Time interval findings (edited analysis)

[Ben Egbert]

Thanks Jim, I will read that link. I am primarily print guy and I am always looking for a great image to print. I seldom watch videos or TV for that matter. I am getting interested in time lapse so things could change. But it's basically for showing here I don't have any other audience.

 

[Ben Egbert]

Ok, you guys are shaming me into shooting raw time intervals. But I need to figure out how to make a decent time lapse with this technique. In camera TL on the R5 blows away any result I get with the RDSR using intervals. I have two issues. one, just getting a TL finished at the right speed. Second, when I so get one done, it still looks jerkier than the in camera version.

I have used Photoshop to create tL from intervals and besides being clunky I still get lots of jerkiness. With Movavi, I often get good (but still jerky) TL from intervals, but many times I get results that are all over the map.

For example I used the exact same files last night to do 388 images at 6 second intervals from RAW that were converted to JPG. Some times either .1 or .2 worked, other times it was 4 to 6 times longer, or very fast. I was careful to start from scratch each time so I was not getting an adding effect. I will do this test over until I figure out what I am doing wrong. I now know that .1 or .2 to works, so it's something I am doing during selection.

I also think I need to stay at 3 or lower second intervals.

 

[JimFox]

Ben, whatever works best for you is what you should do. Just don't overthink it. Keep it simple.

I experimented with the in camera timelapse at Joshua Tree as I finished up, shooting 20 mins which resulted in 5 sec clips as I had mentioned. And on both my D810 and D850 it was able to very nicely capture the stars at night. It's something I will test more.

To do stills though at night is not hard, because you are only taking 1 photo everyone 20 to 30 secs. Doing stills also allows flexibility in that if there is unwanted light pollution like in my case I had cars doing u-turns twice while I was shooting and their headlights totally washed out the Joshua Trees. so for 6 frames I had to mask back in the ground layer. If it was an in camera timelapse I couldn't do that. But for you, if you are shooting out somewhere that's not an issue, in camera timelapse on the R5 could be the way to go.

For my beach timelapses, I did in camera timelapse because it was during the day and I wanted to catch the waves and also any people walking on the beach in front of me.

So to me, there is a case to be made for both methods. They are both tools in our tool pouch that we can choose which works best.

Also as a note, I have found with doing stills and shooting Raw (as normal) and then converting them to jpg can be an issue if the sky is totally blue. I am getting horrible banding. I have found Movavi does read tif files, so I am going to try that next. But for my star photos, I converted them all to jpg and pulled them into Movavi. The stars break up the sky enough that there is no banding.

 

[Ben Egbert]

Jim, after my last post, I tried 3 more TL/s the first was good, the second was not, and the third was good because I think I know what's causing them.

When you start a TL in Movavi, the default duration is 4 seconds. If I press 0 to overwrite the 4, a new value of 00:00.010 appears. I overwrite the next zero with a 1, but in the meantime Movavi is setting the value of 00:00.01 and you get a very shrunken TL. Even if you have it set to 00:00.010 if the tl has not regrown to the appropriate length, you must wait to start export, otherwise you will get the value of 00:00.010 processed.

I am going to see if I can reset the 00:00.100 as the defualt.

 

[JimFox]

I haven't seen any way to change the default settings, if you find it let me know.

You are changing the Duration time in the wrong order. Change the .000 to .100 first. Then change the 00.04.100 to 00.00.100 next. Otherwise you get that wonky .010

I have noticed sometimes the duration doesn't get refreshed after doing that. If it doesn't, move the cursor to in front of the 1 in 00.100 and then click the down arrow next to it. It will change it to .099 or .098 but that's as close as to .100 we will never know the difference.

 

[Ben Egbert]

Right Jim, there is no way to change the default and I also came up with the idea to change the millisecond value first. The key it to wait until the time line looks about right.

 

[Ken Rennie]

Ben I know little about film making but my Maths skills are still pretty good and it looks to me that you are sometimes swopping capture time and viewing times.

Please correct any mistakes you find here, but I wanted to share some stuff I have learned doing time intervals with my 5DSR.

• The number of shots divided by the desired Feet per second (fps) will give you the duration of the video. So 600 shots/30 fps equal 120 seconds or 2 minutes play time. The interval that you shot the frames does not enter into the equation, but will have an effect on smoothness.

600 shots at 30 frames per second gives you 20 seconds of play time

• 
  
• In Movavi, you can adjust the duration, and this is where you set the time. Simply take the number of images and divide by 30 or 24 or 15 (whatever you choose) and use that value for the duration.
• 
  
• Shooting at 3 second intervals for 600 shots using 30fps will yield 1 minute, if you shoot at 6 second intervals, you will get 2 minutes for the same settings.

600 shots at 30fps still lasts for 20 seconds. The interval between the shots changes the length of time that you capture eg 3 second intervals for 600 shots =1800s or 30 minutes but this will be shown in 20 seconds. 6 second intervals for 600 shots =3600s or 60 minutes. My maths graduate daughter would point out that 600 shots = 599 intervals but my figures are close enough

• 
  
• I have not determined which interval is best 3 or 6 seconds. It depends on the speed of changes in your subject. 6 second intervals will take a full hour to get 600 images, where 3 sec intervals will only take half an hour.
• 
  
• My 5DSR can be set to manual mode and using auto iso so that exposure adjustments are made during the shoot. Very useful for sunset/sunrise. This keeps shutter and aperture constant while letting iso rise or fall as required. A bit of forethought is required to start with a bright/dark exposure to make sure you have the range for the ending darker/lighter exposure.

I hope that this helps. Ken

 

[Ben Egbert]

Thanks for pointing this out, I did a spreadsheet for the math and did not check the logic close enough.

I went back to the spreadsheet and found my mistake and fixed it, 20 seconds is correct of course.