Experimental Photo Production SX110/120/130, A470 - page 2 - Creative Uses of CHDK - CHDK Forum supplierdeeply

Experimental Photo Production SX110/120/130, A470

  • 18 Replies
  • 6296 Views
*

Offline c_joerg

  • *****
  • 1147
Re: Experimental Photo Production SX110/120/130, A470
« Reply #10 on: 18 / August / 2017, 02:04:44 »
Advertisements
It looks like you run more than 10000 shoots with one battery?

Your Interval is between 0.61s and 0.67s. So it changes around 10%. Not so much as I expected. In our case it is not the exposure time which makes the variance.  I think it’s more writing on the SD card (with different jpg sizes). On some of my SD cards the variance is much higher.
I would set the Interval to 0.75. Than you had a continuous interval.

@ reyalp
There is some other think in the log file which is unusual. The Aperture jumps from 8.264 to 3.029 (or ND Filter?). I never had seen this before. Not sure if this is a bug in the port or if the camera really changes aperture.

M100 100a, M3 101a, G9x II (1.00c), 2*G1x (101a,100e), S110 (103a), SX50 (100c), SX230 (101a), S45,
Flickr https://www.flickr.com/photos/136329431@N06/albums
YouTube https://www.youtube.com/channel/UCrTH0tHy9OYTVDzWIvXEMlw/videos?shelf_id=0&view=0&sort=dd

*

Offline c_joerg

  • *****
  • 1147
Re: Experimental Photo Production SX110/120/130, A470
« Reply #11 on: 18 / August / 2017, 02:07:09 »
It looks like the A470 has only one Temperature Sensor. The Optical and Sensor Temperature are the same. I have never seen this before.
M100 100a, M3 101a, G9x II (1.00c), 2*G1x (101a,100e), S110 (103a), SX50 (100c), SX230 (101a), S45,
Flickr https://www.flickr.com/photos/136329431@N06/albums
YouTube https://www.youtube.com/channel/UCrTH0tHy9OYTVDzWIvXEMlw/videos?shelf_id=0&view=0&sort=dd

Re: Experimental Photo Production SX110/120/130, A470
« Reply #12 on: 18 / August / 2017, 16:16:07 »
It looks like you run more than 10000 shoots with one battery?
To be sure I've found that my Canons (A470/SX110/120/130) will run for 200 minutes on two high class rechargeables - in my case eneloop black. That would make 20,000 photos on a pair of batteries. On this kayak tour I had to choose a place to stop and change the batteries, which is why I didn't quite let the batteries run out. But the next development is on its way - Fat Boy - six eneloop batteries (3 * parallel, 2 * serial) which should last for 10 hours.

Your Interval is between 0.61s and 0.67s. So it changes around 10%. Not so much as I expected. In our case it is not the exposure time which makes the variance.  I think it’s more writing on the SD card (with different jpg sizes). On some of my SD cards the variance is much higher.
I would set the Interval to 0.75. Than you had a continuous interval.
OK, I didn't realise that setting the interval would absolutely control the interval length. If so, then we'll see about next time (I wanted to extend the interval length in any case, 20 minutes of kayaking time lapse is far too much).
BTW I was accidentally shooting at minimal resolution (640 * 480). I'll try next time at full 6 MP, and see if an interval of 0.75 s copes.

Thanks for your feedback.

*

Offline reyalp

  • ******
  • 13387
Re: Experimental Photo Production SX110/120/130, A470
« Reply #13 on: 18 / August / 2017, 23:50:01 »
There is some other think in the log file which is unusual. The Aperture jumps from 8.264 to 3.029 (or ND Filter?). I never had seen this before. Not sure if this is a bug in the port or if the camera really changes aperture.
For cameras that only have an ND filter, it's normal for the value to be reflected in the "aperture" propcase.
Don't forget what the H stands for.


Re: Experimental Photo Production SX110/120/130, A470
« Reply #14 on: 08 / November / 2017, 11:07:58 »
Hi all,

Another example of what is possible with CHDK.

SX120 Powershot using CHDK for:
 - Edge overlay (original 3 shot panorama);
 - Exposure bracketing (9-fold, +1⅓/-1⅓ in ⅓ steps) of 5 panoramas using the edge overlays as guides;
 - Raw shooting.
 Align_image_stack/enfuse/hugin for panorama production.

Best

*

Offline c_joerg

  • *****
  • 1147
Re: Experimental Photo Production SX110/120/130, A470
« Reply #15 on: 09 / November / 2017, 01:36:50 »
Nice, but at some places (2:20) the video trembles (at least on my computer)
M100 100a, M3 101a, G9x II (1.00c), 2*G1x (101a,100e), S110 (103a), SX50 (100c), SX230 (101a), S45,
Flickr https://www.flickr.com/photos/136329431@N06/albums
YouTube https://www.youtube.com/channel/UCrTH0tHy9OYTVDzWIvXEMlw/videos?shelf_id=0&view=0&sort=dd

Re: Experimental Photo Production SX110/120/130, A470
« Reply #16 on: 09 / November / 2017, 10:34:41 »
Thanks, c_joerg, for the kind criticism. I am using ffmpeg to videofy image sequences:

ffmpeg -y -i im_%05d.jpg -c:v libxvid -qscale:v 3 -r 25 -pix_fmt yuv420p "{$name}$i OL.avi"

In my opinion lower quality xvid or alternatively the H264 codec leads to too much pixellation of photographic images; the judder that you observe is the cost of higher quality.

Any alternative suggestions?

Best, Lee

*

Offline c_joerg

  • *****
  • 1147
Re: Experimental Photo Production SX110/120/130, A470
« Reply #17 on: 10 / November / 2017, 03:21:27 »
Any alternative suggestions?
 

I have not much experience with ffmpeg Which framerate you are using? 25?
I’m using virtualdub to make a avi video.

Maybe there is a similar problem like this:

On my last videos, I made a mistake: the original video has 25fps. For the upload I used Premiere Elements. The upload function has unfortunately converted the video to 29.97fps and this is not really intelligent. Approximately every fifth frame was doubled.
The videos do not look as smooth on YouTube as the original video. This can be seen especially in zooming and fast movements.
« Last Edit: 10 / November / 2017, 05:38:24 by c_joerg »
M100 100a, M3 101a, G9x II (1.00c), 2*G1x (101a,100e), S110 (103a), SX50 (100c), SX230 (101a), S45,
Flickr https://www.flickr.com/photos/136329431@N06/albums
YouTube https://www.youtube.com/channel/UCrTH0tHy9OYTVDzWIvXEMlw/videos?shelf_id=0&view=0&sort=dd


Re: Experimental Photo Production SX110/120/130, A470
« Reply #18 on: 17 / November / 2017, 15:43:58 »
The frame rate is constant (25 fps) throughout all production phases. But it appears that AVS4you's Video Editor which I use for the final montage recodes all the frames, and even with the same codec, introduces stuttering.

I also tried ffmpeg to simply concatenate & copy individual sequences and they came out OK, so re-rendering is the culprit. Unfortunately this means I will have to develop the program I am using to generate the individual sequences to the point where the AVS video editor is superfluous and that might take some time.

Best

 

Related Topics