get_histo_range strangeness - page 6 - Script Writing - CHDK Forum supplierdeeply

get_histo_range strangeness

  • 55 Replies
  • 1817 Views
Re: get_histo_range strangeness
« Reply #50 on: 06 / June / 2022, 12:11:03 »
Advertisements
@philmoz

Quote
The ETTR hint is only calculated for under exposed images. I have tried a few ways of calculating for an over exposed image; but I can't find anything that I think is usable.

Like you, and others I'm sure, I've thought about this and concluded there is no general and robust way of calculating the number of stops one is overexposed.

The histogram is scene dependent and once you over expose the scene-specific information is lost, ie it gets added to the 100% saturated bin. The information is non recoverable, other than saying there are x% of the sampled points in the fully saturated bin. But I see no way of converting this stops/Ev.

Likewise, if the exposure is underexposed, a similar thing can be said, ie the scene specific information is lost.

The only two scenarios, where you can obtain an over or under exposure Ev value is when the histogram clearly has 0% in the highlight or shadow extreme bins.

IMHO and obviously ;-)

Cheers

Garry

Re: get_histo_range strangeness
« Reply #51 on: 06 / June / 2022, 13:05:47 »
@philmoz

Quote
The Ximr builds have an updated version of the experimental raw EV histogram.

Just updated my g5x and g7x.

G5x looks OK, but the g7x doesn't seem to want to play, ie: under/over histo warnings not showing or being very flaky; the zebra wont show.
« Last Edit: 06 / June / 2022, 14:05:19 by pigeonhill »

*

Offline Caefix

  • *****
  • 900
  • Sorry, busy deleting test shots...
Re: get_histo_range strangeness
« Reply #52 on: 06 / June / 2022, 13:35:11 »
... but the g7x doesn't seem to want to play, ie: under/over histo warnings not shows or being very flaky; the zebra wont show.
Edit2: SX270 same...   Other issue ...  ???
https://chdk.setepontos.com/index.php?topic=14550.msg148364;topicseen#new

Edit: Offtopic, but the unexplored interaction with my md_detect.c
https://chdk.setepontos.com/index.php?topic=405.msg147922#msg147922 seems not quite uninteresting...
« Last Edit: 07 / June / 2022, 15:21:45 by Caefix »
All lifetime is a loan from eternity.

*

Offline reyalp

  • ******
  • 13840
Re: get_histo_range strangeness
« Reply #53 on: 06 / June / 2022, 19:15:07 »
G5x looks OK, but the g7x doesn't seem to want to play, ie: under/over histo warnings not showing or being very flaky; the zebra wont show.
The range issue mentioned previously for G7 X also applies to Zebra, so you need to adjust the thresholds. The over/under warning on the histogram is also affected, but I don't think there's a workaround.
Don't forget what the H stands for.


Re: get_histo_range strangeness
« Reply #54 on: 07 / June / 2022, 13:45:20 »
G5x looks OK, but the g7x doesn't seem to want to play, ie: under/over histo warnings not showing or being very flaky; the zebra wont show.
The range issue mentioned previously for G7 X also applies to Zebra, so you need to adjust the thresholds. The over/under warning on the histogram is also affected, but I don't think there's a workaround.


Got it, many thanks.

*

Offline philmoz

  • *****
  • 3441
    • Photos
Re: get_histo_range strangeness
« Reply #55 on: 24 / June / 2022, 19:47:18 »
I updated the raw ev histogram to add an option to save the histogram data for each image file separately.
In playback mode the appropriate histogram can then be displayed for the currently selected image.
Note: this requires finding a variable with the current playback image number. It is only supported on the G12, G1X, G5X and G7X2 at the moment.

CHDK ports:
  sx30is (1.00c, 1.00h, 1.00l, 1.00n & 1.00p)
  g12 (1.00c, 1.00e, 1.00f & 1.00g)
  sx130is (1.01d & 1.01f)
  ixus310hs (1.00a & 1.01a)
  sx40hs (1.00d, 1.00g & 1.00i)
  g1x (1.00e, 1.00f & 1.00g)
  g5x (1.00c, 1.01a, 1.01b)
  g7x2 (1.01a, 1.01b, 1.10b)

 

Related Topics