MicrostockGroup Sponsors


Author Topic: Longer exposure or higher ISO - Technical question  (Read 9560 times)

0 Members and 1 Guest are viewing this topic.

Beppe Grillo

« on: May 15, 2014, 04:12 »
0
Considering a still life, so with no mobile subjects.

When making long exposures the noise tends to increase.
When using higher ISO the noise tends to increase too.

The question: is it better to increase the exposure time or the ISO value?

concretely:
40 sec> 100 ISO
10 sec> 400 ISO
05 sec> 800 ISO

What will give the best result?
Or the results will be identical?


« Reply #1 on: May 15, 2014, 04:19 »
0
Increase the light :)

« Reply #2 on: May 15, 2014, 04:22 »
0
I would say exposure time once the light coming in isn't artificial like the ISO is

ACS

« Reply #3 on: May 15, 2014, 04:34 »
0
Probably the 100 iso-40 secs would be cleaner, but the contrast would be too low.

You can try to paint with light, by using mobile light sources and external flashes, this will help to reduce the exposure time.

Ron

« Reply #4 on: May 15, 2014, 05:11 »
+1
40 sec> 100 ISO  would give the best result IMO

« Reply #5 on: May 15, 2014, 06:00 »
+4
Why not try all three and then tell us what you found out?

Dook

« Reply #6 on: May 15, 2014, 06:28 »
0
You didn't mentioned what camera you have. Modern DSLRs are amazing at 800ISO, so I would choose 5sec/800ISO combination with newer cam.

Beppe Grillo

« Reply #7 on: May 15, 2014, 07:31 »
0
Consider that my question is theoretical.
I would just like  to know what could be the result according to your experience(s)
In this way when I will have to chose the best way I will not have to chose.

Increase the light :)

Unfortunately it is not always possible ;)
Imagine a still life under moon light.
Or a night landscape in the country, what solution will give the best result?

Why not try all three and then tell us what you found out?

Yes, it could be the solution, if I could have the possibility to try (and the time)
Then I should have the possibility to check the result on a good monitor.

You didn't mentioned what camera you have. Modern DSLRs are amazing at 800ISO, so I would choose 5sec/800ISO combination with newer cam.

Nikon D610
I know that it is good used at high ISO, but this does not tell me if it is not better using long exposure times


___
I tend to think that longer exposure is better too.
« Last Edit: May 15, 2014, 07:35 by Beppe Grillo »

Dook

« Reply #8 on: May 15, 2014, 07:35 »
0


[/quote]

Yes, it could be the solution, if I could have the possibility to try (and the time)
Then I should have the possibility to check the result on a good monitor.


[/quote]

Come on, you just need 55 sec for that! :)

« Reply #9 on: May 15, 2014, 07:45 »
+2
For the record, in a recent email exchange with SS, they recommend whenever possible, a lower ISO.

« Reply #10 on: May 15, 2014, 07:46 »
+1
Hi Beppe,

I just went out last weekend and did some slow shutter speed shooting down at a local pier.  I used the lowest possible ISO (100) and very long shutter speeds.  Here are a couple of examples.  These were done at 4:30 - 5:00 am with no (or a very, very slight hint that the sun is coming up - which gave me the blue sky with a 270 second exposure.  I know you were looking at shorter shutter speeds but I am posting because as long as your histogram is spot on, noise is almost non-existent.  I am shooting with a Nikon D7100 24 mp.  As I mentioned, the key is to get the exposure nailed using your histogram and only conduct minimal post processing (unless you are going to significantly downsize). These images required very little post processing.  I was light painting.

The shot of the pier (full shot) was at 7am using a big stopper 10x ND filter and that shot, too, had no noise, but the ones where my histogram was off (under exposed) got all noisy when I tweaked them in Photoshop, so I scrapped them.  A major key to noise reduction is proper exposure.


« Last Edit: May 15, 2014, 07:49 by Mantis »

Beppe Grillo

« Reply #11 on: May 15, 2014, 09:07 »
+1
Thank you Mantis!
(Nice images)

I have done a little fast test (to see @100% in Photoshop).
The result is very surprising, the difference is abyssal

« Last Edit: May 15, 2014, 09:11 by Beppe Grillo »

ShadySue

  • There is a crack in everything
« Reply #12 on: May 15, 2014, 09:44 »
+1
To do your own experiment, you just have to take two consecutive photos and change the settings between each exposure. I think that would be far more valuable than what other people might think. A lot of 'information' passed around is speculation, and even if based on experimentation, if they weren't using your set-up (particularly camera) their results might not correspond with yours.
Think yourself lucky you have a choice, in that your subject isn't moving!

Beppe Grillo

« Reply #13 on: May 15, 2014, 09:57 »
0
To do your own experiment, you just have to take two consecutive photos and change the settings between each exposure.


Yes, your are right.

But
see my post above yours ;)



Think yourself lucky you have a choice, in that your subject isn't moving!

Yes of course to have the possibility of choice is not negligible.
You could have some more possibilities too  with the help of a good taxidermist? :D
« Last Edit: May 15, 2014, 10:02 by Beppe Grillo »

Ron

« Reply #14 on: May 15, 2014, 09:59 »
-2
Right top corner is worst, the other three are similar.

ShadySue

  • There is a crack in everything
« Reply #15 on: May 15, 2014, 09:59 »
0
To do your own experiment, you just have to take two consecutive photos and change the settings between each exposure.

See my post above yours ;)

Hmmm, that was another post which was late in appearing (one happened a few days back). It doesn't happen often, but it's annoying when it does and you reply to something which has already been resolved.
(Facebook is much worse than this - often I see 'announcement' of a TV prog hours after it was broadcast, though the actual post was made much earlier.)
I didn't even get the 'someone has posted ...' warning, which I got, eg for Ron's post above.

Beppe Grillo

« Reply #16 on: May 15, 2014, 10:13 »
0
Right top corner is worst, the other three are similar.

The one in the bottom right corner is the same than the one in the top right corner but with a very higher value of denoise in Lightroom. I don't think that it is a good solution.

The good solution is low ISO + long exposure, as you have previously said.
But I am surprised that there is a so huge difference between 100 and 800 with the D610 which is known to be a very good camera when used with high ISO

« Reply #17 on: May 15, 2014, 10:54 »
+1
It's not quite that simple. With several long exposures, the camera will get warm - and then the results are getting worse. Then higher ISO might be the better solution.

« Reply #18 on: May 15, 2014, 10:54 »
0
Right top corner is worst, the other three are similar.

The one in the bottom right corner is the same than the one in the top right corner but with a very higher value of denoise in Lightroom. I don't think that it is a good solution.

The good solution is low ISO + long exposure, as you have previously said.
But I am surprised that there is a so huge difference between 100 and 800 with the D610 which is known to be a very good camera when used with high ISO
There is some low-frequency chromatic noise in the background of the bottom right - presumably put there by Lightroom as it tried to even everything out.

Beppe Grillo

« Reply #19 on: May 15, 2014, 11:25 »
0
It's not quite that simple. With several long exposures, the camera will get warm - and then the results are getting worse. Then higher ISO might be the better solution.

This is a good point.
So, after how many long exposures the camera will get too much warm and generate more noise, and how much time the sensor will need to turn to a "normal" temperature?

A solution could be little higher ISO and not so long exposure

As you say it is not so simple as it could appear
« Last Edit: May 15, 2014, 11:29 by Beppe Grillo »

« Reply #20 on: May 15, 2014, 11:59 »
+2
It looks like you had sharpening up to 50 - that's very high and could account for the noise. Sharpening at 50 seems really high - with a subject like yours, there should only be a need for very minimal sharpening, IMHO.

Noise can also be affected by which camera profile you use. I shoot with a Nikon D700 and a D5100 and find that particularly where a subject has a lot of shadows or my light wasn't optimum, using "Camera Neutral" will give me the lightest image since any increase in exposure or shadows done in LR will increase the possibility of creating noise. You can then tweak the contrast and saturation slowly to see how it affects the image.

Even though newer cameras are quite good at high ISOs, if you're shooting an image with large areas of solid color, using a lower ISO will generally give you a cleaner image, but this will change as the camera gets hotter and long exposures do increase the noise. Best idea is to bracket and see which gives you the best shot.

« Reply #21 on: May 16, 2014, 20:36 »
0
I don't think I've ever had to use such a long shutter speed, even for night shots. My blue hour landscape shots are usually in the 6 to 15 second range at f8, f11 or f16. Six to 8 seconds is enough time to smooth out water. So I don't see the point of the question. Maybe if you wanted to do star trails, but in that case, it's better to stack a lot of exposures so you don't get hot pixels.

If you're doing a still life indoors, it's better to add light with strobes or a window and reflector.

« Reply #22 on: May 16, 2014, 20:39 »
0
It looks like you had sharpening up to 50 - that's very high and could account for the noise. Sharpening at 50 seems really high - with a subject like yours, there should only be a need for very minimal sharpening, IMHO.

Noise can also be affected by which camera profile you use. I shoot with a Nikon D700 and a D5100 and find that particularly where a subject has a lot of shadows or my light wasn't optimum, using "Camera Neutral" will give me the lightest image since any increase in exposure or shadows done in LR will increase the possibility of creating noise. You can then tweak the contrast and saturation slowly to see how it affects the image.

Even though newer cameras are quite good at high ISOs, if you're shooting an image with large areas of solid color, using a lower ISO will generally give you a cleaner image, but this will change as the camera gets hotter and long exposures do increase the noise. Best idea is to bracket and see which gives you the best shot.

If you're shooting raw, it doesn't matter what profile you use except for how you want the image to look on the screen on the back of your camera. If you want the camera to record as much information as possible in the shadows and highlights, and keep it saved for you, you should shoot raw anyway.

« Reply #23 on: May 16, 2014, 21:14 »
-3
Considering a still life, so with no mobile subjects.

When making long exposures the noise tends to increase.
When using higher ISO the noise tends to increase too.

The question: is it better to increase the exposure time or the ISO value?

concretely:
40 sec> 100 ISO
10 sec> 400 ISO
05 sec> 800 ISO

What will give the best result?
Or the results will be identical?

At a constant aperture, increasing the ISO from 100 to 400 will only halve the exposure time, not quarter it as you suggest above.

You can use either method and effectively negate any resulting noise by simply shrinking the image.

« Reply #24 on: May 16, 2014, 21:19 »
0
I was talking about the profile used in LR - since, when you shoot in RAW you can change it afterwards. I know it really doesn't matter the in camera settings unless you're shooting jpegs.


 

Related Topics

  Subject / Started by Replies Last post
8 Replies
5640 Views
Last post May 30, 2008, 02:15
by Alatriste
3 Replies
3649 Views
Last post June 04, 2013, 16:37
by mtkang
15 Replies
6209 Views
Last post April 09, 2014, 16:48
by DonLand
3 Replies
4424 Views
Last post February 26, 2018, 12:45
by christiano
6 Replies
5864 Views
Last post May 22, 2018, 13:58
by Zero Talent

Sponsors

Mega Bundle of 5,900+ Professional Lightroom Presets

Microstock Poll Results

Sponsors