Access Autofocus Parameters? - page 2 - Script Writing - CHDK Forum
« previous next »
Pages: 1 [2]   Go Down

Access Autofocus Parameters?

  • 17 Replies
  • 6958 Views
*

Offline Microfunguy

  • ****
  • 4037
    • StereoData Maker
Re: Access Autofocus Parameters?
« Reply #10 on: 13 / April / 2013, 09:08:17 »
Advertisements
Quote from: Davo on 13 / April / 2013, 07:32:20
my A810 seems to be outperforming my astro cam that cost 6 times more  :D

Just keep an eye on those coronal mass ejections, the BIG one cannot be far away  :)
Logged
*

Offline reyalp

  • ******
  • 14140
Re: Access Autofocus Parameters?
« Reply #11 on: 13 / April / 2013, 18:10:14 »
Quote from: Davo on 12 / April / 2013, 05:56:05
It occurred to me that the camera must already have a similar routine built in - if only I could find it. I will keep digging.
Of course this must be true. This could be a good project for someone who want to get into reverse engineering the firmware.

One place where this information must be evaluated is in the canon firmware autofocus code. Some signposts might be found in the firmware by looking for strings related to "AF*"
I see "AFScanDriver.c", "AFDataManager.c", "AFIntSrvTask, "AFInteg.c", "SinglAF.c", "AFParam.c", "AFCalc.c" etc.
Also nearby "Peak Exist" "No Peak" (in an AFCalc.c function)
Logged
Don't forget what the H stands for.
*

Offline Microfunguy

  • ****
  • 4037
    • StereoData Maker
Re: Access Autofocus Parameters?
« Reply #12 on: 13 / April / 2013, 19:40:31 »
You could locate a bright star exactly in the centre of the lcd screen (use a cross-hairs grid with open centre).
Detect the pixel location of the star.
Detect the number of pixels above nominal black level around the brightest pixel.
Display that number.
Adjust telescope focus until that number is a minimum.
Logged
*

Offline Davo

  • ***
  • 189
Re: Access Autofocus Parameters?
« Reply #13 on: 14 / April / 2013, 07:22:45 »
Quote
Detect the pixel location of the star. Detect the number of pixels above nominal black level around the brightest pixel.

You should become a software designer - that is exactly how one of the established focus meters works. Unfortunately, the star I observe most tends to fill the LCD screen  ;)

Quote
This could be a good project for someone who want to get into reverse engineering the firmware.

I think I will give this a go based on the pointers you gave. Also, I really want to see what sort of control there is over video recording. Stand by for heaps of noobie developer questions.
Logged
*

Offline Microfunguy

  • ****
  • 4037
    • StereoData Maker
Re: Access Autofocus Parameters?
« Reply #14 on: 14 / April / 2013, 07:49:19 »
Quote from: Davo on 14 / April / 2013, 07:22:45
the star I observe most tends to fill the LCD screen

???
How can that possibly be, a star is effectively a point source and you see its diffraction image.
Are you saying the diffraction spikes almost fill the LCD screen ?

Quote
I think I will give this a go

It is human nature to encourage rather than discourage.
Believe me, you could spend an enormous amount of time on this and not get anywhere.
I would estimate your chances of success as less than one percent but if you have plenty of time available prove me wrong .  :)
Logged
*

Offline ahull

  • *****
  • 634
Re: Access Autofocus Parameters?
« Reply #15 on: 14 / April / 2013, 08:45:35 »
microfunguy, I think the star in question is relatively close  ;) best observed on a bright sunny day. 
« Last Edit: 14 / April / 2013, 08:48:27 by ahull »
Logged
*

Offline Microfunguy

  • ****
  • 4037
    • StereoData Maker
Re: Access Autofocus Parameters?
« Reply #16 on: 14 / April / 2013, 09:22:55 »
Quote from: ahull on 14 / April / 2013, 08:45:35
I think the star in question is relatively close  ;)

Aaaahh ............ as they say in the US of A  "my bad".   :)

(my bad what I am not sure).

Logged
*

Offline outslider

  • *****
  • 762
Re: Access Autofocus Parameters?
« Reply #17 on: 14 / April / 2013, 15:20:21 »
Quote from: Microfunguy on 14 / April / 2013, 07:49:19
Believe me, you could spend an enormous amount of time on this and not get anywhere.
I wonder many researchers have heard the phrase starting new project:)
Logged
if (2*b || !2*b) {
    cout<<question
}

Compile error: poor Yorick
Pages: 1 [2]   Go Up
« previous next »
 

Related Topics


SimplePortal © 2008-2014, SimplePortal