UGHHHHHH! HELP!

Nice camera, but not a lens I'd use for indoor photos without flash.

This is a long post. There is a summary at the end.

The problem you are seeing in these photos, the grainyness and softness, is the visible effect of noise. Noise is variation in signal from expected value. In photography, "expected" generally means the average value of surrounding pixels. When a pixel differs it stands out as brighter, darker or having a different hue.

Noise in photos comes from two sources. The camera adds some noise, but in your photos almost all the noise was variation present in the light itself even before it was captured by the camera. This variation in light is random and originates due to the quantum nature of the emission of light from its source.

The way you overcome the negative visual effects of variation in light is to collect more light in the photo. The more light you collect, the more the random variations in the light average out to something closer to the expected value.

The amount of light collected by the camera in a photo depends on three things:
  1. The amount of light falling on the sensor, per unit area
  2. The surface area of the sensor from which the image is made, and
  3. The efficiency of the sensor in converting light falling on it into an electrical charge.
Unless you change your camera, or crop an image, the latter two items are constants, so you should be most concerned with the first listed item. The amount of light falling on the sensor per unit area is called the exposure. (Many people misuse the term 'exposure" to mean the brightness of the image, Exposure and brightness are related, but on digital cameras they are not the same thing.) The exposure is determined by exactly three parameter:
  1. The amount of light in the scene (scene luminance)
  2. The T-stop the lens is set at (F-stop is a close approximation and the factor we can actually control)
  3. The length of time the sensor is exposed to light (shutter speed)
To collect as much light as possible, then, we try to do three things:
  1. Get as much light in the scene as possible, by using flash, lamps, reflectors, opening window shades and doors...
  2. Use as wide an aperture as possible.
  3. Use as slow a shutter as possible,
One can get a wider aperture on a fsater lens. And if we are open to getting a new camera, get one with as large and efficient a sensor as possible.

All of these approaches are subject to constraints, however. Often the photographer is unable or not permitted to add light to the scene. Too wide an aperture might result in too shallow a Depth of Field (DoF) or too much blur from aberrations, and lenses have limits to their max aperture. Too slow a shutter may result in blur from subject movement or camera movement.

Lets look at your two images and see what you might have done to reduce the noise in them.

The first one is shot at 170mm, 1/100 f/6.3 ISO 2500 with -1 EC. The rule of thumb for shutter speed fast enough to avoid camera motion blur is to shoot at least as fast as 1/(focal length x crop factor). At 170mm that means you should be shooting at no slower than 1/255. However you have a VR lens so you should be able to get at least two stops slower, say 1/60 and still avoid camera motion blur. If you catch the trombonist when the slide is not moving, then 1/60 should adequately freeze subject motion. Your aperture was set to f/6.3. You don't need the extra DoF this gives instead of f/5.6, The young woman behind the trombone's bell doesn't need to be sharply focused. So you could have captured one stop more light if you had shot at 1/60 f/5.6. My inadequate EXIF reader tells me you were in S mode, but not whether you had Auto-ISO set. I suspect you didn't, or if you did that you had set max ISO to 2500. Why would the camera set the aperture to f/6.3 if it was fee to set a higher ISO? For shots like these use auto ISO and set a higher limit. Doing so might have prevented the camera from narrowing aperture to f/6.3. IDK why you had EC of -1. There are no highlights that need protecting, and if anything the scene is lighter than average and thus would need a positive EC, if any. You are shooting in RAW. this is good if you know how to develop and you use the right profiles. You have greater control over shadow and highlight processing, white balance, brightness adjustment, sharpening and noise reduction. Light levels are too low to worry about ETTR. All RAW files need some sharpening and many need noise reduction.

The second photo was taken at 150mm, 1/320 f/5.6 ISO6400, again with -1 EC. Violinist's bows are almost always moving so the shutter speed may well be appropriate, and that faster shutter is the main reason for the extra noise in this image. If you think you can catch her when the bow is momentarily still, try 1/60 again, and get 2 and 1/3 stops more exposure, dropping ISO back down to 2500. The aperture is at max. Nothing you can do there. Use Auto-ISO and lose the -1 EC.

So with your current equipment, you might have improved the first photo's noise by one stop. That's not a lot. If you can catch the violinist motionless, you might get a bit over 2 stops improvement. That's better, but catching a violinist motionless isn't easy, so maybe there was no improvement to be had. To get better noise performance than this you need a faster lens of the correct focal range, or a camera with a larger or more efficient sensor. Neither is an inexpensive solution.

A 70-200mm f/2.8 VR lens would offer you two stops more light and cover the focal lengths you are using, while continuing to allow you to shoot stationary subjects at about 1/60. This will double your Signal to Noise Ratio. That's often a noticeable improvement in noise. You'd need a monopod or tripod with this lens. A monopod works best in confined space or when you need to adjust shooting position.

Somebody suggested that you use an 85mm f/1.8G lens and crop to get the same framing, That might be less of a solution than it first seems. f/1.8 is 3 and 1/3 stops faster than f/5.6, but when you crop you are throwing away light. The cropped image will be noisier than the uncropped original. To get the same framing with an 85mm from the same distance, you will have to crop to 1/4 of the original. That throws away 2 stops of the 3 1/3 stop advantage. You are also throwing away 3/4 of the pixels, so the image will not be as sharp. The 85mm f/1.8G is sharper than your 18-200mm, but it's linear resolution isn't anywhere near twice as high, so you'd be taking a noticeable sharpness penalty to get the 4/3 stop noise improvement. IMO, if you display large enough to see the noise problems, the loss of resolution will be even more visible than the noise improvement. The 85mm f/1.8 G doesn't have VR so you'd have to increase your slowest shutter from 1/60 to 1/160. ( 1/(85 * 1.5) = 1/127.5. There is no such shutter speed and the slowest speed faster than this is 1/160.) 1/160 is 4/3 stops faster than 1/60, so when subject motion allows you to shoot at 1/60 but lens stabilization doesn't, you have no net gain in noise performance with the 85mm.

One other lens you could look for is the Sigma 50-150mm f/1.8. It is discontinued so you'd have to buy used. It's specs are right (or close enough - you may have to crop a little bit, but the faster aperture makes up for it in this case) but IDK how well it performs in terms of sharpness and autofocus. Your 18-200 isn't l that sharp at 150-200mm, so the Sigma is unlikely to be significantly worse, and may be better. I'd expect the Sigma'S AF to be slow though, because 150mm f/1.8 requires large elements. That may be why Sigma replaced it in their lineup with a 50-100mm f/1.8. Elements would mass less than half as much. The 50-100 though would suffer from similar problems as the 85mm for this application.

Don't bother looking for a different APS-C camera to get higher efficiency. Efficiency differences between contemporary cameras of the same format are usually not more than about 1/3 stop, and the D500 is one of the more efficient APS-C models. If you want to see a noise improvement from a different camera, you'd need to switch to FX. A D750 currently costs a bit less than your D500, which is about the same price as a 70-200mm f/2.8 VR II. It will give about 1.1 stops better noise performance. A D610 will cost a few hundred dollars less, and give the same noise performance but its AF and metering won't be as good. However, if you move to FX you will need longer focal lengths to get the same framing. If you are currently shooting in a range from 135mm to 200mm on DX you'd need 200-300m on FX You won't find an f/2.8 zoom covering that range, only f/4 or slower so there is no gain to be had by getting an FX camera to shoot at these subject distances.

The 70-20mm f/2.8 VR II lens will give you a bigger improvement for about the same cost as a different body. You can consider an older or used model or a lens from a third party, as long as it has some form of optical image stabilization

Summary
  • Use RAW if you know how, and apply sharpening and noise reduction judiciously. Otherwise shoot JPEG.
  • Set ISO to 100 then set Auto-ISO with a high upper limit.
  • Shoot in S mode unless you find the camera not giving you max aperture (then shoot in M and set aperture to max.)
  • Use VR.
  • Set Shutter to 1/60 when you can catch the subject not moving. Adjust to faster otherwise, to freeze subject motion, but never use a faster shutter than you need to.
  • Set EC to 0.
  • Get a faster lens that covers 135mm - 200mm. For this application a 70-200mm f/2.8 with VR/VC/OIS is best, with a monopod.
 
Last edited:
Those are below expected camera performance. You can download samples at the same ISO settings and check. What were the adjustments to the shots that you applied while converting raw files? Or was it some rather extreme manipulation over JPEGs?
Are you saying that the camera should have performed better? Ie something's wrong with the camera?
 
Those are below expected camera performance. You can download samples at the same ISO settings and check. What were the adjustments to the shots that you applied while converting raw files? Or was it some rather extreme manipulation over JPEGs?
Are you saying that the camera should have performed better? Ie something's wrong with the camera?
It is more probable that something is wrong with the RAW processing. Can you discuss your sharpening and noise reduction techniques?
 
Those are below expected camera performance. You can download samples at the same ISO settings and check. What were the adjustments to the shots that you applied while converting raw files? Or was it some rather extreme manipulation over JPEGs?
Are you saying that the camera should have performed better? Ie something's wrong with the camera?
Unlikely that the camera is at fault. Your images aren't terrible, just not as good as they could be.

Were the images JPEG straight from the camera, or were they raw files processed by LR or Nikon converter? Any JPEG tweaks used in-camera? Any adjustments PP?

You probably need to experiment more to get the best set of parameters. For example, have you addressed the Exposure Compensation issue? There's very little justification for having it set at -1.0. If you are using JPEG, that's about the limit of recoverability.
 
Last edited:
Those are below expected camera performance. You can download samples at the same ISO settings and check. What were the adjustments to the shots that you applied while converting raw files? Or was it some rather extreme manipulation over JPEGs?
Are you saying that the camera should have performed better? Ie something's wrong with the camera?
Without knowing the answers to my questions I can't tell.
 
Wow, this advice is all over the place. Lets start with the fact that the D500 is a fine & capable camera. You don't need full frame. Now the catch is that any camera is simply a tool used to record an exposure. A camera records light. Period. And, as you know, you are shooting in some very poor light; it is the light that is bad and fundamentally the source of all frustration.

Next is indeed the lens. You have a slow zoom lens. When we say slow, we mean that as you zoom the lens the aperture stops down to f/5.6. The aperture is the "window" allowing light through the lens to the camera sensor. f/5.6 is a somewhat small aperture. By comparison, an aperture of f/2.8 will allow FOUR times more light though. This translates into much lower ISO numbers, which in turn translates to less noise. The catch: the most recent Nikon 70-200 f/2.8 zoom is almost $3000. The previous generation can still be had new for around $2100. And third party options will save you more money (although for hand held shots I would certainly want any lens to have some sort of stabilization). Someone else pointed out that you could get an 85mm f/1.8 prime, with the caveat being that you would now need to crop a good deal to get the same framing as the 170mm focal length. The flip side is that f/1.8 lets in over twice as much light at f/2.8! And I doubt this images are being blown up and printed as 16x20 (or even 8x10) prints so cropping should be fine.
I would suggest a 135mm f/2 prime lens. For a fully automatic lens, the Nikon. Or if you can live with manual focussing, the much cheaper Samyang.
 
Those are below expected camera performance. You can download samples at the same ISO settings and check. What were the adjustments to the shots that you applied while converting raw files? Or was it some rather extreme manipulation over JPEGs?
Are you saying that the camera should have performed better? Ie something's wrong with the camera?
Nothing is wrong with the camera.

That horizontal line could be caused by the battery being low, by a faulty card (maybe just dirty contacts), or by a faulty card reader.
 
You post makes vary good sense. Given D500 sensor is using capacitor switching, and when available light is the exposure limiting factor, one can start at ISO 400.

But we are doing a bit of generalization and guessing here. It is practical when seeking advice to present the images straight out of the camera, to rule out any possible problems down the stream.

Not seeing originals, really hard to tell. The camera is a 21 megapixel one, the images are close to 6 megapixel. With no cropping this amounts to downsizing by 1.7x. But the pixel structure on the images here looks like those images are crops posted at 100-150%. I see a surprising amount of noise reduction artifacts on the OP samples, seems important to know the source of those artifacts.

What a photographer does in this case is turning on fluorescent lights at home, putting up some test scene (including some dolls for skin tone for example), and finding the exposure that provides the acceptable results. Framing and cropping are major concerns for low light shooting conditions.

Comparing the results to the results of other people is a good way to narrow down on the problem.
Nice camera, but not a lens I'd use for indoor photos without flash.

This is a long post. There is a summary at the end.

The problem you are seeing in these photos, the grainyness and softness, is the visible effect of noise. Noise is variation in signal from expected value. In photography, "expected" generally means the average value of surrounding pixels. When a pixel differs it stands out as brighter, darker or having a different hue.

Noise in photos comes from two sources. The camera adds some noise, but in your photos almost all the noise was variation present in the light itself even before it was captured by the camera. This variation in light is random and originates due to the quantum nature of the emission of light from its source.

The way you overcome the negative visual effects of variation in light is to collect more light in the photo. The more light you collect, the more the random variations in the light average out to something closer to the expected value.

The amount of light collected by the camera in a photo depends on three things:
  1. The amount of light falling on the sensor, per unit area
  2. The surface area of the sensor from which the image is made, and
  3. The efficiency of the sensor in converting light falling on it into an electrical charge.
Unless you change your camera, or crop an image, the latter two items are constants, so you should be most concerned with the first listed item. The amount of light falling on the sensor per unit area is called the exposure. (Many people misuse the term 'exposure" to mean the brightness of the image, Exposure and brightness are related, but on digital cameras they are not the same thing.) The exposure is determined by exactly three parameter:
  1. The amount of light in the scene (scene luminance)
  2. The T-stop the lens is set at (F-stop is a close approximation and the factor we can actually control)
  3. The length of time the sensor is exposed to light (shutter speed)
To collect as much light as possible, then, we try to do three things:
  1. Get as much light in the scene as possible, by using flash, lamps, reflectors, opening window shades and doors...
  2. Use as wide an aperture as possible.
  3. Use as slow a shutter as possible,
One can get a wider aperture on a fsater lens. And if we are open to getting a new camera, get one with as large and efficient a sensor as possible.

All of these approaches are subject to constraints, however. Often the photographer is unable or not permitted to add light to the scene. Too wide an aperture might result in too shallow a Depth of Field (DoF) or too much blur from aberrations, and lenses have limits to their max aperture. Too slow a shutter may result in blur from subject movement or camera movement.

Lets look at your two images and see what you might have done to reduce the noise in them.

The first one is shot at 170mm, 1/100 f/6.3 ISO 2500 with -1 EC. The rule of thumb for shutter speed fast enough to avoid camera motion blur is to shoot at least as fast as 1/(focal length x crop factor). At 170mm that means you should be shooting at no slower than 1/255. However you have a VR lens so you should be able to get at least two stops slower, say 1/60 and still avoid camera motion blur. If you catch the trombonist when the slide is not moving, then 1/60 should adequately freeze subject motion. Your aperture was set to f/6.3. You don't need the extra DoF this gives instead of f/5.6, The young woman behind the trombone's bell doesn't need to be sharply focused. So you could have captured one stop more light if you had shot at 1/60 f/5.6. My inadequate EXIF reader tells me you were in S mode, but not whether you had Auto-ISO set. I suspect you didn't, or if you did that you had set max ISO to 2500. Why would the camera set the aperture to f/6.3 if it was fee to set a higher ISO? For shots like these use auto ISO and set a higher limit. Doing so might have prevented the camera from narrowing aperture to f/6.3. IDK why you had EC of -1. There are no highlights that need protecting, and if anything the scene is lighter than average and thus would need a positive EC, if any. You are shooting in RAW. this is good if you know how to develop and you use the right profiles. You have greater control over shadow and highlight processing, white balance, brightness adjustment, sharpening and noise reduction. Light levels are too low to worry about ETTR. All RAW files need some sharpening and many need noise reduction.

The second photo was taken at 150mm, 1/320 f/5.6 ISO6400, again with -1 EC. Violinist's bows are almost always moving so the shutter speed may well be appropriate, and that faster shutter is the main reason for the extra noise in this image. If you think you can catch her when the bow is momentarily still, try 1/60 again, and get 2 and 1/3 stops more exposure, dropping ISO back down to 2500. The aperture is at max. Nothing you can do there. Use Auto-ISO and lose the -1 EC.

So with your current equipment, you might have improved the first photo's noise by one stop. That's not a lot. If you can catch the violinist motionless, you might get a bit over 2 stops improvement. That's better, but catching a violinist motionless isn't easy, so maybe there was no improvement to be had. To get better noise performance than this you need a faster lens of the correct focal range, or a camera with a larger or more efficient sensor. Neither is an inexpensive solution.

A 70-200mm f/2.8 VR lens would offer you two stops more light and cover the focal lengths you are using, while continuing to allow you to shoot stationary subjects at about 1/60. This will double your Signal to Noise Ratio. That's often a noticeable improvement in noise. You'd need a monopod or tripod with this lens. A monopod works best in confined space or when you need to adjust shooting position.

Somebody suggested that you use an 85mm f/1.8G lens and crop to get the same framing, That might be less of a solution than it first seems. f/1.8 is 3 and 1/3 stops faster than f/5.6, but when you crop you are throwing away light. The cropped image will be noisier than the uncropped original. To get the same framing with an 85mm from the same distance, you will have to crop to 1/4 of the original. That throws away 2 stops of the 3 1/3 stop advantage. You are also throwing away 3/4 of the pixels, so the image will not be as sharp. The 85mm f/1.8G is sharper than your 18-200mm, but it's linear resolution isn't anywhere near twice as high, so you'd be taking a noticeable sharpness penalty to get the 4/3 stop noise improvement. IMO, if you display large enough to see the noise problems, the loss of resolution will be even more visible than the noise improvement. The 85mm f/1.8 G doesn't have VR so you'd have to increase your slowest shutter from 1/60 to 1/160. ( 1/(85 * 1.5) = 1/127.5. There is no such shutter speed and the slowest speed faster than this is 1/160.) 1/160 is 4/3 stops faster than 1/60, so when subject motion allows you to shoot at 1/60 but lens stabilization doesn't, you have no net gain in noise performance with the 85mm.

One other lens you could look for is the Sigma 50-150mm f/1.8. It is discontinued so you'd have to buy used. It's specs are right (or close enough - you may have to crop a little bit, but the faster aperture makes up for it in this case) but IDK how well it performs in terms of sharpness and autofocus. Your 18-200 isn't l that sharp at 150-200mm, so the Sigma is unlikely to be significantly worse, and may be better. I'd expect the Sigma'S AF to be slow though, because 150mm f/1.8 requires large elements. That may be why Sigma replaced it in their lineup with a 50-100mm f/1.8. Elements would mass less than half as much. The 50-100 though would suffer from similar problems as the 85mm for this application.

Don't bother looking for a different APS-C camera to get higher efficiency. Efficiency differences between contemporary cameras of the same format are usually not more than about 1/3 stop, and the D500 is one of the more efficient APS-C models. If you want to see a noise improvement from a different camera, you'd need to switch to FX. A D750 currently costs a bit less than your D500, which is about the same price as a 70-200mm f/2.8 VR II. It will give about 1.1 stops better noise performance. A D610 will cost a few hundred dollars less, and give the same noise performance but its AF and metering won't be as good. However, if you move to FX you will need longer focal lengths to get the same framing. If you are currently shooting in a range from 135mm to 200mm on DX you'd need 200-300m on FX You won't find an f/2.8 zoom covering that range, only f/4 or slower so there is no gain to be had by getting an FX camera to shoot at these subject distances.

The 70-20mm f/2.8 VR II lens will give you a bigger improvement for about the same cost as a different body. You can consider an older or used model or a lens from a third party, as long as it has some form of optical image stabilization

Summary
  • Use RAW if you know how, and apply sharpening and noise reduction judiciously. Otherwise shoot JPEG.
  • Set ISO to 100 then set Auto-ISO with a high upper limit.
  • Shoot in S mode unless you find the camera not giving you max aperture (then shoot in M and set aperture to max.)
  • Use VR.
  • Set Shutter to 1/60 when you can catch the subject not moving. Adjust to faster otherwise, to freeze subject motion, but never use a faster shutter than you need to.
  • Set EC to 0.
  • Get a faster lens that covers 135mm - 200mm. For this application a 70-200mm f/2.8 with VR/VC/OIS is best, with a monopod.
 
Those are below expected camera performance. You can download samples at the same ISO settings and check. What were the adjustments to the shots that you applied while converting raw files? Or was it some rather extreme manipulation over JPEGs?
Are you saying that the camera should have performed better? Ie something's wrong with the camera?
Nothing is wrong with the camera.

That horizontal line could be caused by the battery being low, by a faulty card (maybe just dirty contacts), or by a faulty card reader.
I was not referring to that line. I'm not saying something wrong with the camera.

On a side note, two-pixel lines on JPEGs are seldom a result of I/O malfunctions, more often they come from software glitches or an operator's error. Again, nothing concrete can be said without looking at the original out of camera file.
 
Not seeing originals, really hard to tell. The camera is a 21 megapixel one, the images are close to 6 megapixel. With no cropping this amounts to downsizing by 1.7x. But the pixel structure on the images here looks like those images are crops posted at 100-150%. I see a surprising amount of noise reduction artifacts on the OP samples, seems important to know the source of those artifacts.
IDK why you had EC of -1. There are no highlights that need protecting, and if anything the scene is lighter than average and thus would need a positive EC, if any.

. Elements would mass less than half as much. The 50-100 though would suffer from similar problems as the 85mm for this application.

Summary
  • Use RAW if you know how, and apply sharpening and noise reduction judiciously. Otherwise shoot JPEG.
  • Set ISO to 100 then set Auto-ISO with a high upper limit.
  • Shoot in S mode unless you find the camera not giving you max aperture (then shoot in M and set aperture to max.)
  • Use VR.
  • Set Shutter to 1/60 when you can catch the subject not moving. Adjust to faster otherwise, to freeze subject motion, but never use a faster shutter than you need to.
  • Set EC to 0.
  • Get a faster lens that covers 135mm - 200mm. For this application a 70-200mm f/2.8 with VR/VC/OIS is best, with a monopod.
Plenty of good advice, but no meaningful response from the OP.

The EC question has been asked at least three times, so it seems that the OP doesn't understand what this is all about. When struggling with low light, -1 EC is certainly not needed. I can't remember ever using as much as -1 EC under any lighting conditions (not counting HDR runs).

There was another thread recently which delved into all sorts of strange rabbit holes when it was clear that the OP had set EC to -2 and that this was the main trouble.
 
Last edited:
Not seeing originals, really hard to tell. The camera is a 21 megapixel one, the images are close to 6 megapixel. With no cropping this amounts to downsizing by 1.7x. But the pixel structure on the images here looks like those images are crops posted at 100-150%. I see a surprising amount of noise reduction artifacts on the OP samples, seems important to know the source of those artifacts.
IDK why you had EC of -1. There are no highlights that need protecting, and if anything the scene is lighter than average and thus would need a positive EC, if any.

. Elements would mass less than half as much. The 50-100 though would suffer from similar problems as the 85mm for this application.

Summary
  • Use RAW if you know how, and apply sharpening and noise reduction judiciously. Otherwise shoot JPEG.
  • Set ISO to 100 then set Auto-ISO with a high upper limit.
  • Shoot in S mode unless you find the camera not giving you max aperture (then shoot in M and set aperture to max.)
  • Use VR.
  • Set Shutter to 1/60 when you can catch the subject not moving. Adjust to faster otherwise, to freeze subject motion, but never use a faster shutter than you need to.
  • Set EC to 0.
  • Get a faster lens that covers 135mm - 200mm. For this application a 70-200mm f/2.8 with VR/VC/OIS is best, with a monopod.
Plenty of good advice, but no meaningful response from the OP.

The EC question has been asked at least three times, so it seems that the OP doesn't understand what this is all about. When struggling with low light, -1 EC is certainly not needed. I can't remember ever using as much as -1 EC under any lighting conditions (not counting HDR runs).
I'm afraid there is more to it than just EC. It might be that the brightness was raised in Photos 2, too.

I would say posting originals is absolutely necessary. Also, the advice to experiment at home is IMHO a very good one.

--
http://www.libraw.org/
 
Last edited:
Those are below expected camera performance. You can download samples at the same ISO settings and check. What were the adjustments to the shots that you applied while converting raw files? Or was it some rather extreme manipulation over JPEGs?
Are you saying that the camera should have performed better? Ie something's wrong with the camera?
Without knowing the answers to my questions I can't tell.
I used Affinity Photo and lightened and sharpened, but that's about it.
 
Those are below expected camera performance. You can download samples at the same ISO settings and check. What were the adjustments to the shots that you applied while converting raw files? Or was it some rather extreme manipulation over JPEGs?
Are you saying that the camera should have performed better? Ie something's wrong with the camera?
Without knowing the answers to my questions I can't tell.
I used Affinity Photo and lightened and sharpened, but that's about it.
EXIF says:

CreatorTool: Photos 2.0

ProfileDescription: Apple Wide Color Sharing Profile

That does not look to me like Affinity Photo signature. It may be that Photos kicked in behind the scenes and added some confusion.

Lightened - by how much? Why not to post the originals? The major source of the noise is exposure being too low, so that's the crucial piece in the puzzle, to see unmodified image. However if you already see it, you can work on increasing exposure. Just mind the crop, try to fill the frame when shooting.

With your camera raising ISO past 400 does not help noise, you need to increase exposure, not ISO.
 
The EC question has been asked at least three times, so it seems that the OP doesn't understand what this is all about. When struggling with low light, -1 EC is certainly not needed.
I'm afraid there is more to it than just EC.
Agreed, but getting the exposure in the right ball-park will help. Maybe the OP should try bracketing exposure?
I would say posting originals is absolutely necessary. Also, the advice to experiment at home is IMHO a very good one.
Yes, posting OOC JPEGs would be quite informative. To save bandwidth, crops would do.

Firing off plenty of shots would also be helpful. I hadn't counted them before, but I fired off about 1200 frames at an airshow to bring home a handful of pleasing images. Some experimentation was involved, so maybe next time I won't have to spray and pray. :-D
 
This is a problem for me, and it drives me crazy - somewhat grainy and not sharp. I am often taking pictures of plays/musicals, and then orchestras/jazz groups. The lighting is never great, so I know I'm in a tough situation - but I still can't get it right. Any comments/suggestions would be appreciate. It's not terribly unclear, but it's not right. Thanks for any help.
You've had a lot of good advice but I'd like to point out something fundamental about photography: the best results come if everything is well-matched.

You've spent a lot of money on an excellent camera but used it with a very basic lens. Starting from scratch you'd have done better to buy a cheaper (but still very good) camera such as the D7xxx series and a better lens. However, now you have the camera let's look at lens options.

Nothing will be cheap and give you the extra light you need but there are some possibilities that aren't super-expensive. Usually prime lenses are cheaper than zooms (although obviously less flexible) and third party lenses are cheaper than original-maker ones.

The Sigma 150mm F2.8 APO Macro EX DG OS HSM is much cheaper than the longer Nikkor zooms and at f/2.8 is 4 times brighter than the lens you use.

The Sigma 70-200mm F2.8 DG OS HSM is slightly more expensive than that but still much cheaper than the longer Nikkor zooms.
 
If you examine the two photos, the kid's faces are out of focus and the instruments are sharp. Is that really what you want to have in focus?

Fix that, and see if your other issues don't get a bit better!
 
Those are below expected camera performance. You can download samples at the same ISO settings and check. What were the adjustments to the shots that you applied while converting raw files? Or was it some rather extreme manipulation over JPEGs?
Are you saying that the camera should have performed better? Ie something's wrong with the camera?
These photos are what you would expect from your gear with your knowledge and experience. Someone with more knowledge and experience with your gear could do better whether its making a choice of exposure (aperture and shutter speed), ISO, AF setting, post processing, or all of the above. Anybody could also do better with better gear (faster, better lenses, sensor suited for task, etc.), if they know how to use.

With all of that being said, they aren't bad by any means. Noise is better than motion blur in my book. Motion blur is caused by too slow of shutter speed for a moving subject or focal length induced camera shake...where the photographer can't keep the camera/lens still. So at 200mm for example, you would want 1/1.5x focal length which would be 1/300 to overcome camera shake, 100mm 1/150 etc. . Experience folks can hand hold for less of a ratio. About 1/125 is my minimum for a general subject that could have normal movement. 1/500 for action movement. A lot of variables and trade offs. You just have to know what they are and what the impact will be. That comes with experience.
 
Noise is better than motion blur in my book.
Absolutely. With some good software, you can process-out a lot of color noise problems. However, if you have much motion blur, you are stuck with it.
 
I would suggest a 135mm f/2 prime lens. For a fully automatic lens, the Nikon. Or if you can live with manual focussing, the much cheaper Samyang.
I would NOT recommend the Nikon 135mm f/2. It's a DC lens and these tend to be very soft wide open. My 105 DC wasn't sharp until f/4. Don't get me wrong, the DC lenses are phenomenal portrait lenses, but not great low-light lenses.
 
This is all super helpful and I appreciate it the embarrassing thing is the -1 EC was completely unintentional. It must have been set like that from the last time I used the camera - I just forgot about it, or was experimenting. I guess I do need a better lens, although at this point can't afford one! Will be putting the other suggestions to use. Thanks so much
Not seeing originals, really hard to tell. The camera is a 21 megapixel one, the images are close to 6 megapixel. With no cropping this amounts to downsizing by 1.7x. But the pixel structure on the images here looks like those images are crops posted at 100-150%. I see a surprising amount of noise reduction artifacts on the OP samples, seems important to know the source of those artifacts.
IDK why you had EC of -1. There are no highlights that need protecting, and if anything the scene is lighter than average and thus would need a positive EC, if any.

. Elements would mass less than half as much. The 50-100 though would suffer from similar problems as the 85mm for this application.

Summary
  • Use RAW if you know how, and apply sharpening and noise reduction judiciously. Otherwise shoot JPEG.
  • Set ISO to 100 then set Auto-ISO with a high upper limit.
  • Shoot in S mode unless you find the camera not giving you max aperture (then shoot in M and set aperture to max.)
  • Use VR.
  • Set Shutter to 1/60 when you can catch the subject not moving. Adjust to faster otherwise, to freeze subject motion, but never use a faster shutter than you need to.
  • Set EC to 0.
  • Get a faster lens that covers 135mm - 200mm. For this application a 70-200mm f/2.8 with VR/VC/OIS is best, with a monopod.
Plenty of good advice, but no meaningful response from the OP.

The EC question has been asked at least three times, so it seems that the OP doesn't understand what this is all about. When struggling with low light, -1 EC is certainly not needed. I can't remember ever using as much as -1 EC under any lighting conditions (not counting HDR runs).
I'm afraid there is more to it than just EC. It might be that the brightness was raised in Photos 2, too.

I would say posting originals is absolutely necessary. Also, the advice to experiment at home is IMHO a very good one.

--
http://www.libraw.org/
 

Keyboard shortcuts

Back
Top