Closed GoogleCodeExporter closed 9 years ago
Could you please post all your settings (both EAEB and Intervalometer)? Thanks!
Original comment by eduardo....@gmail.com
on 13 Dec 2011 at 7:07
Sure thing. :) Thanks for the timely response.
For my basic camera settings I have manual focus and burst mode enabled. AEB
and Flash exp comp. are off. The camera is in M mode with aperture 8 (I figure
this wasn't too important) and Tv is Bulb.
In CHDK...Safety shift is off, Auto ISO is off, and AEB is off. The
intervalometer script works fine. The settings are:
2s delay: No
Time(s): 0
EAEB: Yes
Shots (5-15)
For the EAEB script I have:
2s delay: ON
Frames: 5
Step(EV) +1
I keep trying multiple combinations for Bulb Min and Bulb Max. So far when bulb
max is 30s or faster, the script works great...as long as bulb min is 5 steps
quicker...2sec or faster. The Tv during exposures usually start at bulb min and
then get quicker.
My problem comes in when I attempt to make bulb max 1min or more. I'be been
trying many combinations where I calculate bulb max to be 5 or more stops
quicker than bulb min. I managed to get it working once at 1min with bulb min
2s. My results were six exposures of:
(bulb max 1min, bulb min 2s)
1) 1min
2) 30s
3) 8s
4) 2
5) 1/2s
6) 1/8s
I do not know why there were six exposures or why bulb max was 1/8s. How is
this calculated by the camera? Originally I tried it at 1min and 4s, but it
stopped after one photo with Tv 1min.
I tried Bulb Max at 2min and Bulb min at 8sec and got three images:
1) 2min (the camera then displayed the image and waited 1min before continuing.
The blue script light was still blinking)
2) 2s (not sure why)
3) The third image was stopped at 12 min and it appeared that it would have
gone for longer.
After I tried to replicate the 1min-2s EAEB, but it stopped after two photos:
1) 1min
2) 2s
I tried multiple times, but again it just took one image and the script stopped
running:
1) 59s
I then tested again with Bulb Max at 30s, but it only took one photo! After
that I attached my CF card to my reader and deleted the "settings" and
"presets" files so that the scripts would reset.
My main questions are:
1) What should the bulb min and max settings be if I want to take an EAEB with
maximum Tv of 1min? At 3,5,7 exposures?
2) How does the camera determine exposures between the bulb min and max values?
I assumed that the camera simply takes exposures from min to max at values in
difference of 1 step.
Ex: 1min-4s...Exposure Tv:
1) 1min
2) 30s
3) 15s
4) 8s
5) 4s
With this I would assume to take a photo where the proper exposure is 15s.
Also, it seems that at times the camera shoots +/- and at times solely "+"
values.
3) What other camera settings should I use for EAEB with exposures over 1min?
THANK YOU SO MUCH. I hope this helps. :) I am continuing other combinations
to understand EAEB mode more.
Original comment by davidden...@gmail.com
on 14 Dec 2011 at 7:55
PS...at times where I take an exposure of 4min, there is often a 2min delay
between shots. For an exposure of 8min, there is often a 4min delay.
Sometimes there is no delay. Is this normal or expected? Thanks. :)
Original comment by davidden...@gmail.com
on 14 Dec 2011 at 8:04
I think your understanding of how EAEB and the intervalometer are expected to
work is flawless, all your suppositions are spot-on; so this is must be a bug,
and I will come later when it has benn fixed. Thanks for the detailed report.
Original comment by eduardo....@gmail.com
on 14 Dec 2011 at 10:38
Thanks a ton Eduardo. :)
I tried all other builds with the same result. As of now I am increasing the
iso and the number of intervals to work with a maximum shutterspeed of 30s. It
works great, with no delay between shots and the no need to touch the camera
during exposures.
I noticed that whether I change the step to +1/2,1,or 2 that there was no 15s
exposure.
Ex:
5 exposures, 1/2step, Bulb Max 30s, Bulb min 1s
1) 30s
2) 8s
3) 4s
4) 2s
5) 1s
If I change the step inteval to +1 instead of 1/2, the same exposures are
calculated. I assume that the step values are over ridden by the Bulb Min and
Max Values. :)
Also, if I change the Bulb Min Tv to 2s, I end up with 4 exposures instead of
5. I can work around the missing 15s exposure, but thought I'd let you know. At
times when the script stops the Tv for the setting in Manual mode changes from
Bulb to the last exposure time. Maybe it gets stuck in a loop?
Originally I was planning on taking 25 (5 sets of 5 +2ev bracketed exposures)
at iso 200. This would leave me with 5HDR images that I would stack in
Photoshop to reduce noise. On a full moon I set Bulb Min to 8min and Bulb Max
to 2s. From what I understand (and perceive) I would be left with an image with
resulting ISO similar to 12.5.
Now I intend to raise my iso to 1600 and take 75 (15 sets of 5 +1ev bracketed
exposures) exposures. I now would have 15HDR images to stack, and
mathematically I should greatly reduce noise, but I am somewhat weary of the
added computations required by photoshop. It seams as though there may be more
room for error. Also, my bulb Min would be 30s, while my bulb max would be 2s
and I feel I may not get enough information and too many dark images.
I like pursuing this theory of mine for night photography. You have made it
possible for me, as I have been dreaming of it for many years. I also have
downloaded the SRC files and tried to make sense of them, script wise. Is there
a specific file I should look at? Maybe something I can read to help me
understand better?
Thank you so much.
Original comment by davidden...@gmail.com
on 17 Dec 2011 at 12:14
Just some info on this:
* Exposures longer that 30s are done 'by hand' (400plus opens the shutter,
waits, and closes it), but exposures of 30s or less are done by the camera
(400plus tells the camera the exposure time, and then just fires the shutter);
this explains why the different behavior.
* In BULB mode, 400plus only takes into account "Bulb min" and "Bulb max"
parameters; EV separation is fixed at 1EV, and the number of shots is
determined by the "distance" from "Bulb min" to "Bulb max".
The "interesting" part of the EAEB script is the "sub_ext_aeb" routine (line
201 at scripts.c), but the shutter is controlled by "shutter_release_bulb"
(line 405 at utils.c). Notice how the code to open the shutter is the same as
the code to close it: "press_button(0xB6)"; so, if we somehow "lose" one of
this codes (if the camera is busy, for example), the script will get out of
sync and close the shutter when it should open it. Notice also how the code to
detect whether the camera is busy changed in version 20111111...
Original comment by eduardo....@gmail.com
on 17 Dec 2011 at 9:36
Could you please test _next_ (revision 1107 or greater) nightly build? It is
working for me now, but I could only do a couple of tests. If this solves the
issue, I will release 20111111-2 shortly, as this is clearly a regression. Many
thanks.
Original comment by eduardo....@gmail.com
on 17 Dec 2011 at 11:07
Dear Eduardo...:)
I tested Revision 1107 thoroughly and it works gloriously! Tested Tv up to 32
min and 9 exposures!!! Thank you so incredibly much for your attention, help,
and solution.
I've tried nearly every tether app (Capture One, Lightroom, OnOne DSLR remote)
and even DSLR.Bot for the iphone to achieve long exposure bracketing. This is
hands down the best and most convenient as I don't have to carry around any
extra equipment.
The tether apps did not allow Bulb exposures easily.
DSLR.bot works great when I made my own IR remote, but the bracketing function
was horrible. I did notice that the Rebel XTi had a .3sec delay required
between exposures, in bulb mode. Anything quicker would lead to problems.
(Might be helpful for future releases.)
I can not emphasize enough how incredibly helpful this is towards my
photography. I thank you so very much for your help.
Badassery,
Richard Swagger
Original comment by RickGran...@gmail.com
on 20 Dec 2011 at 2:39
I have just uploaded 20111111-2; this is the same as 20111111-1, plus the fix
that you tried on revision r1107 (oh, and the fix just adds a 1/10s delay after
every BULB shot in EAEB, BTW). Thanks for your detailed report, your thorough
tests, and your encouraging words... always happy to help!
Original comment by eduardo....@gmail.com
on 20 Dec 2011 at 11:30
Please, forget about 20111111-2, I left some debugging code there; use
20111111-3 instead. Sorry for the inconvenience.
Original comment by eduardo....@gmail.com
on 21 Dec 2011 at 7:04
Original issue reported on code.google.com by
RickGran...@gmail.com
on 13 Dec 2011 at 3:27