P52 Week 32: Long Exposure (August 6 - August 12)

[Darmok N Jalad]

I actually woke up in the middle of the night and did the time math without even thinking about it. And yes, 7500 images would be obscene to process. (For anyone else wondering, it's the inverse of your shutter speed times the number of seconds you want your final image to be equivalent to. So for my example of 1/2500 I'd need 2500 images to capture 1 second's worth of exposure and then if I wanted a three second exposure, I'd take three times 2500 for 7500 images total.)

And despite being a "party trick" as you call it, some people like me have little need for ND filters on a regular basis and they are expensive for a one-off every couple of years. It's an interesting proposition for the rare occasion that someone might feel the need to get smooth water or clouds.

I did buy a variable ND filter a few years ago but it never occurs to me to pack it when I go somewhere I might need it, so I'll have to try this some time.

Yeah, plus if you shoot lots of different stuff like I do, it gets in way. It’s a lot like when you have the wrong lens on the camera for a particular situation. Just yesterday this happened. I took my UWA for some lake shots, and naturally, a GBH was there and close enough that my long zoom would have been perfect.

 

[katbel]

Yep, more exposures in that 3 minute window. Maybe try 5 or 6 (or a few more) per minute, just as an experiment. Obviously, it's going to depend on the movement of the clouds or water (as example subjects).

Can't wait to try, but since yesterday the sky is grey and it's raining a little , that is very good for us but not for LE

 

[katbel]

Here is my morning experiment with clouds, that were moving a little more.
The first one is the original, to show the difference
The second is the work of stacking 36 photos taken in 10 min, following @r.harris1 suggestions

P.s. There have been days were the clouds didn't need a long exposure ...this is why I put the first shot to differentiate it from the final LE

 

[mollyc]

View attachment 2244242 View attachment 2244244 Here is my morning experiment with clouds, that were moving a little more.
The first one is the original, to show the difference
The second is the work of stacking 36 photos taken in 10 min, following @r.harris1 suggestions
P.s. There have been days were the clouds didn't need a long exposure ...this is why I put the first shot to differentiate it from the final LE

oh what a great comparison!!

 

[katbel]

oh what a great comparison!!

Thanks for the encouraging words but I just found Long Exposure clouds photos of what I should aim to

https://i.pinimg.com/originals/23/90/ca/2390ca7bd3f52b2d3b1bdf607b04fb74.jpg

Long-exposure-photograph-.jpg

 

[r.harris1]

View attachment 2244242 View attachment 2244244 Here is my morning experiment with clouds, that were moving a little more.
The first one is the original, to show the difference
The second is the work of stacking 36 photos taken in 10 min, following @r.harris1 suggestions

P.s. There have been days were the clouds didn't need a long exposure ...this is why I put the first shot to differentiate it from the final LE

Absolutely fantastic!

 

[katbel]

Absolutely fantastic!

Thanks but I barely see what is good in that photo. We have similar skies in November ...without the need of LE
What do you see?

 

[thirsty_monk]

Here is my take on a long exposure

ISO 100, 120.5 sec, f22

 

[katbel]

Here is my take on a long exposure

View attachment 2244392

ISO 100, 120.5 sec, f22

Very clever!

 

[katbel]

This morning had a better chance with the clouds, shorter time more photos

 

[r.harris1]

Thanks but I barely see what is good in that photo. We have similar skies in November ...without the need of LE
What do you see?

I see sky movement against the solid city, which would be what I look for from a long exposure. I get a sense of "time passing". The better image for me is the single short exposure one though as it has more drama (just a preference thing). Ideally, with an exposure as long as 10 minutes, it could be worth setting up an intervalometer to rattle off a shot more frequently. Maybe every 5 seconds. That would be a lot of images to process, obviously, but the goal is to have as small a gap between photos as makes sense within the total time span. It's trying to imitate the sensor being exposed for 10 minutes (or whatever).

To be clear, this isn't a substitute for ND filters and does fall apart once you try and do longer exposures, simply because of the number of images to process. It is, in fact, a "party trick", though useful in some circumstances.

 

[katbel]

Two different LE experiments with a lamp: 8sec f29 the first one
The second 8 sec f8

 

[katbel]

One more out of focus LE: still 8sec f8

 

[r.harris1]

One other thing on long exposures vs frame averaging. An actual long exposure is additive on your sensor. It keeps collecting photons as long as the shutter is open. It’s easier to capture drama in a real long exposure, for example. With frame averaging, if two or three of the images have a great piece of sun peeking through the clouds and 25 don’t, it will get averaged out. So there are uses for both techniques.

Another party trick that you can do with either technique is to average people or cars out of a scene (if it’s a busy area where people / cars are moving). I don’t have an example of my own handy, but this is a great early photograph by Louis Daguerre, one of the earliest practitioners of photography. It only has two people in it, it took 10 minutes to expose and all of the others were “averaged out”. https://mymodernmet.com/daguerre-oldest-paris-photo/

The two people were still long enough to record on the recording medium.

 

[mollyc]

Okay, guys, I am back from a week at the beach. I purposely planned out long exposures for this week way back in December, after we had booked my trip. But stupid me completely forgot to pack my tripod. I had SO MUCH camera gear because my daughter wanted me to do her senior pictures at the beach, and I managed to get those and some other images for myself, but since I didn't have a tripod I missed out on a lot of stuff this trip.

So, @r.harris1's party trick as he calls it, really came in clutch for me. I tried two different times and the second one I rather liked. The first one was a good experiment to see how it worked conceptually, but the resulting photo was pretty boring.

This one came out pretty well. Had I used a tripod I would have set up my composition differently and angled down to capture all of the little fences and probably moved the horizon. As it was, I just set my camera on a railing and used the intervalometer. I did 30 exposures at 1/30s per frame every 10 second, then merged them all together in the directions from the original link from rharris.

To rharris' point above, over the course of the four minutes or however long it took for these, there was actually a LOT of movement of people, across the sand and also playing in the water. I accidentally ran the wrong command when I was doing a merge for this and really I must have captured about 100 distinct people locations over the time period, but with the median command they all disappeared except for the people who stayed still under their umbrellas. My other image was kind of interesting; it looked more like a time lapse than a long exposure. I deleted it, but maybe I'll rerun it and post it later.

I also did some ICM but they aren't edited yet so I'll post those in the next day or two.

 

[cogi0490]

Since I have currently not that much time, I went this week with the most simple approach.

But thank you @r.harris1 for the nice introduction of the, to me unknown, technique - definetely have to try this one in the future!

 

[_timo_redux_]

Attempting to "drag my shutter" out a tram window yesterday.

 

[mollyc]

six second ICM image.

 

[katbel]

Last night I played with the city lights LE and ICM
The lights brought the buildings facade in front in this one

 

[katbel]

This one was taken almost at the same time, 30 sec later but with f6 instead of f8 10 sec

 

[mollyc]

View attachment 2245382 Last night I played with the city lights LE and ICM
The lights brought the buildings facade in front in this one

View attachment 2245385
This one was taken almost at the same time, 30 sec later but with f6 instead of f8 10 sec

I really love the second one

 

[Janichsan]

Jumping on the bandwagon which slightly derailed this topic I gave the frame averaging a try.

Nicely smoothed water seems to be tricky.

Movement clearly looks more like a series of images than smooth blurs. Still an interesting effect, though.

The technique seems great to remove unwanted moving persons and objects, though.

Before (one of five pictures taken 10 seconds apart):

After:

 

[mollyc]

Jumping on the bandwagon which slightly derailed this topic I gave the frame averaging a try.

Nicely smoothed water seems to be tricky.

Movement clearly looks more like a series of images than smooth blurs. Still an interesting effect, though.

The technique seems great to remove unwanted moving persons and objects, though.

Before (one of five pictures taken 10 seconds apart):

After:

For the images that show the movement of the people....how many frames did you take and how far apart were they? It seems like they might have been too close together. Although the end result is quite interesting. A traditional long exposure may have kept some of them in there as well if they were moving slowly.

 

[Janichsan]

For the images that show the movement of the people....how many frames did you take and how far apart were they? It seems like they might have been too close together. Although the end result is quite interesting. A traditional long exposure may have kept some of them in there as well if they were moving slowly.

The first one at the crossing were 14 shots in burst mode, about 0.1 seconds apart. The second one has 11 frames, 10 seconds apart.

 

[mollyc]

The first one at the crossing were 14 shots in burst mode, about 0.1 seconds apart. The second one has 11 frames, 10 seconds apart.

Yes, so I think the short time period between images for the first one explains why you got that time lapse look. There was not enough movement of the extra people to end up with blank spaces for averaging. When you use a longer time frame, more of the background becomes visible over time and the people get erased out because it's evident they are moving throughout.