JVC Autocal Software V12 calibration for 2019 RS1000/RS2000/RS3000/NX5/NX7/NX9

[Manni]

Well, since I was mentioned here, I can also answer it directly.

If you look at a matrix in HCFR you put the values from left to right and top to bottom in the xml file
My xml file looked like this:

  <matrixOfSensor>
    <double>0.02063751220703125</double>
    <double>0.00086295604705810547</double>
    <double>0.000039804726839065579</double>
    <double>-0.00034344196319580078</double>
    <double>0.021785736083984375</double>
    <double>0.00023213028907775876</double>
    <double>-0.0003108978271484375</double>
    <double>-0.00026074051856994629</double>
    <double>0.021076202392578125</double>
  </matrixOfSensor>

(Spyder pointed towards Lens, reference meter pointed towards screen)

Thanks, but for those of us using Calman or Lightspace, we need to know the format of the file, i.e. which field of the matrix goes where. I assume it’s in R G B order. Is that what you follow with the Klein or HCFR?

And by the way, thanks for sharing, I can’t believe it was here from V7...

[KarlKlammer]

As a Table, the lines in the xml file would look like this.
It is in R-G-B order.



BTW, the guy who wrote the i1 Display DLL for Autocal pointed me to the xml file some time ago. So I don't claim to have found this. I just experimented with it last week to verify that it really works. It was with Autocal 11. Later I reinstalled Autocal 7. And matrixOfSensor was also there, though I can't verify that it works for Autocal 7.

For those who did not see my screenshots in the other forum, here they are again:

Autocal for Gamma and Color with Spyder without matrix



Autocal for gamma and color with Spyder and matrix

[jj-34]

Is that right that you made your correction matrix with simultaneous measures and the spyder X facing the projector and the Klein facing the screen ?

[jj-34]

@Manni : I made the #1000 reply in this forum, did I win something ......? 

[KarlKlammer]

Is that right that you made your correction matrix with simultaneous measures and the spyder X facing the projector and the Klein facing the screen ?

It was a Spyder 5. But yes, the rest is, how I did it.

[David Vaughn]

It was a Spyder 5. But yes, the rest is, how I did it.

Karl,

Can you write detailed instructions on how to do this? I "think" I know how I can do it, but I'm not 100% sure. 

From your charts what it looks like is you can do an autocal that's essentially 99.9% perfect, which eliminates the need to do a manual calibration. Is this correct?

[jj-34]

It was a Spyder 5. But yes, the rest is, how I did it.

I see, that explains the sensor facing the projector.

[dcchan2]

From your charts what it looks like is you can do an autocal that's essentially 99.9% perfect, which eliminates the need to do a manual calibration. Is this correct?

I was also able to get excellent results using this method.  Note, however, it is just a convenience and cannot improve the colour autocal beyond what you can do with the i1Pro2 directly; it has no effect on gamma autocal.

[David Vaughn]

Is the benefit to be able to do color and gamma at the same time? So essentially it's just a time saver?

[KarlKlammer]

Karl,

Can you write detailed instructions on how to do this? I "think" I know how I can do it, but I'm not 100% sure.

From your charts what it looks like is you can do an autocal that's essentially 99.9% perfect, which eliminates the need to do a manual calibration. Is this correct?

I'll try it (English is a foreign language for me. So please excuse my mistakes )

For this procedure to make sense, you must have access to a meter that is more accurate than a Spyder.
In my case it is a K10-A profiled to an i1 Pro 2.

I admit, that I may have caused some confusion by writing that I had my Spyder facing the projector. Of course a Spyder X will have to be placed facing the screen. And then the matrix will definitely look more like jj-34 has shown. But since Autocal <12 needs the Spyder 5 facing the projector, I found it more practical to make the measurements for the matrix the way I wrote earlier.

An XYZ correction matrix will tell Autocal, how to correct the measurement data it receives from the Spyder.
So the first step is to create this matrix. There are several options. I can't go into much detail here, since my workflow is based on LightSpace/ColourSpace and the Klein SDK, and I don't have much experience with other solutions. Maybe someone else can chime in.
There is of course HCFR where you can build a matrix based on comparison measuerments between the Spyder and a reference meter. jj-34 has shown the window containing the matrix a few posts earlier.
CalMan also has the option to create a matrix.
If you only have access to raw measurements from Spyder and reference there is an Excel sheet, I found in the other forum. And Calman also has an external tool called "CalMAN 2018 Four Color Matrix Tool".

So once you have built the correction matrix, you have to put the nine values in the table into the lines of xml file in your windows user profile. If you have done this, you are ready to go and can use Autocal the way you usually do.
When you have a big luminance difference like in my matrix, the Spyder will most likely be shown outside the rectangle in the Autocal meter placement window. But based on my experience, you can disregard that unless Autocal shows 0 measurements in its log.

If you are a perfectionist, there is always some room for improvement. But I think the result is very good. My verification measurements after Autocal showed an average dE 2000 of 1.17 and a maximum of 1.74.

[Manni]

I can't go into much detail here, since my workflow is based on LightSpace/ColourSpace and the Klein SDK

Thanks for the detailed explanation, I'm sure it will help many who are not familiar with using a matrix correction for profiling.

As your workflow is with Lightspace/Colourspace, please could you share the exact correspondance between your Colourspace matrix and the xml file? Just a screenshot of both with the same values would be enough.

I understand that you're using the Klein tool to "correct" the Klein with the i1pro2, but then you must be creating a matrix in Colourspace with the Klein as the reference and the Spyder as the field meter. That's the matrix correction I'm interested in here, as well as the corresponding values in the XML.

From there, we can confirm the format of the XML file.

Having just the values in the XML file doesn't really help if we don't know what the values mean.

I'm sure we'll fine with trial and error, but having this info would save time.

Thanks again!

[Manni]

Is the benefit to be able to do color and gamma at the same time? So essentially it's just a time saver?

If you have a Spyder and an i1pro2, yes, it's just about convenience and time saving, but if you have a real reference spectrometer (which the i1pro2 or the unprofiled Klein aren't), for example a Jeti1511, which isn't supported by the Autocal, it means you can get the Autocal to produce near reference results with the Spyder.

Of course, the other advantage is for those who don't have an i1pro2 but and a better meter than the Spyder that isn't supported by the Autocal. That means they can use their i1d3 or their Colormunki or whatever to correct their Spyder and get much better results with the Autocal. Before the Autocal started supporting the i1pro2, the only way to correct the Spyder was to use a spreadsheet method, and that was a bear...

[dcchan2]

For this procedure to make sense, you must have access to a meter that is more accurate than a Spyder.
In my case it is a K10-A profiled to an i1 Pro 2.

Is there any specific reason for using the K10-A (profiled to an i1Pro2) to profile the Spyder, instead of using the i1Pro2 directly?

[Manni]

Is there any specific reason for using the K10-A (profiled to an i1Pro2) to profile the Spyder, instead of using the i1Pro2 directly?

It makes sense to profile the K10A to the i1pro2 in order to use the Klein in general, but in this case no, in most cases there is no benefit to go through the K10A. You would get the same or better results profiling the Spyder to the i1pro2 directly, but I guess if the Klein is already profiled to the i1pro2 using the Klein tools, it's faster to just use the Klein as a reference meter if you don't want to get the i1pro2 out. You shouldn't lose much if the first profile was good, but you won't gain anything purely for profiling.

The i1pro2 can struggle to read some colors (especially red and blue) even with the 100% patches used for matrixing, but whatever you do to resolve this when creating the Klein profile (zooming down, etc) you could do when creating a Spyder profile directly from the i1pro2.

[KarlKlammer]

I did not use the i1 Pro 2 directly, because the K10-A already has a correction profile saved in its internal memory (and I was too lazy unpacking it). So I use the internal profile and not a LightSpace/ColourSpace profile.

The whole procedure was an experiment. I usually don't do a color Autocal. Instead I let my Radiance and two LUTs take care of everything except the gamma drift.



The reason I mentioned the Klein SDK was that I hijacked a tool that I have written based on it to create the matrix.

So here is a screenshot of the tables in my tool.

JVC Autocal Software V12 calibration for 2019 RS1000/RS2000/RS3000/NX5/NX7/NX9

JVC Autocal Software V12 calibration for 2019 RS1000/RS2000/RS3000/NX5/NX7/NX9

As I am only hijacking it, the captions don't match the tables. So the lower left table contains the imported K10-A WRGB measurements from ColourSpace (already corrected due to the internal profile). The lower right table contains the imported Spyder WRGB measurements. The upper table contains the matrix that the Klein SDK has calculated from the lower two tables.




Edit: I added red numbers in the cells in the upper table of the first screenshot to match the lines in the xml file.