Shot Histogram Request - page 21 - CHDK Releases - CHDK Forum
« previous next »
Pages: 1 ... 16 17 18 19 20 [21] 22 23 24 25 26 ... 47   Go Down

Shot Histogram Request

  • 467 Replies
  • 129129 Views
*

Offline srsa_4c

  • ******
  • 4451
Re: Shot Histogram Request
« Reply #200 on: 30 / March / 2013, 14:57:25 »
Advertisements
Quote from: lapser on 30 / March / 2013, 14:42:15
Line 326 of the script says:
  tvd0=tdv -- for log

It should say tvd0=tvd
Better :)
It now appears to work (shoots continuously, exposition is adjusted).
Logged
*

Offline lapser

  • *****
  • 1093
Re: Shot Histogram Request
« Reply #201 on: 30 / March / 2013, 15:09:34 »
Quote from: srsa_4c on 30 / March / 2013, 14:57:25
Better :)
It now appears to work (shoots continuously, exposition is adjusted).
Great!

I wonder what the focus parameter would do if you set it to 1? That's supposed to set it to the hyperfocal distance, but it tries to go into manual focus first. I need to figure out how to focus correctly on cameras without manual focus.

The minimum and maximum ISO will probably be a little off. The first line of the script should be different for each camera. I think I can figure out that info in my conversion routines in the future.

svdiff=59 -- real to market difference

Thanks for the help in getting it to work on a new camera!
Logged
EOS-M3_120f / SX50_100b / SX260_101a / G1X_100g / D20_100b
https://www.youtube.com/user/DrLapser/videos
*

Offline ahull

  • *****
  • 634
Re: Shot Histogram Request
« Reply #202 on: 30 / March / 2013, 16:28:37 »
Thanks for your help. Sadly too late for this evening. I should get another chance to try it tomorrow.

Meanwhile, here is yesterdays effort with a different script. (Quite a few exposure issues as you can see).

Test timelapse from SD-300 (Ixus 40) using CHDK
Logged
*

Offline lapser

  • *****
  • 1093
Re: Shot Histogram Request
« Reply #203 on: 30 / March / 2013, 19:06:14 »
Quote from: ahull on 30 / March / 2013, 16:28:37
Meanwhile, here is yesterdays effort with a different script. (Quite a few exposure issues as you can see).
You have a beautiful spot for time lapsing. Is there much light pollution at night? Where are you located?

The online specs for the camera says it can take over 2 shots per second in continuous mode. If you enter an interval of "5", that would be 0.5 seconds per shot, or 2 shots per second. The first number printed on the screen is the time left over. If it's positive, it means you're shooting at 2 shots per second. When it goes negative, it's not keeping up with the shot rate.

Be sure that you set the review time to 0 seconds, or that will slow you up. I try to set everything I can in the camera's menus to off. I use "cloudy" for white balance with sunsets, even on sunny days. Auto white balance doesn't look good with time lapse.

I just did the "smoothing" factor, and used it once, but it looks like it worked. I used a smoothing factor of "8" for sunset.

Most of my time lapses are of the computer screen. You can go back and forth from the screen to a dark spot and watch the exposure follow, if it's working right.

Setting the log file to "1" is what I recommend. It saves a log with an 8 digit name of "month-day-hour-minute.LOG" The modified print_screen(n) function names the log file "nnnnnnnn.LOG" if n>9999.

I'm looking forward seeing your future creations. Good luck.
Logged
EOS-M3_120f / SX50_100b / SX260_101a / G1X_100g / D20_100b
https://www.youtube.com/user/DrLapser/videos
*

Offline ahull

  • *****
  • 634
Re: Shot Histogram Request
« Reply #204 on: 31 / March / 2013, 06:46:59 »
Hi lapser, that timelapse was shot facing north, the hills in the distance form part of the "Highland boundary fault" dividing the Scottish Highlands from the Lowlands. Light pollution is more of a problem to the south where the light from Edinburgh and Glasgow creeps over the hills.

Looking north we do sometimes see the northern lights, but I haven't as yet managed to film them.

I have a battery pack setup which I intend to attach to a camera in a weatherproof box and set up on a fencepost not far from here. Probably I will use that little Ixus 40, as it gives good results, and gives good battery life. I can shoot about 1000 frames on the internal NB-4L battery.  I do have a couple of other options when it comes to CHDK cameras, but based on previous results, I will probably stick with this set up. Another advantage of the Ixus models is that they are well built, with good optics and sensors, and can shoot images suitable for HD timelapses.

Furthermore, they can be picked up for very little on e-bay, so in the unlikely event that someone pinches it, while it is unattended in some remote spot, I am not too much out of pocket.

Not sure if I will get a chance tonight, but I do intend to test your script as soon as I get an opportunity. I'll post the results here of course.
« Last Edit: 31 / March / 2013, 06:54:06 by ahull »
Logged
*

Offline ahull

  • *****
  • 634
Re: Shot Histogram Request
« Reply #205 on: 31 / March / 2013, 12:01:31 »
 ??? Hmmm now we get one shot, then the camera crashes. I will check everything later.  srsa_4c I presume it worked for you on an Ixus 40, if so, must be something odd on my setup.
Logged
*

Offline srsa_4c

  • ******
  • 4451
Re: Shot Histogram Request
« Reply #206 on: 31 / March / 2013, 12:06:12 »
Quote from: ahull on 31 / March / 2013, 12:01:31
??? Hmmm now we get one shot, then the camera crashes. I will check everything later.  srsa_4c I presume it worked for you on an Ixus 40, if so, must be something odd on my setup.
There's a known way to crash the cam, by setting a too low ISO override (the source mentions crash below ISO 58).
My script params were as attached, cam in P mode, continuous drive.
« Last Edit: 31 / March / 2013, 12:31:18 by srsa_4c »
Logged
*

Offline lapser

  • *****
  • 1093
Re: Shot Histogram Request
« Reply #207 on: 31 / March / 2013, 12:28:38 »
Try setting all the overrides to off. My script is setting all the exposure parameters, so it might be conflicting with the CHDK overrides.

The script recognizes the camera exposure compensation setting, so if you want the entire sequence darker or lighter, you can use that.

Try setting infinity focus on the camera before starting the script. Don't set the focus parameter in the script. I think it might crash the camera.

I also turn as much as I can off in the camera menus, like IS and such. The mode should not be "auto". Try it with Av (aperture priority) if the camera has that feature. I think P mode should also work, since everything is in half shoot. But it will use the aperture setting of the first shot. You probably want the widest aperture possible. Maybe you could put your hand over the camera for the first shot to get the widest aperture (infinity focus mode), and that would work.
Logged
EOS-M3_120f / SX50_100b / SX260_101a / G1X_100g / D20_100b
https://www.youtube.com/user/DrLapser/videos
*

Offline lapser

  • *****
  • 1093
Smoothing Test
« Reply #208 on: 31 / March / 2013, 13:21:34 »
I went to a dark place last night (literally, not figuratively) and pointed the G1X at the sky while it went from light to dark.

The smoothing algorithm worked well to keep the exposure changes to 0 or -1 for each shot. There's no back and forth oscillations like I had before. The -1 changes do tend to bunch up together in groups of 3 because I don't do any changes if the exposure is within plus or minus 2, to prevent oscillations.  I think I can modify this so it's OK to do a -1 or +1 change if the last change was in the same direction. This should prevent the oscillations, and the bunching up in groups of 3 (I think). The bunching up is more visible at the beginning of the video when the exposure is changing more slowly.

[EDIT] After a little more thought, I realized that the oscillations and bunching up of -1 changes are due to the inaccuracy of the shutter speed changes. A -1 change in Tv96 is not precise, which shows up in the shot meter, and "pumping" of the dark areas in the video.  But a change of -1 in Sv96 is much more precise, and doesn't produce the "pumping" or "bunching". The script switched to Sv96 changes and held shutter speed at 1/2 second during the middle part of the video. That's why it "pumps" more at the beginning.

I've attached the log file. I added a description in the file of what the numbers mean. Here's the (kind of boring) video:

http://www.youtube.com/watch?v=lbBMdFa9Cmk#ws
« Last Edit: 31 / March / 2013, 14:15:41 by lapser »
Logged
EOS-M3_120f / SX50_100b / SX260_101a / G1X_100g / D20_100b
https://www.youtube.com/user/DrLapser/videos
*

Offline ahull

  • *****
  • 634
Re: Shot Histogram Request
« Reply #209 on: 31 / March / 2013, 15:30:20 »
Setting up the camera *exactly* as per srsa_4c's settings (I assumed P mode was Portrait, as this cam has an Auto and Manual mode, but nothing specifically called P If this is not the correct mode srsa_4,  let me know)... and it works  :D. It seems a little flakey if I use any other settings.

By the time I had checked everything, it was just past sunset time, so I am going to let it shoot a little twilight, for as long the battery lasts. I just sat it on the fence post, no waterproof housing, slightly risky as there is a chance of frost. I notice the script doesn't switch off the display, not sure if this is possible, but it would help preserve the battery life a bit if it is. 

If I get any usable results I'll post them soon.

Edit:  :D Looks like it works (and having said there is little light pollution, I lied  :P, in the village there is some, hence the rosy pink glow in the foreground near the end of the sequence).


Test timelapse 2 from SD-300 (Ixus 40) using CHDK Test build (ixus40_sd300-100k-1.2.0-r2619_130330) and lapser's script.

PS. Ignore a couple of dodgy frames at the start of the video, not sure how they sneaked in there, not the fault of the script, entirely the fault of the camera man.
« Last Edit: 31 / March / 2013, 17:17:19 by ahull »
Logged
Pages: 1 ... 16 17 18 19 20 [21] 22 23 24 25 26 ... 47   Go Up
« previous next »
 

Related Topics