582 replies to this topic

[lock042]

 

 

 

This is awesome! Will it be applicable to sequences, too? So that one could apply it on calibrated lights and then do auto per-image background extraction?

 

Now this is done!!

Attached Thumbnails

• 

[BQ Octantis]

Hi Cyril,

 

Can you point me to the right build? I downloaded the latest in the pipeline for "master", but it refuses to run on an admin account on Apple silicon or at all on Intel…

 

BQ

[lock042]

This is because the app is not signed.
You need to use th xattr command

[Domdron]

Now this is done!!

Super, will try asap.

[BQ Octantis]

This is because the app is not signed.
You need to use th xattr command

I did, which is how it is able to run in the admin account. But it does nothing when I try to fire it up in the user account. This command is unnecessary in Catalina (or High Sierra)—you just right-click and select "Open".

 

Oddly enough, when I retried on my Intel test Mac this morning it opened just fine! I'll have to get time on my wife's laptop to check the M2 again…

 

The CCM panel works as expected. It just so happens yesterday's sky was almost ideal D65 conditions (practically cloudless, with the noon sun at 50°), and I was itching to get a shot of my CC24 facing straight up from a ladder in the far field:

 

 

However, matrix entry is much more tedious than pixel math—so the key trade-off over pixel math is being able to apply this to each sub. Entries are limited to 8 digits (including signs, decimals, and e+0x characters); pasting longer values results in truncation vice rounding. There are presets, but none are anything I would use. I have a text file with the pre-formatted pixel math for 16 different scenarios (by lens, sRGB, Planckian, AdobeRGB…), so if I could save my own presets this panel would be a force multiplier. Lastly, CCMs are often normalized to separate out the white balance factors, but this panel combines them. That's not an issue for me—my color calibration script spits out all:

 

M =

   70.2603  -19.9099    7.0141
   -6.8682   23.3685   -9.5739
    1.4687   -7.7844   31.3205

M_prime =

   2.080411  -1.418848   0.338438
  -0.203368   1.665323  -0.461955
   0.043490  -0.554746   1.511256

K = 14.032
kDkF =

   2.4067   1.4769

 

But it would be handy to be able to tweak the gain and r/g, b/g factors directly.

 

Where would you like this feedback?

 

BQ

Edited by BQ Octantis, 17 March 2024 - 07:25 AM.

[BQ Octantis]

Oddly enough, when I retried on my Intel test Mac this morning it opened just fine! I'll have to get time on my wife's laptop to check the M2 again…

Since everyone was asleep, I just took the M2. Turns out, if I fire up 1.3.0 first and then quit out of it, 1.3.1 will launch.

[lock042]

We could increase the number of digits if needed. However, I'm not sure it will make any difference with the rounding errors.

[BQ Octantis]

Nah, I can eliminate the leading zero to get enough digits. The only issue is having to type those buggers out every time I want to use a CCM! I currently cut and paste lines into pixel math that correspond to each row.

 

BQ

Edited by BQ Octantis, 19 March 2024 - 03:50 PM.