Messages

My conclusion over the years:
Use ML-ETTR to make the best possible picture, Postprocessing (deflicker, ramping, adjusting exposure to taste) is done easiest and best with dedicated software.

I can only agree...

Wonder what will be next: 100D?

Then this will probably be the last two-digit APS-C DSLR from Canon ... 

Do you have a link to said manual?

Only the German version 'EOS_M50_Help_Guide_DE.pdf'
https://www.canon.de/support/consumer_products/products/cameras/digital_slr/eos-m50.html?type=manuals&manualid=tcm:83-1667469
Page 88 or search about 12 bit.

The M5 also has 12 bit  in continuous mode...

[M50 bit depth]

M50 bit depth

Single - 14-bit
Continuous Low - 12-bit?
Continuous High - 12-bit?

The manual says that the data for continuous recordings are converted into 12 bits...

Over on CHDK I use this: https://chdk.fandom.com/wiki/Execute_Lua_scripts_with_hostlua_%26_notepad%2B%2B

PS Forgot to say, no good for ML as it's only integer math Lua

But the tool is really fantastic. I can test it synonymous script without a camera if I have no camera available (on trips). If a camera is connected, then I can upload the scripts by CHDKPTP. For someone who makes many mistakes while programming, the time savings are enormous

Why? Plot the same graphs as (steps, 1 / distance) on linear scale; the result will be a lot closer to a straight line.

Right...
So, if you know upper and lower distance you could implement a function which gives you a focus distance for each step...

@a1ex
I hope I do not disturb your break...

Is it possible (after AF) to read the absolute number of focus steps?

Code Select Expand

call_event_proc("EFLensCom.MoveFocus",steps,speed)
On the M3 I can also change the speed. Do you know which influence the speed has or where are the limits?

Quick test with the 50/1.8 STM and 5D2.

I did some investigation with the M3 as well
https://chdk.setepontos.com/index.php?topic=13756.20

- the focus distance encoder has a small number of unique positions

Max 31 positions on my efs 55-250 mm  is stm

I find it interesting that all my lenses have the same maximum lower focus distance

Focus Distance Lower max    : 8191cm
Focus Distance Upper max    : 65535 cm => inf

- aperture does not affect the results (so, "Lower" and "Upper"

I notice this as well..

- the mean value of the two distances might be a better estimation

I think these values are just any table values. They are not very accurate either. My G1x provides much more accurate distance values

I'm playing with my EOS M3 and trying to set the focus
https://chdk.setepontos.com/index.php?topic=13756.msg140015;topicseen#new

In the M3 there is information about Subject Distance Near / Far. I suspect that this is the same information that you find in the EXIF data under 'Focus Distance'. Maybe this information will help you further.

However, this information is probably not very accurate.
If I use small steps, the EXIF Information are not changing

Code Select Expand

call_event_proc("EFLensCom.MoveFocus",steps10)

It looks like that the DoF Information in the EXIF simpy calulatete by

Code Select Expand

Subject distance = (Focus Distance Upper  - Focus Distance Lower) / 2


EXIF Info from 6D

Camera Model Name               : Canon EOS 6D
Lens Model                      : EF24mm f/2.8 IS USM
Aperture Value                  : 2.8
Circle Of Confusion             : 0.030 mm
Hyperfocal Distance             : 6.76 m
Focus Distance Lower            : 2.1 m
Focus Distance Upper            : 3.67 m
Depth Of Field                  : 2.97 m (2.03 - 5.00 m)

Another idea for a power supply. A powerbank with quick charge. With 2 diodes and 2 resistors you have to generate directly 9V / 2A. With that, the cameras are getting along well.
https://blog.oxplot.com/quickcharge/

Yes, but these are only useful if you operate in LiveView. Outside LiveView, i.e. in shooting mode with display off, the power draw is lower than in playback mode with display on (both configurations being available in vanilla firmware).

Yes of course, I forgot this advantage of a DSLR ... 

Tip: with an external power adapter,

https://www.magiclantern.fm/forum/index.php?topic=21660

I am trying to do a long-term time lapse of about 90 days. 

I am also trying a  long-term time lapse ...
Currently I do that with my Powershoot and CHDK. There I switch between the recordings in the playback mode which saves a lot of energy.
https://chdk.setepontos.com/index.php?topic=9969.540
Would that be possible with ML?
I cannot find the post anymore but with the EOS M3 / M10 there was a function, where you could set the camera for a certain time in standby. Is there something like this with ML?

Instead I would recommend some single board computer (Arduino for example) with control to power input in your setup. It is able to turn on and off power as wanted (requires almost no programming skills) and trigger events (=actuations) as needed.

But especially for long periods you have to pay attention to the Arduino too (millis () overflow after 49 days).

No idea but there is no reason you shouldn't stick to it. And I mean it as a person who usually hates all the "auto" stuff. The implementation of the auto algorithm is really smart. It combines ~binary search to find the coarse range with fine sweeping afterwards.

The auto mode takes only a very long time

Try to rather not use the +/-100 mode as it sweeps with the step of 5, meaning that the final accuracy is only +/-5. The +/-20 method can also deliver a worse result than the auto if the values of +/-20 are still not beyond the borders of focus confirmation.

I know that

BTW, from my experience: I don't find the Dot Tune method the ultimate solution for focus problems. I'm used to still apply additional corrections based on photographing of real subjects and not focus charts. It seems to make a difference.

The problem is not the method rather that the focus through the viewfinder varies greatly depending on the target.

The attached plot shows 10 shots of the Liveview focused and 10 shots of the viewfinder (Same chart). It is easy to see that the sharpness of the live view almost does not vary at all. The shots on the viewfinder vary very strong. The plot is the worst case with a Siemensstar. Shown is the relative degree of focus of an image determined as described here:
https://de.mathworks.com/matlabcentral/fileexchange/27314-focus-measure


My idea was: Comparison (with a script on PC) of an image that was focused on the live view with 21 shots that were focused in the range + -10 on the viewfinder. But here it was the same, after each test, the focus was different. You have to live with the fact that the focus on the viewfinder varies.

To see the impact of the of different charts, I attached two score charts. The upper diagram comes from the Siemensstar, the lower one comes from the target proposed by the DotTune method.

I'd still like the logs, if possible.

Can I upload the log directly?
https://drive.google.com/open?id=1rqfUPcqdmjB7iuPuFNBYeOXLKz8rAvPF

For example, if you hold the camera still, focused on the chart, and you change the AFMA value to some extreme values (say +/- 100 or whatever the menu allows), do you still get focus confirmation? After some threshold, you should no longer get the confirmation.

That's works very well. Max range in the Canon Menu is 20.

Maybe the focus chart is "too sharp" and you still get AF "confirmation" even if the image is somewhat out of focus?

As can be seen from Reply # 267, I use different charts. The different charts and lighting have a big impact on the variance of the viewfinder focus. Here are my previous experiences:
A Siemensstar in artificial light is the worst case. Here the focus has the most varianz. It gets a bit better in sunlight. Horizontal and vertical lines have less variance. Horizontal lines work best, presumably because only one component of the cross sensor is used.

Here you can see how the setup is at the Canon service for the adjustment:
https://www.dslr-forum.de/showpost.php?p=15313563&postcount=40

I assume it might be a bug in DotTune, which I experienced on 70D.

It looks quite like this...

Workaround: change the DotTune to use ~"same settings for all lenses" instead of separate setting for each lens.

This works very well even in the 20 range . Thanks for the hint.
Is it correct that every time I start the camera, the scan range is set to Auto?

I just checked the 50mm

- if you change the value manually from ML menu, is the change visible in Canon menu?

Yes

- if you change it from Canon menu, is the change visible in ML menu?

Yes

Should I still use the logging? 20s is not much to start the dot_tune...



I only started with the tests because the focus on the viewfinder varies a lot from shot to shot depending on the target and the light ...

I have done some more tests with an EF 24mm f2.8 IS USM. Also, I have done this test with different targets and lighting situation. However, I could not test in daylight yet. No matter what range I set, it always comes out a very flat constant score graph. What does that mean?

With the 24mm I sometimes get the value -2 at -100 +100, which cannot be confirmed in a 40 or 20 range.

Is it actually possible for you to freeze the results at the end? Or make a snapshot on the end?
Especially if you scan in automatic mode it takes a long time. It would be nice if the results were longer in the end.

I tried to run dot_tune on my 6D with EF 50mm f1.8 STM. I used different targets in around 2m distance.   
I used 4 passes and AFMA mode 'This lens, prime/both'. I also used all Scan types.

I always get a really constant score graph and result is always 0. I never seen a score graph like in Reply #14. How can it be?
How important is the light source?

Edit: the EOS M50 appears to run EOS firmware (other recent models, i.e. M3, M5, M6, M10 and M100, are based on PowerShot firmware). Looking for a volunteer to try the LED blinking test on this camera, too

Does anyone know if Canon Basic is available on the M50 or EOS R(as with the M3)?

For me, the perfect solution to this problem was a printout of the finished housing layout and the purchase of 40 batteries at 2500 mAh each. Well and their connections.

As long as you do not have to get on a plane ....

I run the intervalometer with 6s and Auto ETTR. I got a jitter from around +- 0.2s.
Can it come from Auto ETTR?


I haven't changed anything on installation since last time...

Exposure matching is exposure matching.  Doesn't matter how different the images are, whether those images are captured with 1/3 stop or 1/2 stop exposure settings

Just for understanding.
This video is made with 1/3 stops end I can't correct the pumping sun in post because the core of the sun is blown out.

If I were to create these recordings with ETTR then you could probably correct it but then I would lose a lot of information especially in the dark areas.

I really like the CHDK part where I can change exposure in 1/96 stops...

The red "line" is where the 90th percentile falls in your image (in other words, 90% of pixels are darker than that and 10% are brighter). If this falls onto a flat area (such as the sky), that's a good choice for deflickering.

Ok, I understood that now...

I have now examined the entire scene. Once the complete picture and then only the sky.
Each mean, median and 90% percentile. Curves are normalizes by an offset.

My result:
If you use only the sky, then there are no differences between mean, median and 90% percentile. Also the 90% percentile for the full picture makes no different (of course the data comes from sky only).
If you use mean or median for the full picture you get some errors (Around picture 914). Not much (around 0.05EV) but enough to see it on a clear sky. I mean I would have scene where it was clearer to see. This video shows flicker from 0.04EV

.

I still think that if you have the opportunity to choose a sector (Like in LRTimelapse) that is the safest method.

Code Select Expand

im = read_raw(f.name);
Where can I find the implementation from 'read_raw' ?

I'm doing it in the following way. Converting in a linear TIF

Code Select Expand

dcraw -4 -D –T
And than I open the TIF.





 

Ran this deflicker script modified as follows:

How do you run the code?
Looks very similar to my 'MATLAB' Code but it's not 'MATLAB'.
Do you have a code like this for focus peaking?

Code Select Expand


# dcraw -v IMG_0951.CR2
black = 512

# value not critical
white = 16383

Have you tried to get white and black with EXIFTOOL from CR2? I have only found black values.

Result: only some minor flicker in the sky (when metering on the turbulent water). Couldn't locate the scene in the youtube clip to compare. Very interesting phenomenon, btw!

OK sorry for that.
This video has very much post-processing to get the turbulent water shooth. As soon as the water touches the feet of the tripod, it will vibrate. If you want I can give you the original video.

Just a short overview about post-processing:
- Deflicker with LocalExposure2012 in Lightroom
- images aligned (image registration).
- median on 3 images to eliminate birds and individual events.
- smooth by averaging several pictures (motion blur).

Videos without lens correction from P&S looks awful....   

Code Select Expand


prctile (g1(:), 90) %  instead of median(g1(:))


I have not really understood what that has an impact if I use prctile instead of median...

in this particular case, plain averaging (mean) gives very good results, too:

So you think in that case it make no different if I just use the sky or the full scene with mean?

Unfortunately, that was the only one Scene I have only in JPG...

Enclosed the last 80 files of a scene from this YouTube (1:40) video.

Furthermore, there is a text file with file number and EV values in the ZIP file.
Tell me if you need more files... upload takes a while...

https://drive.google.com/open?id=1mmd-aIvkpbSYKLeRcRyCECtPq_c0F2I3

On the plot you can see how the flicker increase when the water goes higher...






Update Second link
https://drive.google.com/open?id=1QA78WXg5guVBJluY8EfhO-aeW4I2VM2m

Can you share a CR2 from the youtube clip, or maybe a few consecutive flickering ones?

Yes of cource. I try to upload on my google drive tonight...

There you should use median, not average. Even if it's a mostly flat area, a single specular highlight (just an example) will bias the average, but will have no effect on the median.

Understood. I have not thought about that yet. Here I have always used the mean value.

I'm not talking about median on temporal axis; there you can ramp the exposure values as you wish, as long as you don't remove the high-frequency components (required for deflickering).

Just for understanding:
- The purple curve are the exposure values from RAW (selected area)
- The green curve is the calulatet background curve
(For that curve over time I use median and average. LRTimlapse use only average.)
- The red curve is the delta from green and purple curve. The deflicker values which comes into LocalExposure2012.




img upload