Low light shooting

If you would like to post, you'll need to register. Note that if you have a BCG store account, you'll need a new, separate account here (we keep the two sites separate for security purposes).

Hi All, i currently use a D500 with 500 & 300pf lenses predominantly for wildlife, as the nights are getting longer (based in the UK) a lot of my photography is late afternoon/early evening! Im currently trying to get some decent images of the local kingfishers! which are very fast flying so in low light its not the easiest!
My question is im after a second body is there a camera better than the D500 for lowlight shooting I've been looking at the D5 (D6 out of my budget) & D850!? or open to any other suggestions
Thanks
Andy
 
Hi All, i currently use a D500 with 500 & 300pf lenses predominantly for wildlife, as the nights are getting longer (based in the UK) a lot of my photography is late afternoon/early evening! Im currently trying to get some decent images of the local kingfishers! which are very fast flying so in low light its not the easiest!
My question is im after a second body is there a camera better than the D500 for lowlight shooting I've been looking at the D5 (D6 out of my budget) & D850!? or open to any other suggestions
Thanks
Andy
Assuming you can adequately fill the frame with your subject just about any modern full frame camera will do better in low light situations than a crop sensor camera. That includes the D5 but also the D850, D4S, D4, D750 or others. But that filling the frame part is important. If you can't get close to your subjects either physically or optically and will end up cropping to the DX format or smaller then you won't gain any advantage over your existing D500 and in the case of every Nikon camera except the D5/D6 you'll actually come out a bit behind in the low light/noise game any time you have to crop to the DX format or beyond.

So it depends a lot on how well you can fill the frame with your Kingfisher shots. If you can reliably get close enough to fill a full frame camera with your lenses then yes going to a full frame FX camera can help. But if you'll end up cropping heavily to get back to the kind of subject size you can achieve with your D500 then your existing camera is very hard to beat.
 
I own a D500, and I can tell you there are many bodies that are better in lowlight. I also own a D4 and it is fantastic. ISO 12,800 is no problem. I used to have a D750 and it is no too bad, and a cheaper option. I shoot almost exclusively birds, and often crop heavily. I've never seen an increase in noise beyond what was already there. If noise is a concern download a free trial of Topaz DeNoise. It is fantastic, and you will be shooting at higher ISO's than ever before.

Taken this morning at 12,800 iso.

D42_2319-small.jpg
You can only see EXIF info for this image if you are logged in.
 
Last edited:
Thank you all for your kind words and guidance I don't normally crop too much and would like a body for each of my 500 & 300pf's to take out with me
Think i will get either the D4s or a D5
Thank you once again
Regards
Andy
 
Agreed but upon the D850 its lowlight performance is worse than the D500.
I'm not sure what you're saying here but if you can achieve the same relative subject size in their respectively different frame sizes and you evaluate both images at the same output size the D850 has somewhat better noise performance than the D500. The trick when comparing cameras of different resolution or different sensor sizes is to compare at the same output size and resolution and with that perspective the D850 comes out ahead until you start substantially cropping into the frame. By the time you crop down to a DX sensor format the D500 definitely comes out ahead.

Here's a look at dynamic range for the full D850 sensor vs the full D500 sensor from DXoMark:
1601311365363.png
 
Hi All, i currently use a D500 with 500 & 300pf lenses predominantly for wildlife, as the nights are getting longer (based in the UK) a lot of my photography is late afternoon/early evening! Im currently trying to get some decent images of the local kingfishers! which are very fast flying so in low light its not the easiest!
My question is im after a second body is there a camera better than the D500 for lowlight shooting I've been looking at the D5 (D6 out of my budget) & D850!? or open to any other suggestions
Thanks
Andy

I own a D500, and I can tell you there are many bodies that are better in lowlight. I also own a D4 and it is fantastic. ISO 12,800 is no problem. I used to have a D750 and it is no too bad, and a cheaper option. I shoot almost exclusively birds, and often crop heavily. I've never seen an increase in noise beyond what was already there. If noise is a concern download a free trial of Topaz DeNoise. It is fantastic, and you will be shooting at higher ISO's than ever before.

I got intrigued by the discussion that @Capturingtheunseen.com initiated some time ago in one of his threads and exactly for the reason you have been opening this one I have just got hands on a mint used D4S. Sorry to others for repeating myself but you can tweak data by means of software, but you can't cheat physics. The prime parameter for avoiding noise in the first place is sensor pixel area and this will remain more or less the same unless a technological breakthrough will come with a totally new sensor technology. Thus a FX body will always have better lowlight performance compared with a DX body of the same resolution - well, and at least with a similar generation of the cameras. You might want to take a look here regarding sensor pixel area ...

I'd love to swap my D7200 for a D500 primarily because of the combination of reach, frame rate and buffer size.
However, if you really have to push the limits in terms of low light it has to be FX I guess. Which is logical because unless the innovation wall for current CMOS sensor technology can be overcome the pixel area size remains to be the central parameter for noise tolerance.

D7200 --> 3,92 µm --> 15,37 µm²
D500 --> 4,2 µm --> 17,64 µm² --> +15% to the D7200 --> fits to moderate noise improvement of D500 compared to D7200
D4/s --> 7,3 µm --> 53,29 µm² --> +200% to the D500 --> well, what will that mean in the end :sneaky:.

For me it was the same question as for you. Without putting my wallet on a DX body I couldn't get the "reach" for a D5 or D6 :p. As you can see I in fact took a step back in resolution. So why on earth should one get a camera with less pixels ??? Well, reason one was just explained.

IF lowlight = TRUE THEN DO
MAXIMISE pixelsize_available
ENDIF

The other reason is @Steve - well, to be exact it is his explanations regarding the relation between motion blur sensitivity and pixel size.

So, if you don't need to print huge and you want sharp pictures in low light, go for a FX camera with low(er) resolution.
But of course there is a downside to it. If you shoot the 500PF you'd loose 250mm reach compared to a DX body and you can't compensate for that with a TC on the FX body. And this is not about of loosing 50mm, it is about loosing most of your AF sensors because your camera will recognize the combo as an f8 lens.
With your primary targets this could become a bit difficult, unless you can live with "true" 500mm reach without having to crop too much.

Unfortunately I didn't get much opportunity to shoot and test yet, but what I can say up to now is that with my D4S
  • I gain at least 2 stops of low light tolerance in terms of noise if looking at unprocessed RAWs even against my D750, which is still better or in the best casse almost equal to DX sensor camera and
  • I can realize just from the quick and dirty tests that my keeper rate, especially handholding, is clearly better than before and because weight and balance hasn't changed so much, the only reason can be what @Steve has explained about motion blur.
I know it could be a tough decision, but maybe there is a chance for you to get hands one of these bodies on a rental basis or from a friend and you just give it a try ... even after having mine only a couple of days, I am almost certain you'll love it.

All the best ...
 
Last edited:
I got intrigued by the discussion that @Capturingtheunseen.com initiated some time ago in one of his threads and exactly for the reason you have been opening this one I have just got hands on a mint used D4S. Sorry to others for repeating myself but you can tweak data by means of software, but you can't cheat physics. The prime parameter for avoiding noise in the first place is sensor pixel area and this will remain more or less the same unless a technological breakthrough will come with a totally new sensor technology. Thus a FX body will always have better lowlight performance compared with a DX body of the same resolution - well, and at least with a similar generation of the cameras. You might want to take a look here regarding sensor pixel area ...



For me it was the same question as for you. Without putting my wallet on a DX body I couldn't get the "reach" for a D5 or D6 :p. As you can see I in fact took a step back in resolution. So why on earth should one get a camera with less pixels ??? Well, reason one was just explained.

IF lowlight = TRUE THEN DO
MAXIMISE pixelsize_available
ENDIF

The other reason is @Steve - well, to be exact it is his explanations regarding the relation between motion blur sensitivity and pixel size.

So, if you don't need to print huge and you want sharp pictures in low light, go for a FX camera with low(er) resolution.
But of course there is a downside to it. If you shoot the 500PF you'd loose 250mm reach compared to a DX body and you can't compensate for that with a TC on the FX body. And this is not about of loosing 50mm, it is about loosing most of your AF sensors because your camera will recognize the combo as an f8 lens.
With your primary targets this could become a bit difficult, unless you can live with "true" 500mm reach without having to crop too much.

Unfortunately I didn't get much opportunity to shoot and test yet, but what I can say up to now is that with my D4S
  • I gain at least 2 stops of low light tolerance in terms of noise if looking at unprocessed RAWs even against my D750, which is still better or in the best casse almost equal to DX sensor camera and
  • I can realize just from the quick and dirty tests that my keeper rate, especially handholding, is clearly better than before and because weight and balance hasn't changed so much, the only reason can be what @Steve has explained about motion blur.
I know it could be a tough decision, but maybe there is a chance for you to get hands one of these bodies on a rental basis or from a friend and you just give it a try ... even after having mine only a couple of days, I am almost certain you'll love it.

All the best ...
Thank you for taking the time out to explain this! unfortunately i won't be able to hire one and no one i know shoots with them, but as i need an extra body i will be getting one of these Ive seen D4s & D5 both at reasonable prices I've now just got to make my mind up in which one i go for!!
Thanks
Andy
 
I got intrigued by the discussion that @Capturingtheunseen.com initiated some time ago in one of his threads and exactly for the reason you have been opening this one I have just got hands on a mint used D4S. Sorry to others for repeating myself but you can tweak data by means of software, but you can't cheat physics. The prime parameter for avoiding noise in the first place is sensor pixel area and this will remain more or less the same unless a technological breakthrough will come with a totally new sensor technology. Thus a FX body will always have better lowlight performance compared with a DX body of the same resolution - well, and at least with a similar generation of the cameras. You might want to take a look here regarding sensor pixel area ...



For me it was the same question as for you. Without putting my wallet on a DX body I couldn't get the "reach" for a D5 or D6 :p. As you can see I in fact took a step back in resolution. So why on earth should one get a camera with less pixels ??? Well, reason one was just explained.

IF lowlight = TRUE THEN DO
MAXIMISE pixelsize_available
ENDIF

The other reason is @Steve - well, to be exact it is his explanations regarding the relation between motion blur sensitivity and pixel size.

So, if you don't need to print huge and you want sharp pictures in low light, go for a FX camera with low(er) resolution.
But of course there is a downside to it. If you shoot the 500PF you'd loose 250mm reach compared to a DX body and you can't compensate for that with a TC on the FX body. And this is not about of loosing 50mm, it is about loosing most of your AF sensors because your camera will recognize the combo as an f8 lens.
With your primary targets this could become a bit difficult, unless you can live with "true" 500mm reach without having to crop too much.

Unfortunately I didn't get much opportunity to shoot and test yet, but what I can say up to now is that with my D4S
  • I gain at least 2 stops of low light tolerance in terms of noise if looking at unprocessed RAWs even against my D750, which is still better or in the best casse almost equal to DX sensor camera and
  • I can realize just from the quick and dirty tests that my keeper rate, especially handholding, is clearly better than before and because weight and balance hasn't changed so much, the only reason can be what @Steve has explained about motion blur.
I know it could be a tough decision, but maybe there is a chance for you to get hands one of these bodies on a rental basis or from a friend and you just give it a try ... even after having mine only a couple of days, I am almost certain you'll love it.

All the best ...
 
Could you provide a reference to the article by @Steve regarding motion blur with different sensors/cameras?
It's a long part of the shutter speed chapter in my Exposure and Metering book. Basically, the blur is always the same, but higher res sensors are more capable of showing it to you.
 
When I compared the D500, D850, and D5 cameras photographing birds in flight the D5 was significantly better at locking focus when a bird was approaching the camera directly with its flight path. I would expect the D6 to be better than the D5 although both are low resolution cameras producing a DX file that is little better than the D1x in terms of image resolution should any cropping be needed. This is a situation where the 500mm PF is a great lens to have.
 
Hi All, i currently use a D500 with 500 & 300pf lenses predominantly for wildlife, as the nights are getting longer (based in the UK) a lot of my photography is late afternoon/early evening! Im currently trying to get some decent images of the local kingfishers! which are very fast flying so in low light its not the easiest!
My question is im after a second body is there a camera better than the D500 for lowlight shooting I've been looking at the D5 (D6 out of my budget) & D850!? or open to any other suggestions
Thanks
Andy
Hi Andy
Ive had the D4s for some years now and have no regrets, sure, it doesn't have a bucket of pixels , only 16mp, but they are very good ones :)
You can even produce noise with a bad exposure (always)
My budget pick right now would be a D5, (dual cards that are the same, preferably XQD)
-was a big upgrade with the D5, so you will have the best of both worlds
Just a note on noise,
We all come across noise shooting wildlife because light is usually not premium when shooting wildlife, well, a lot of times horrible :)
I would recommend getting Steve's Noise Reduction Video, it is absolutely amazing, yes! that is right, the best noise vid out there,

No matter what camera you have, get the exposure wrong and the noise will be amplified
 
Hi Andy
Ive had the D4s for some years now and have no regrets, sure, it doesn't have a bucket of pixels , only 16mp, but they are very good ones :)
You can even produce noise with a bad exposure (always)
My budget pick right now would be a D5, (dual cards that are the same, preferably XQD)
-was a big upgrade with the D5, so you will have the best of both worlds
Just a note on noise,
We all come across noise shooting wildlife because light is usually not premium when shooting wildlife, well, a lot of times horrible :)
I would recommend getting Steve's Noise Reduction Video, it is absolutely amazing, yes! that is right, the best noise vid out there,

No matter what camera you have, get the exposure wrong and the noise will be amplified
I agree about Steve's noise reduction video -- it describes a great technique that has helped me tremendously. But I also keep hearing about Topaz AI noise reduction -- has anyone had enough experience with that program to know if the "AI" offers a step up? Some people seem to be very impressed with it.
 
Yes, it's an excellent program and does some magical work, but always bear in mind that a lot depends upon what you start with and how you perceive the changes to the image as you tweak the settings whilst you play with it. Also remember it is a resources hog and puts a big demand on your graphics controller (GPU) and the PC itself. These 4 things are essential for running an efficient PC.
  1. CPU - the microchip that runs everything
  2. GPU - the graphics card - for photography, I would always recommend a dedicated card rather than using the built in onboard GPU.
  3. RAM Memory - what the computer uses for processing.
  4. Storage Memory (hard drive) - never fill your drive to capacity. Drives need free space to work with and I always recommend having a minimum of 2 drives. One with nothing but the operating system and any others for storage.
Regarding RAM, I normally run 32gb and Topaz flies but I am currently stuck with an old PC with only 16gb RAM and low level onboard GPU and it sucks big time.
 
I agree about Steve's noise reduction video -- it describes a great technique that has helped me tremendously. But I also keep hearing about Topaz AI noise reduction -- has anyone had enough experience with that program to know if the "AI" offers a step up? Some people seem to be very impressed with it.

I don't have much experience with it because I am still busy moving from the Nikon software to Capture One for post processing. But I have tried a couple of things within the 30 days trial period, so please take my thoughts as a first impression.

It seems to produce really good results but I am not yet able to compare them with the results of other tools, being it the built-in noise reduction tool of the raw conversion and processing software out there like Photoshop/Lightroom, Capture One, Capture NX-D and such alike.

However, beside the sheer result in terms of noise reduction there are more aspects to think about.
Yes, like @Oztours said, you need a bit of patience if your PC has not the power ...

Yes, it's an excellent program and does some magical work, but always bear in mind that a lot depends upon what you start with and how you perceive the changes to the image as you tweak the settings whilst you play with it. Also remember it is a resources hog and puts a big demand on your graphics controller (GPU) and the PC itself.

..., but I think the way to look at it is comparing the total time needed for denoising photo.

Step 1:
The first priority is always to avoid causing noise or at least keep it to a minimum when taking the picture in the first place (Extra processing time = 0 :)).

Step 2:

If you still have too much of it the next step would be to use the built-in function of your post processing software that you can use
  • either flat, i.e. apply it plain on the complete picture (Extra processing time = very small :)) ...
  • or selective on certain parts of your image by means of masks/layers (Extra processing time = small :) to considerable :cautious:,. depending on your routine level and complexitiy of the picture).
At this stage let's assume you use it plain up to the level where you don't have negative effects on overall IQ like e.g. loosing fine details

Step 3:

If this is still not good enough the next two option are

  1. Using the selective noise reduction of your processing software
  2. Using a sparate denoising tool like Topaz Denoise.
Both options have their pros and cons.

Option 1 ...
... is a bit more demanding, because you need to spend some time with setting up the masks/layers. For people living in their post procsssing software this might not be a big deal but for others it could - and that is where things like Topaz Denoise become tempting for me :unsure:.
On the other hand you are able to keep a complete non-destructive worklflow based on your RAW's under just one piece of software having all processing done based on the original RAW file.

Option 2 ...
... is easier to handle, as you just drop a complete image in the software's throat and it does the job for you. Topaz Denoise is quite clever in providing proposals for setting the parameters in order to get a very good result without having to dive in the details. The downside is that you have to break up the technical workflow. What does this mean ?

Topaz Denoise can read my Nikon RAW files (NEF) directly, but can store only in file formats outside the RAW workflow, i.e. (JPG, PNG, TIFF 8 or 16 bit, DNG). However, the main post processing software works non-destructive, i.e. stores all processing steps in sidecar files leaving the original RAW flie untouched. Topaz will thus open a NEF file always in its original state ignoring the processing you might have already done to it. The alternative would be to do all the processing apart from final noise reduction in the main processing software, then export your image and use the exported image as input for Topaz Denoise. The implications of this will be that
  • the noise will remain present while you do your processing and may optically get in your way, and
  • you will "process" the noise as well (e.g. sharpen it)
I you want to work the other way round in order to avoid this you would need to run your original image through the denoising tool and then do the rest of the processing. But then you have to do it based on a DNG or a 16 TIFF instead of your RAW file and you could have lost information stored in it.

As I said I didn't have the chance to learn everything yet, but something I also realized was that Topaz Denoise has bent the colour settings in images I procesed with it, but I am not sure whether this has something to do with the settings of my processing software and/or color profiles.

I would be really interestes to hear from experienced Topaz Denoise users how they organize their workflow if they are NOT using Adobe software as main post processing software, but a software working based on the native Raw formats of the camera producers. The DNG format was invented and patented by Adobe with the intention of creating something like a PDF standard for RAW files of imaging systems, but fortunately they never got even close to this point (having standards is a goods thing but monopolies never are ...).

You might want to take a look at two articels written by Nasim Mansurov at Photographylife about RAW versus DNG to get an idea.
I am still very interested in Topaz Denoise myself but I haven't had a chance to go deep enough to find out what would be the best way for me to use it.
Currently I think the first way I try is:
  1. Do the complete processing of my images first including using the built-in denoise function of my processing software up to the limit where I would have to start working selectively with masks / layers in order avoid killing details with it, so that I can reduce noise already in step one as far as possible.
  2. If the result is o.k., do final export to JPG, else export the processed RAW image to 16bit TIFF
  3. Running this 16 bit TIFF through Topaz Denoise and save as JPG from there.
Once I have learned to properly use the layers/Mask functions of my new processing software I would try selective denoising with the built-in tool, compare with the results I get with Topaz and then decide which way to go.

I am working in the area of proces design and IT solutions for industrial manufacturing management and there is an inflationary trend to calling everything AI that is too difficult for the sales and marketing guys to explain to the client :D.

Another consideration is storage capacity. If you use the Denoise tool occasionally just fpr critical pictures it shoudln't matter, but i f I run a NEF with 20-24MB through Topaz Denoise the result is as DNG or TIFF file with around 140 MB, i.e. 6 to 7 times bigger !

Looking forward to how this discussion continues :sneaky:
 
I prefer using my D500 for wildlife / birding over my D850, which does have a better SNR to just past ISO 51200, because it's easier to fill the frame with the DX body yielding more detail even in lower light conditions. The technique I use to reduce noise is to take several shots then align and average them together. For example, if you take 2 images at ISO 3200 and average them, you will get an ISO 1600 result. If you take and average 4 shots you will get an ISO 800 result. This assumes your subject is not moving, although you can use some PP tricks to correct for slight movement.

There is another technique to consider if shooting manual with auto ISO. In lower light I'll take a shot or two at the appropriate shutter speed for the focal length of the lens, then take 4-6 shots at 1/2 that shutter speed, then about a dozen shots at 1/2 that shutter speed and maybe even go a little slower beyond that. Each time the ISO decreases by 1/2 and there is a good chance at least one of the images in a sequence, usually in the middle, will be sharp.

I try not to buy new hardware until I have exhausted other options, but that's just me.
 
The technique I use to reduce noise is to take several shots then align and average them together. For example, if you take 2 images at ISO 3200 and average them, you will get an ISO 1600 result. If you take and average 4 shots you will get an ISO 800 result.

Doesn't the noise reduce by the square root of the number of measurements? That is 4 signals - photographs in this case - superimposed would show a twofold reduction in noise. That is certainly true in physical signal measurements and is called signal averaging and which I used to do in my former working life...
cheers,
Alex
 
Thank you all for your kind words and guidance I don't normally crop too much and would like a body for each of my 500 & 300pf's to take out with me
Think i will get either the D4s or a D5
Thank you once again
Regards
Andy
Get the D4s. It is really a lot better in low light than the D5 is. I have shot both and I was really disappointed in the D5 noise, compared to the D4s.
 
I agree about Steve's noise reduction video -- it describes a great technique that has helped me tremendously. But I also keep hearing about Topaz AI noise reduction -- has anyone had enough experience with that program to know if the "AI" offers a step up? Some people seem to be very impressed with it.
Hi Jim i have Topaz A1 noise reduction, and to be honest i prefer to use lightroom but i find that both take the detail away from the picture that's why im looking for a better camera for low light!
Try Topaz for yourself as you get it on a free 1 month no obligation trial!!
Regards
Andy
 
I prefer using my D500 for wildlife / birding over my D850, which does have a better SNR to just past ISO 51200, because it's easier to fill the frame with the DX body yielding more detail even in lower light conditions. The technique I use to reduce noise is to take several shots then align and average them together. For example, if you take 2 images at ISO 3200 and average them, you will get an ISO 1600 result. If you take and average 4 shots you will get an ISO 800 result. This assumes your subject is not moving, although you can use some PP tricks to correct for slight movement.

There is another technique to consider if shooting manual with auto ISO. In lower light I'll take a shot or two at the appropriate shutter speed for the focal length of the lens, then take 4-6 shots at 1/2 that shutter speed, then about a dozen shots at 1/2 that shutter speed and maybe even go a little slower beyond that. Each time the ISO decreases by 1/2 and there is a good chance at least one of the images in a sequence, usually in the middle, will be sharp.

I try not to buy new hardware until I have exhausted other options, but that's just me.
Thanks for this i think im asking the impossible from my camera, shooting a fast small bird in lowlight!! i can just get by when he's stationary on a branch but the diving shots are a no go!
Im going to wait until the new Z6 &7s come out before i make my decision!
 
Hi Andy
Ive had the D4s for some years now and have no regrets, sure, it doesn't have a bucket of pixels , only 16mp, but they are very good ones :)
You can even produce noise with a bad exposure (always)
My budget pick right now would be a D5, (dual cards that are the same, preferably XQD)
-was a big upgrade with the D5, so you will have the best of both worlds
Just a note on noise,
We all come across noise shooting wildlife because light is usually not premium when shooting wildlife, well, a lot of times horrible :)
I would recommend getting Steve's Noise Reduction Video, it is absolutely amazing, yes! that is right, the best noise vid out there,

No matter what camera you have, get the exposure wrong and the noise will be amplified
Thanks Gary i will take a look at the video!
 
Doesn't the noise reduce by the square root of the number of measurements? That is 4 signals - photographs in this case - superimposed would show a twofold reduction in noise. That is certainly true in physical signal measurements and is called signal averaging and which I used to do in my former working life...
cheers,
Alex

Here’s my reference.
 
Back
Top