]



Quote:
Originally Posted by zoyd  /t/1393853/fork-of-hcfr-started-whats-needed/1320#post_22913422


Here is a new build which works with both my D3 and i1pro 2 using the ArgyllCMS 1.4 driver code libusb0 instead of winusb0. This version also has the D3 tweaks to improve low-light precision. Update your i1pro drivers from the Drivers directory and let me know if it works on win7.

Excellent, thanks Zoyd!

Works in Win 7 64 and Win 7 32 bit with i1Pro, DTP94, and original Monaco Optix (c5-ish).


Simultaneous measures work between the DTP94 (w/native xrite driver) and one Argyll device. It doesn't work with my 2 Argyll devices (Monaco Optix/i1Pro), but I rarely find use for this feature - I would rather read the same spot on the display for profiling anyway.

 

[vega509]



Quote:
Originally Posted by jdbimmer  /t/1393853/fork-of-hcfr-started-whats-needed/1320#post_22914552


Excellent, thanks Zoyd!

Works in Win 7 64 and Win 7 32 bit with i1Pro, DTP94, and original Monaco Optix (c5-ish).


Simultaneous measures work between the DTP94 (w/native xrite driver) and one Argyll device. It doesn't work with my 2 Argyll devices (Monaco Optix/i1Pro), but I rarely find use for this feature - I would rather read the same spot on the display for profiling anyway.

simultaneous works for my i1pro and my i1display pro

 

[jdbimmer]



Quote:
Originally Posted by vega509  /t/1393853/fork-of-hcfr-started-whats-needed/1320#post_22914654


simultaneous works for my i1pro and my i1display pro

That's good news. I'm planning on purchasing the D3 soon. Unfortunately, the $50 rebate expired.

 

[zoyd]

Here is a new build to play around with. It incorporates all of the changes that John had been working on, code optimization and reconfiguring the profiling options. I was able to measure and apply a matrix correction but couldn't manage to save or load one, you can manually enter a matrix correction if needed. I also made CIE94 the default deltaE metric, you can use the original (Luv) calculation by selecting it in the advanced preferences. Also, the ability to use vendor dlls has been reenabled so you run either Argyll or original drivers.


HCFRsetup_CIE94.exe

 

[vega509]



Quote:
Originally Posted by zoyd  /t/1393853/fork-of-hcfr-started-whats-needed/1320#post_22916212


Here is a new build to play around with. It incorporates all of the changes that John had been working on, code optimization and reconfiguring the profiling options. I was able to measure and apply a matrix correction but couldn't manage to save or load one, you can manually enter a matrix correction if needed. I also made CIE94 the default deltaE metric, you can use the original (Luv) calculation by selecting it in the advanced preferences. Also, the ability to use vendor dlls has been reenabled so you run either Argyll or original drivers.


HCFRsetup_CIE94.exe

is it possible to zip the HCFR folder? the install program wants to remove all the other versions.


thanks zoyd

 

[zoyd]

no problem - HCFR Calibration.zip

 

[vega509]



Quote:
Originally Posted by zoyd  /t/1393853/fork-of-hcfr-started-whats-needed/1320#post_22916568


no problem - HCFR Calibration.zip

I believe there may be a deltaE calculation error on the grayscale 0% and 10%.


3.0.4.0 cuv



3.0.4.0 cie94



the gamut deltaE looks a little off, but I haven't done the math.


thank you once again for all the time you have put into this.


vega

 

[Make73]



Quote:
Originally Posted by zoyd  /t/1393853/fork-of-hcfr-started-whats-needed/1320#post_22916212


...you can manually enter a matrix correction if needed...

Where i can do that?

Sensor matrix?


I hope we get matrix load/save option back because many use matrix at every measuring.


So i1D3 with plasma should calibrate/sync with 50% IRE window?

Earlier there was instructions that it should be done with 30% IRE window.

 

[zoyd]



Quote:
Originally Posted by vega509  /t/1393853/fork-of-hcfr-started-whats-needed/1320#post_22916613


I believe there may be a deltaE calculation error on the grayscale 0% and 10%.


the gamut deltaE looks a little off, but I haven't done the math.


thank you once again for all the time you have put into this.


vega

grayscale dE uses Luv only which is standard practice so switching between Luv/CIE94 the number should not change. 0% dE needs to be removed in future version.



You are right, gamut dE for CIE94 has an error, will fix later.



@Make73: Sometimes 30% isn't bright enough for the probe to sync, need to add a success indicator somewhere.

 

[vega509]



Quote:
Originally Posted by Make73  /t/1393853/fork-of-hcfr-started-whats-needed/1320#post_22916994


Where i can do that?

Sensor matrix?


I hope we get matrix load/save option back because many use matrix at every measuring.


So i1D3 with plasma should calibrate/sync with 50% IRE window?

Earlier there was instructions that it should be done with 30% IRE window.

you can enter the sensor matrix in the Measures Sensor Configure panel.

 

[Make73]



Quote:
Originally Posted by vega509  /t/1393853/fork-of-hcfr-started-whats-needed/1320#post_22917357


you can enter the sensor matrix in the Measures Sensor Configure panel.

Ok, but if it could be possible to add option like it was "Manually edit..." Load/Save .mhc file

Now it don´t save matrix with .chc file

 

[zoyd]

It's difficult to go backwards, I'll see what can be done with the current scheme. In the meantime I have updated the previous link zip file with a new testing version.


changes:


-fixed dE CIE94

-fixed dE calculation in luminance/sat measures pages

-fixed luminance calculation in luminance/sat graph

-removed black level dE calculation

-added BT.1886 gamma target (replaces previous "optimized regression")


note that the BT.1886 target values use your white and black level measurements from the grayscale page. If you don't trust your meter to give you a good black level reading then enter it manually in the 0% stimulus Y box.



HCFR 20Calibration.zip

 

[rahzel]

I have to agree with Make73... the load ccmx option was nice. But I really appreciate the work you (and John) have done zoyd, thanks again.


So in your 'CIE94' versions, HCFR will use CIE94 for color gamut and CIELUV for grayscale? (edit: this seems to be the case, because my grayscale measurements are the same, but my color gamut numbers are different)

Also, BT.1886 is new to me. With this option, my original calibration targeting 2.35 is now ~2.5! Is this normal? Should we be targeting a similar range (ie 2.2-2.4) using bt.1886?

 

[LarryInRI]

rahzel,


Here is a link that will give you more than enough info about BT.1886.
http://www.avsforum.com/t/1409045/how-power-law-gamma-calibration-can-lead-to-crushed-blacks


Larry

 

[rahzel]



Quote:
Originally Posted by LarryInRI  /t/1393853/fork-of-hcfr-started-whats-needed/1350#post_22919779


rahzel,


Here is a link that will give you more than enough info about BT.1886.
http://www.avsforum.com/t/1409045/how-power-law-gamma-calibration-can-lead-to-crushed-blacks


Larry

Thanks Larry. Ya, I've browsed through that thread before, but not in great detail. I guess I was just lazy and wanted a quick answer.

But I'll give it a read.

 

[LarryInRI]

rahzel,


Then try this thread: http://www.avsforum.com/t/1450925/what-is-bt-1886-gamma .


Larry

 

[zoyd]



Quote:
Originally Posted by rahzel  /t/1393853/fork-of-hcfr-started-whats-needed/1350#post_22919671


I have to agree with Make73... the load ccmx option was nice. But I really appreciate the work you (and John) have done zoyd, thanks again.


So in your 'CIE94' versions, HCFR will use CIE94 for color gamut and CIELUV for grayscale? (edit: this seems to be the case, because my grayscale measurements are the same, but my color gamut numbers are different)

Also, BT.1886 is new to me. With this option, my original calibration targeting 2.35 is now ~2.5! Is this normal? Should we be targeting a similar range (ie 2.2-2.4) using bt.1886?

yes, all grayscale dE is done in CIELuv space and color in CIE94 unless you select the option to revert back to CIELuv for color. When you select BT.1886 gamma reference it does two things.


1. Draws a BT.1886 reference gamma curve (white squares) on the gamma plot, you should try and get yours to lie on top of that if possible.

2. Calculates the target Y in the gray scale measures page in order to achieve 1.


If you have a very low mll the target curve will be almost flat at gamma=2.4, if you have a higher mll the target curve will dip down at the lower stimulus levels. This will help to avoid crushed blacks but you need a multipoint gamma adjustment to do this. Remember that you need to take your viewing environment into account as well, a 2.4 gamma may not be suitable if there is much ambient light. If all you have is a single gamma control I would shot for an average of the BT.1886 curve for your display.

 

[rahzel]

My ST50 has an MLL of ~0.0028fL. I prefer a higher gamma because my room has controlled lighting.


I only have a 2pt system with a gamma slider so I'm aware that I'm not able to get my gamma curve to lineup with the bt.1886 curve. I had my slider at 2.6 before, which made the gamma around 2.33avg using the power law formula. With the bt.1886 formula, it puts it around 2.5avg. Setting the slider to 2.4 puts the gamma closer to the bt.1886 target (which is ~2.32 avg). So the bt.1886 gamma curve is fixed? You can't change the target? When I change the target in HCFR, it doesn't move the reference curve. I know my ST50 doesn't have a multipoint adjustment, but my LG LCDs do and I plan on re-calibrating them soon. And again, since I have controlled lighting, I prefer a higher gamma.


Anyway, here is the gamma curve for my ST50 using the bt.1886 formula:

Quote:
Originally Posted by LarryInRI  /t/1393853/fork-of-hcfr-started-whats-needed/1350#post_22919994


rahzel,


Then try this thread: http://www.avsforum.com/t/1450925/what-is-bt-1886-gamma .


Larry

Thanks again Larry.

 

[vega509]

There is an error when trying to save the meter correction file.




I have tried every directory in the HCFR folder with the same error.

 

[zoyd]



Quote:
Originally Posted by rahzel  /t/1393853/fork-of-hcfr-started-whats-needed/1350#post_22920534


So the bt.1886 gamma curve is fixed? You can't change the target?

It's determined by your white and black level but will never exceed 2.4. With BT.1886 reference you have to think about gamma calibration a little differently. With the other options (display, display+black compensation etc.) you are fitting functions to your measurements to yield a single parameter (gamma). With BT.1886 you are adjusting Y(x) to match a predetermined function that is calculated based on your white and black level.