Low (and I mean LOW) light, low ISO

MatijaG

Well-known member
Messages
100
Reaction score
8
Location
SI
This is of course is an area where full manual control is the only way, and the rear display seems to show better pictures than I get when I actually download them to my computer., so some of the pictures that I thought would come out OK, did not.

Still, I managed to salvage a couple of pictures.




Note, all pictures were taken with multi-second exposure, lens set to infinity, aperture 8. Because I only had a very flimsy tripod, and no remote, I set the 7100 to take three exposures when triggered by the self-timer.

In almost all the cases I found that the first of each series of three was smeared. So that means I really need to invest in a remote, or a more sturdy tripod, or both.




That limelight really made this picture tricky. If I expose for it, everything else sinks into darkness.





If you look at the red trails closely, they seem to have a wobble. I'm not sure why that is. Did the cars bounce on their suspension? Or is it an effect of my flimsy tripod?




HIghway on-ramp



I was looking for a vantage point over the highway in order to get those brake-light trails. This one wasn't ideal (too far), but it was good enough to capture a decent picture once I cropped it tightly enough.



--
 

Attachments

  • 2551909.jpg
    2551909.jpg
    2.8 MB · Views: 0
  • 2551910.jpg
    2551910.jpg
    2 MB · Views: 0
Generally, better lit more urban areas yield better pictures. The dynamic range there isn't so wide that some stuff is inky black and others are blown out.
 
MatijaG wrote:

In almost all the cases I found that the first of each series of three was smeared. So that means I really need to invest in a remote, or a more sturdy tripod, or both.


That limelight really made this picture tricky. If I expose for it, everything else sinks into darkness.

If you look at the red trails closely, they seem to have a wobble. I'm not sure why that is. Did the cars bounce on their suspension? Or is it an effect of my flimsy tripod?


HIghway on-ramp

I was looking for a vantage point over the highway in order to get those brake-light trails. This one wasn't ideal (too far), but it was good enough to capture a decent picture once I cropped it tightly enough.
In the first one, I wouldn't worry too much about trying to expose outside of the floodlights. If anything, I think that a bit of post to darken the darker tones would make for a more dramatic image.

The second one looks like mirror slap to me, which would show up on the bright light sources but not the ambient background on long exposures. I think that because you were using the timer, you couldn't use the MUp setting (somebody will surely correct me if I'm wrong, but my camera is packed away at the moment). If you look at the tail lights, you can see that the amplitude of the wobble is greatest near you, then diminishes as the cars travel over time. Also, it's hard to tell because it's not that great, but the streetlights look like they could be more indistinct.

ML-L3 remote; Cheapest most effective accessory you can buy. Cheers.

--
 
My shortcut is to frame shots with the viewfinder and then switch to live view to take the shot. You need to cover the eyepiece with that little black thingy if you shoot on a tripod thru the viewfinder. Use mirror up, too. Try dialing in -2/3 EV to keep the highlights from being blown on white lights.

Good shooting, Joe
 
Surely these photos are not that low iso? They are iso 250. Low iso would be 100?



You would have been better off with iso 100. It would give you more dynamic range and less noise. Also the longer shutter speed would actually help. Shutter slap is not really a problem in really long exposures as the camera settles down



were these really 130/10 seconds (13 seconds). if so its not mirror slap its a wobble pod problem

--
http://www.flickr.com/photos/john_clinch/
 
Last edited:
MatijaG wrote:
If you look at the red trails closely, they seem to have a wobble. I'm not sure why that is. Did the cars bounce on their suspension? Or is it an effect of my flimsy tripod?
It cannot be from your tripod since all other points of light are sharp. If the camera were vibrating, then they would have an oblong shape, or be double valued. Look at the red reflectors near the triangular shaped sign. perfectly round.

And there is no way that your tripod would vibrate for 13 seconds with such a constant period. So it must be the cars. But the trails for several cars seem to have pretty much the same wobble period, and that is strange too. I do note that the amplitude of the wobble varies from one car to another.

Interesting.
 
Bob in Baltimore wrote:
MatijaG wrote:
If you look at the red trails closely, they seem to have a wobble. I'm not sure why that is. Did the cars bounce on their suspension? Or is it an effect of my flimsy tripod?
It cannot be from your tripod since all other points of light are sharp. If the camera were vibrating, then they would have an oblong shape, or be double valued. Look at the red reflectors near the triangular shaped sign. perfectly round.

And there is no way that your tripod would vibrate for 13 seconds with such a constant period. So it must be the cars. But the trails for several cars seem to have pretty much the same wobble period, and that is strange too. I do note that the amplitude of the wobble varies from one car to another.

Interesting.
 
The wobbly lights are strange, I agree with bob that its not a tripod problem as background lights are quite round in the shot. But for all the lights on all the roads to be wobbly, not what I'd expect.

I'd recommend trying for crossover light in these images. It varies from day to day due to cloud variations, and latitude to latitude - sometimes can give electric blue skies, other times far from it. You need to be in position between sunset/sunrise and the first hour of dark sky. Within that slot you should get about 10 to 20 mins of crossover light or blue hour where the lights from the cars and buildings are in balance with the sky colour.















--
UK wedding photographer in the Lake District
For my landscapes and fine art photography:
 
I've been doing really similar photography (urban and suburban night shots) for a while now, and the results vary from camera to camera, but I think I get the results I like most observing the following:

-Learn how much the camera allows you to blow the lights so that you can later bring back in post processing just enough detail in highlights so that they are now just large blobs of white (this depends on your own taste). For this kind of shots I found this to be crucial, since this is a upper limit that you can use and aim for. If you expose over that usually building walls and areas around lights quickly loose details.

-I don't know how the lens IS (image stabilization) that you use handles this, but mostly I get better results disabling image stabilization even when using just a wobbly tripod. In fact a couple of times I got really weird results when I inadvertently left IS on when quickly switching from handheld to tripod (or even just by setting the camera on a stone wall or something similar) - this was rare, but I couldn't find other reasons for some really weird motion blur effects observed.

-Some areas will unavoidably remain completely dark if you have large areas in your field of view. As another user pointed out, you're more likely to deal with higher dynamic range in suburban / smaller cities / towns situations since you have more dark and unlit background than in city centers where most walls and pavement is lit by at least some light. This is just an overall observations though since some city back alleys are really really dark (but depending on your area of residence I'm not sure you'd even want to venture there with a D7100 and a tripod :-D )

-Choose ISO and shutter speed depending on the contents of the settings. If the point of the picture is architecture, "highway lights" and similar I find it good to go with lowest ISO (100) and as long the exposure time as necessary to achieve the first point above. If on the other hand you're photographing something that moves against the static background such as people moving in front of buildings or ships moving on the water in ports etc, I usually have to make a choice to get one of the two: * go with a long exposure time so that the lines of the moving object blur out or * go with a much shorter time, so that the moving objects remain at least recognizably distinctive in their shapes - lights on ships are especially difficult here because they leave trails immediately but the ships themselves need much more time to become visible

-Some scenes will require multiple shots and HDR procedures; not to get the HDR look but to bring all the dynamic range into the final picture, no matter what camera you use (static human eye contrast ratio is ~6.5f stops, while dynamic human eye contrast ratio is ~20f stops).

-To get a pleasing (and closest to what was observed by a human eye at the time of taking the picture) final outcome I need to do at least some post processing most of the time. Thus means: shoot RAW in the best quality possible (I think it's 14bit loss-less for D7100). You'll probably need to lift the shadows and blacks.

-Be on the guard for CA effects and the interplay it has with software trying to fix it and night street lights that tend to produce orange-is silhouettes and shadows. Try switching CA removal on and off for your shots and observe the outlines of your objects; sometimes you'll get surprisingly good edge sharpness and/or more even colors, sometimes it will "blow". Your milage will probably vary.

OK, hope that the above will help you or someone else too with their photography. Knowing your camera response is quite important in such lighting conditions, as you already observed...don't rely too much on the LCD display that much for such pictures when it comes to color, do however keep a close watch on areas around lights that overexpose. At least this is what my experience tells me.

On a second note...I've been trying to make my mind for a while now whether to get a D7100 or not and since you're doing the type of pictures that I'm interested in (at least some of the time) I have two questions:
1.)Would you be willing to post an unedited RAW of one of such pictures for me to examine and try with my workflow ? I don't particularly care about the composition etc, but it would have to be 14 bit loss-less and contain at least some completely dark area. The two pictures in the OP would be perfect. This would help me a lot, but I completely understand if you don't what to share your raw files or have time.
2.)What is your opinion of D7100 for such pictures ? Did you come across any unexpected quirks handling it for such settings ? I know you used it in manual, but maybe you at least tried if it would AF, if the metering actually produced a usable exposure measurement etc.

Cheers

Erm....right...what are the chances of you being from the same country from all the places in the world. I apologize to the English readers, but: Lep pozdrav in veliko užitkov z uporabo D7100 želim.
 
Last edited:
Gar Ber wrote:

-I don't know how the lens IS (image stabilization) that you use handles this, but mostly I get better results disabling image stabilization even when using just a wobbly tripod.
Yes, turn of IS when on a tripod. 1) The stabilization sensors and software are not tuned to the frequencies and amplitudes encountered by tripod vibrations. 2) In the absence normal hand-held motions, the S/W may seek, to make sure that it isn't missing something, thereby creating a problem. (I assume that this is sort of like turning up the volume during a silent portion of a recording becasue you think something might be wrong.)
(static human eye contrast ratio is ~6.5f stops, while dynamic human eye contrast ratio is ~20f stops).
Interesting point. Can you point me to where that is documented? I have made my own estimates, but would like a more authoritative source.
 
Bob in Baltimore wrote:
Gar Ber wrote:

-I don't know how the lens IS (image stabilization) that you use handles this, but mostly I get better results disabling image stabilization even when using just a wobbly tripod.
Yes, turn of IS when on a tripod. 1) The stabilization sensors and software are not tuned to the frequencies and amplitudes encountered by tripod vibrations. 2) In the absence normal hand-held motions, the S/W may seek, to make sure that it isn't missing something, thereby creating a problem. (I assume that this is sort of like turning up the volume during a silent portion of a recording becasue you think something might be wrong.)
(static human eye contrast ratio is ~6.5f stops, while dynamic human eye contrast ratio is ~20f stops).
Interesting point. Can you point me to where that is documented? I have made my own estimates, but would like a more authoritative source.
 
Surely these photos are not that low iso? They are iso 250. Low iso would be 100?
You are right. I meant compared to the other shots we have been seeing here, which were an exploration of how high you can pull the ISO and still get useful pictures (fascinating examples all, but I wanted to explore something else). Compared to those, 250 is low. And it's not 100, because I didn't think of locking the ISO, another mistake I need to fix the next time.
were these really 130/10 seconds (13 seconds). if so its not mirror slap its a wobble pod problem
Yes, they were 13 seconds. I tried 10 seconds, and they didn't look good enough on the back screen, so I tried 13, and those looked right. I didn't realize the camera was adjusting the ISO behind my back.
 
Did you turn off the VR?
 
jkjond wrote:

I'd recommend trying for crossover light in these images. It varies from day to day due to cloud variations, and latitude to latitude - sometimes can give electric blue skies, other times far from it. You need to be in position between sunset/sunrise and the first hour of dark sky. Within that slot you should get about 10 to 20 mins of crossover light or blue hour where the lights from the cars and buildings are in balance with the sky colour.
Your pictures are lovely! Yes, I must try to catch the crossover light. Wow. Thank you!
 
Gar Ber wrote:

I've been doing really similar photography (urban and suburban night shots) for a while now, and the results vary from camera to camera, but I think I get the results I like most observing the following:
Your comments have been incredibly helpful, THANK YOU! In fact, I think I'm going to change some of your suggestions into a checklist and have a printed copy with me the next time I go shooting.
This is just an overall observations though since some city back alleys are really really dark (but depending on your area of residence I'm not sure you'd even want to venture there with a D7100 and a tripod :-D )
My other hobby is martial arts. If I wanted to take pictures of a really shady (haha) area, I could always call a few of my students to provide security while I'm busy looking through the viewfinder. ;-)
On a second note...I've been trying to make my mind for a while now whether to get a D7100 or not and since you're doing the type of pictures that I'm interested in (at least some of the time) I have two questions:
1.)Would you be willing to post an unedited RAW of one of such pictures for me to examine and try with my workflow ? I don't particularly care about the composition etc, but it would have to be 14 bit loss-less and contain at least some completely dark area. The two pictures in the OP would be perfect. This would help me a lot, but I completely understand if you don't what to share your raw files or have time.
No problem whatsoever. I don't usually give out my RAWs (most of the people who want my pictures wouldn't know what to do with a raw image, anyway), but I feel you provided so much good information a couple of RAW files is the least I can do. I'll PM you and we can arrange the details outside this discussion.
2.)What is your opinion of D7100 for such pictures ? Did you come across any unexpected quirks handling it for such settings ? I know you used it in manual, but maybe you at least tried if it would AF, if the metering actually produced a usable exposure measurement etc.
I like the D7100 much better than the D60 I used before. The list of things I got wrong for this shoot is quite long (forgot to turn off auto ISO, forgot to turn off VR, didn't have a remote, etc). However, the camera came to my aid: the ISO stayed fairly low, I did some initial focusing by using live view (seemed to work fine, without hunting), and the "take multiple shots in self-timer" feature has been a life saver, since in a lot of the pictures I took, the first shot was messed up, but the subsequent ones were fine (I had the self-timer set to take three shots each time I pressed release).

Not as convenient as the remote, and you can't use MUP with it, but a viable workaround, IMHO.
Cheers

Erm....right...what are the chances of you being from the same country from all the places in the world. I apologize to the English readers, but: Lep pozdrav in veliko užitkov z uporabo D7100 želim.
ROFL! Hvala!
 
In the first one, I wouldn't worry too much about trying to expose outside of the floodlights. If anything, I think that a bit of post to darken the darker tones would make for a more dramatic image.
I tried darkening the image, but then part of the blownout area became yellow (in darktable, maybe I should try it with another post software), so I started decreasing saturation to get rid of it, and wound up with a black and white image I kinda like:









--
 

Attachments

  • 2553634.jpg
    2553634.jpg
    903.1 KB · Views: 0

Keyboard shortcuts

Back
Top