Hi,

I recently did a few timelapse shots with the ProTimer 3 (firmware 28) and my EOS R5. For the first approx. 100 shots of the 700+ Timelapse it dropped one or more frames very frequently, then became very stable after about 100 frames. Here are some the details and please let me know if you need more:

ProTimer 3, Curren firmware
Timelapse Mode, 10 second interval

Holy Grail sunset directly into the sun, temperature 0C to -3 or -4C or so.

The frame dropping occurred in the first part while the sun was still in the frame--as such during this time I did not have QDDB connected. The shutter speed was 1/5000s and the aperture was f/5.6. I was using a Samyang manual 14mm. I generally do not have problems shooting an 8 second frame with a 10 second interval, so I cannot understand why I have frames dropped with a 1/5000th of a second shutter speed and a 10 second interval.

The R5's back screen was folded closed--not in use.

This has me a bit stumped. Maybe it is a weird sleep issue with the R5? Maybe it is an issue with the Samyang? But interestingly enough, the 20 frames after it stabilized at 10s intervals, the shutter was still 1/5000s of second.

Also, about the time it stabilized, I connected the USB-C cable to my android device for QDDB and probably about 20 frames later, the shutter speed started dropping (I manually dropped the shutter speed to get the exposure correct after the sun went behind Mt. Fuji) and from there the auto-holy grail took over and the interval was consistently stable at 10s until the end of the shoot.

Now that I am working through it all--I perhaps have a theory...I wonder if it is possible that with the long time time between the fast shutter and the 10 second interval, the camera went to sleep and struggled to wake up, causing the missed frames--but then once the Android was paired via USB-C and connected to QDDB, that somehow prevented the camera from sleeping? Any thoughts, have you seen this before with this combination? I could not find any writes ups about it in the forum.

I checked two other Timelapse shots I have done recently where the sun was in the frame and I started QDDB after the sun had left the frame.
---In the first one earlier this month, I had a stable interval (10s) the whole time, but the shutter speed as 1/1600th (canon EF lens+adaptor). I am not sure if the phone was attached and on from the beginning of the shot or not with this one.
---The second was the day before the first shot described above (except I was using a canon EF lens+adaptor), same location, and same approach. The phone was not connected until later and as before the 2-3 frames (and as many as 5!) were dropped between actual photos until the phone was attached, then it was stable for the remainder of the shoot.

If it would help, I can provide screen shots of the LRTimelapse showing the shutter and interval, etc.

If this theory is right, can I have QDDB connected and just not start the LRTImelapse module until I after the sun is out of the frame? If I don't tell it do anything during that time? It should not have an affect at all, right?

However--if this theory is right, what do I do for non-holy grail shots? Any ideas? Has anyone else had this issue with the R5? How do I ensure that the interval remains stable without connecting a phone?

Thanks--and I look forward to any advice, suggestions, or known solutions.

Russ

You had autofocus turned off all the time? Autofocus is the most frequent cause of dropped frames.
If that's not the case, it could be the long dark time putting the camera to sleep.
In that case you could experiment with longer standby times in the camera or play with thehttps://lrtimelapse.com/lrtpt/manual/zz "DSLM Standby / Wakeup Time" in the LRT PT's Settings, which is explained in the manual: https://lrtimelapse.com/lrtpt/manual/
In any case try to figure out what's happening before going out next time. Just do tests at home to not mess up important shots.

Another thing to try could be to slightly increase the release time in the PT's Settings, maybe your camera needs a longer release time to trigger reliably. Also explained in the manual, or in the faq: https://lrtimelapse.com/lrtpt/faq/

Gunther, thank you so much for the quick reply. I really appreciate the additional details on how I can approach the problem from the LRT PT side. Today I will run some tests, starting with some the R5s "eco mode" and "power saving" features as well as the LRT PT sides suggestions you have. Ideally I can find a fix while keeping the power saving features. For what it is worth, I can report my findings back here in case other R5 users find similar difficulties. I find this forum the best repository for info and advice on Timelapse photography!

This post was last modified: 2022-02-20, 13:08 by Gunther.

Thanks, keep us posted!

Gunther, thank you for your kind help. This is what I found after a LOT of testing today:

Bottom line, changing the LRT PT3 Release Setting from 200ms to 270 seemed to have the best results with the least amount of increase. I found that having the drive in slow-speed continuous caused significant problems that were resolved by setting the drive mode to single shot. Changing the "Eco" and "Power Savings--auto power off" settings to "off" did not have an affect on the number of frames dropped.

Some of the data:
I mostly tested under the following conditions: F2.8 1/5000s iso 40K, with a 10 second interval for 30 frames, Single Shot Drive Mode, and first curtain electronic shutter, back screen closed, Canon EOS R5 with current firmware as of 22FEB2022.

Release Setting_____ Frames_____ Dropped Frames
200ms______________14/30_______16/30
220ms______________8/30_______22/30
230ms______________12/30_______18/30
240ms______________22/30_______8/30
250ms______________30/30_______0/30
250ms______________26/30_______4/30
260ms______________27/30_______3/30
270ms______________30/30_______0/30
270ms______________30/30_______0/30
270ms______________30/30_______0/30
270ms______________8/30_______22/30 * Test of slow speed continuous drive mode at 270ms vs the original 200ms
270ms______________30/30_______0/30 * back at single shot drive mode
270ms______________30/30_______0/30 * test using electronic shutter
270ms______________30/30_______0/30 * Test using full mechanical shutter
270ms______________99/100_______1/100 *(back to 1st Curtain Electronic Shutter--tested 100 shots. one drop????
270ms______________250/250_______0/250 *Test again against longer set, this time no drops.

I also tested at 300ms, 350ms, 400ms, 450ms but was not seeing consistent results--that is when I noticed the slow-speed continuous drive mode issue and changed the the single shot mode. I noticed this issue because on some of these early tests, at above 300ms, I would occasionally cause two frames to be taken for one trigger. That was odd, so I changed to single and it became consistent. Then I worked to find the shortest release setting that worked consistently. All in all I took 530 frames at 270ms release setting and single shot drive mode and it only dropped 1 frame. Do dropped frame just occasionally happen? Is this cause for concern or more testing looking at say 280ms?

Based on the above, I concluded the best Release Settings would be 270 using single shot drive mode when using LRT PT3.

Next I wanted to test the minimum dark time when not using a screen to monitor--all tests at 270ms Release setting and a 30-frame test.

Shutter Speed_____Interval____dark time_____Frames____Dropped Frames
8s__________________10s___________2s____________30/30_____0/30
5s__________________6s____________1s____________30/30_____0/30
5s__________________5.5s________0.5s____________19/30_____11/30
5s__________________5.7s________0.7s____________22/30_____8/30
5s__________________5.8s________0.8s____________27/30_____3/30
5s__________________5.9s________0.9s____________30/30_____0/30

Conclusion based on this small data set with the conditions described above (not using back screen for monitoring: Minimum dark time 0.9s when not using the screen for monitoring or making adjustments.


I hope this is helpful and welcome any thoughts on the approach / conclusions.

Russ

Thanks for sharing your results.

For Timelapse shooting you always should have set your camera to single shot. That's one of the prerequisites. Sorry didn't mention that because I thought it was self evident.

And I wonder why you would want to shoot timelapse wit 1/5000 of a second. Normally you would do anything to work with long shutter times, mostly using an ND filter at least during the day.
Keeping the dark times short and exposure times long is one of the success factors for good timelapses.

Thank you for the reply! There is a lot to remember for the Timelapse settings, and I had forgotten that it needs to be in single shot mode--but I also never had a problem before now when using my pervious 5Dmk4. Lesson well learned.

The reason the shutter speed was so high for the start of this shoot was that I was shooting the “Diamond Fuji”-- that is the sun setting right on top of Mt Fuji—so I had to use a very fast shutter speed in order to keep preserve some details on Mt. Fuji and the lake in front of it...before the sun went behind Fuji…also, I had the aperture wide open as I didn't want to ramp the aperture after the sun set for the golden and blue hour portions of the shoot. I suppose could have used an 8-stop N/D filter but I didn't think that would work well because I would need to remove it after the sun went down and I thought that the sudden change in exposure would be too much. Finally, I was using for that shot a very manual Samyang 14mm that cannot accept filters and the aperture can only be adjusted with the ring on the lens. (Though I did have a 24-70mm available...).

I just finished processing the TL tonight--even with the skipped frames, it turned out okay. That said, it is clear I still have so much to learn. I am sure I could have done it better. I always welcome your sage advice. I will post a link here once I get it uploaded.

Russ

This post was last modified: 2022-02-20, 20:38 by Gunther.

Ok, for some holy grail situations where you would shoot until the stars, it might be necessary to start with short exposures.
Just keep practicing! Enjoy! :-)

come on Russ, lets see the timelapse :-)

Cymro--Thank you. I figured out just now how to upload to YouTube--but I didn't add any music as copyright issue are more than I want to deal with now so of course the video is a bit boring. I will eventually put it up on Instagram with music. You can see it here: russm.photos (I have several Timelapse reels there) but for now you can see the two Fuji Timelapse videos I made (that caused this whole thread to be created (and you will notice the changes in speed caused by the dropped frames--but it works out in my opinion as the speed change is kinda of cool. Here they are:

Diamond Fuji: https://youtu.be/w2LWhd8alzo
Cloudy Fuji Sunset: https://youtu.be/DjzLj_MCd78

Welcome any thoughts!