Username:

Password:

UKAI is moving to Facebook

Author Topic: Which DSS 32 to 16 bit settings?  (Read 2051 times)

0 Members and 1 Guest are viewing this topic.

Offline fred555

  • New Member
  • *
  • Posts: 17
Which DSS 32 to 16 bit settings?
« on: 19:14:05, 11 January, 2012 »
How does one determine which method and which settings to use when converting the autosave image from DSS in Photoshop?

I have tried mucking about with the local histogram method and the default exposure and gamma method. I am not really sure what I am doing and do not feel I am optimising the initial 16 bit Tiff.

I would gratefully receive any advice.

Cheers

Offline Einari

  • Poster God
  • *****
  • Posts: 2970
    • Einari's astro pages
Re: Which DSS 32 to 16 bit settings?
« Reply #1 on: 19:46:09, 11 January, 2012 »
I don't change anything when converting from 32 to 16 bit in PS.
But YMMV.
Tapio

Celestron C8, C9.25" & Hyperstar, iOptron CEM60-EC, TS 80 triplet,  Coronado PST, SXV-694,  ASI120MM/ASI174MC Cool

Offline Ian Straton

  • Poster God
  • *****
  • Posts: 1131
    • Stoneage Observatory blog
Re: Which DSS 32 to 16 bit settings?
« Reply #2 on: 20:22:03, 11 January, 2012 »
I did some experimentation on this recently (there is thread about my tests on here somewhere) short version: if you combined subs of varying length exposures in dss then "local adaption" will give you really fine control over the tone mappings during the 32 -> 16 bit conversion process.  If you combined images of all the same duration then use the "equalise histogram" option. 

The exposure and gamma option offers poor contrast control relative to local adaption and the highlight compression option will completely knacker your image, so don't use it... ever...

The above statements are based on a series of tests in which I edit the image in 32 bit using whatever controls were available with each option to a highly rigorous standard (ie that looks pretty good to me..) and then exporting to 16 bit and doing a quality comparison (basically looking at the histograms to see if they had gone all spiky.. I generally find that a spiky histogram is very bad for further procesing).   One other outcome of this set of experiments was to not do the conversion in dss, I assume its resampling algorithims are not as sophisticated as Photoshop but certainly it did not produce as good an outcome as doing the conversion in PS.

Offline Roger B

  • Administrator
  • Galactic Poster
  • *****
  • Posts: 9010
    • Capricorn1 observatory
Re: Which DSS 32 to 16 bit settings?
« Reply #3 on: 21:48:39, 11 January, 2012 »
I don't change a thing either from 32 to 16.

Roger     
Scopes - TS65q, W/O 110 Triplet
Mount - EQ6 PRO via  Eqmod,  Rowan belt modded
Cameras - SX694, QSI583WSG,Lodestar Guidecam, GP-Cam, QHY5II.
TS 9mm OAG
Quantum 7x1.25 EFW, Baader LRGB,Ha,OIII,SII Filters 
Quark chromo

SGPro, Pixinsight.

************AND NO TIME TO USE IT************

Offline Ian Straton

  • Poster God
  • *****
  • Posts: 1131
    • Stoneage Observatory blog
Re: Which DSS 32 to 16 bit settings?
« Reply #4 on: 08:55:23, 12 January, 2012 »
Roger and Einari, when you say you don't change a thing that literally cannot be true, you are discarding half the bit depth of the image which is a massive amount of data, something tells PS what to discard, I suspect you mean you just accept the default options when going image>mode>16bit ?  In which case you are accepting neutral settings on the exposure and gamma conversion method, thats fine but depending on what you stacked you could be getting much better results by using one of the other methods or even by adjusting the sliders in exposure and gamma.


Offline Einari

  • Poster God
  • *****
  • Posts: 2970
    • Einari's astro pages
Re: Which DSS 32 to 16 bit settings?
« Reply #5 on: 18:12:36, 12 January, 2012 »
After reading this I experimented with some autosave.tifs with different conversion method (what was suggested here). Couldn't find difference but I have spared the wisdom so I will try again.
Tapio

Celestron C8, C9.25" & Hyperstar, iOptron CEM60-EC, TS 80 triplet,  Coronado PST, SXV-694,  ASI120MM/ASI174MC Cool

Offline Ian Straton

  • Poster God
  • *****
  • Posts: 1131
    • Stoneage Observatory blog
Re: Which DSS 32 to 16 bit settings?
« Reply #6 on: 19:12:41, 12 January, 2012 »
Einari, it very much depends on what you have stacked, if all your subs have the same exposure then you are not adding to the bit depth when you stack them so 16 of the available 32 bits per channel are essentially zero so it doesn't matter (except for the highlight compression option, this thing sucks!) if you have stacked different exposures then you will extended the dynamic range in the image beyond the 14 or 16 bit depth of your camera, then the conversion method becomes really important because you are in the realm of High Dynamic Range images and they need very different handling.  (see the various HDR threads currently doing the rounds in any other images..)

Offline Einari

  • Poster God
  • *****
  • Posts: 2970
    • Einari's astro pages
Re: Which DSS 32 to 16 bit settings?
« Reply #7 on: 06:43:55, 13 January, 2012 »
Ian, I thought so too.
Tapio

Celestron C8, C9.25" & Hyperstar, iOptron CEM60-EC, TS 80 triplet,  Coronado PST, SXV-694,  ASI120MM/ASI174MC Cool

 

ukbuysellRemote Imaging from AustraliaSharpSkyblank APTUKAI on Facebook
TinyPortal © 2005-2012