Long Exposure Noise at Base ISO on Sony A7Riv

Platographer

Leading Member
Messages
998
Reaction score
356
Below is the OOC (extra fine) jpeg of a 3.2-second exposure I took in South Beach with my Sony A7Riv and Sony 16-35mm f/2.8 GM at f/10 and 28mm. My concern is that, even though I took this photo at ISO 100, there seems to be a lot of color noise in the sky. Also, the water looks off to me, like it doesn't look as good as it should. I can't really describe why, as obviously I don't expect the water to be sharp at this shutter speed (motion blurred water is the goal of long exposures). But do the water and sky look as good as you would expect at base ISO and 3.2-second SS? I would appreciate any input as to whether this looks as good as it should (without regard to composition). If anyone has a comparable jpeg of a semi-long exposure of water with the A7Riv that you are willing to upload in full resolution, that would be much appreciated as well.

View attachment 4d08c96c33b641369482c1bbedcc7177.jpg
 
Last edited:
Not much problem here. At sunset there are lots of color, both cold and warm tones. Judged from the ship in the horizon and the sea mark, also turbulent air might come into play. So you can't expect all pixels in a limited area to record the same tone.

Did you use long exposure noise reduction? This is always a good advice if you are taking long exposures in temperatures above, say, 15-20 degrees Celsius, since heat will result in more noise.
 
Not much problem here. At sunset there are lots of color, both cold and warm tones. Judged from the ship in the horizon and the sea mark, also turbulent air might come into play. So you can't expect all pixels in a limited area to record the same tone.

Did you use long exposure noise reduction? This is always a good advice if you are taking long exposures in temperatures above, say, 15-20 degrees Celsius, since heat will result in more noise.
I have long exposure noise reduction turned off. Does it affect the raw file? If it doesn't somehow get baked into the raw file, turning it on would make sense, as I shoot raw and jpeg.
 
It does affect the raw file, and you still probably want it on. If you prefer to apply it in post, you should shoot a same length exposure with the lens cap on, to capture a dark frame to be subtracted during post.

--
A7R2 with SEL2470Z and a number of adapted lenses (Canon FD, Minolta AF, Canon EF, Leica, Nikon...); A7R converted to IR.
 
Last edited:
  • Like
Reactions: Lan
It does affect the raw file, and you still probably want it on. If you prefer to apply it in post, you should shoot a same length exposure with the lens cap on, to capture a dark frame to be subtracted during post.
 
Not much problem here. At sunset there are lots of color, both cold and warm tones. Judged from the ship in the horizon and the sea mark, also turbulent air might come into play. So you can't expect all pixels in a limited area to record the same tone.

Did you use long exposure noise reduction? This is always a good advice if you are taking long exposures in temperatures above, say, 15-20 degrees Celsius, since heat will result in more noise.
I have long exposure noise reduction turned off. Does it affect the raw file? If it doesn't somehow get baked into the raw file, turning it on would make sense, as I shoot raw and jpeg.
Yes, long exposure noise reduction is applied to the data before the raw data is written to the memory card.
 
Last edited:
It does affect the raw file, and you still probably want it on. If you prefer to apply it in post, you should shoot a same length exposure with the lens cap on, to capture a dark frame to be subtracted during post.
Interesting. Is it okay to move the camera during long exposure noise reduction?
After the initial exposure yes it can move. The second noise reduction exposure is taken with the shutter closed (or lens cap on if doing it manually)
 
  • Like
Reactions: Lan
Not much problem here. At sunset there are lots of color, both cold and warm tones. Judged from the ship in the horizon and the sea mark, also turbulent air might come into play. So you can't expect all pixels in a limited area to record the same tone.

Did you use long exposure noise reduction? This is always a good advice if you are taking long exposures in temperatures above, say, 15-20 degrees Celsius, since heat will result in more noise.
I have long exposure noise reduction turned off. Does it affect the raw file? If it doesn't somehow get baked into the raw file, turning it on would make sense, as I shoot raw and jpeg.
Yes, long exposure noise reduction is applied to the data before the raw data is written to the memory card.
That's good to know. I think it is worth using most of the time. Does long exposure noise look like high iso luminosity or color noise? It seems like it may be something different entirely.
 
The noise levels don't look unexpected here.

LENR does affect the raw data. I always shoot with LENR *on* whenever I do long exposures; and I also shoot uncompressed raw.

My suggestion is that you test the effects for yourself at home though; just do a few test shots in a somewhat darkened room with LENR on, and LENR off. Have a look and see what you think?

I would be tempted to change your raw converter though; DxO PhotoLab v4 Elite's DeepPRIME NR really brings out the best in A7R4 files. There's a free 30 day trial on the DxO website, and it's available for both PC and Mac. Their latest NR is amazing :)
 
The noise levels don't look unexpected here.

LENR does affect the raw data. I always shoot with LENR *on* whenever I do long exposures; and I also shoot uncompressed raw.

My suggestion is that you test the effects for yourself at home though; just do a few test shots in a somewhat darkened room with LENR on, and LENR off. Have a look and see what you think?

I would be tempted to change your raw converter though; DxO PhotoLab v4 Elite's DeepPRIME NR really brings out the best in A7R4 files. There's a free 30 day trial on the DxO website, and it's available for both PC and Mac. Their latest NR is amazing :)
That's good to know. The example photo I posted was only a 3.2 second exposure though. Even with such a short exposure, this is an expected amount of noise? I turned on LENR for my last shoot, but it only seems to activate with bulb mode. I just shot some night photos with bracketing and some of the bright shots were 15-20 seconds, but there wasn't a "processing" period.
 
The noise levels don't look unexpected here.

LENR does affect the raw data. I always shoot with LENR *on* whenever I do long exposures; and I also shoot uncompressed raw.

My suggestion is that you test the effects for yourself at home though; just do a few test shots in a somewhat darkened room with LENR on, and LENR off. Have a look and see what you think?

I would be tempted to change your raw converter though; DxO PhotoLab v4 Elite's DeepPRIME NR really brings out the best in A7R4 files. There's a free 30 day trial on the DxO website, and it's available for both PC and Mac. Their latest NR is amazing :)
That's good to know. The example photo I posted was only a 3.2 second exposure though. Even with such a short exposure, this is an expected amount of noise? I turned on LENR for my last shoot, but it only seems to activate with bulb mode. I just shot some night photos with bracketing and some of the bright shots were 15-20 seconds, but there wasn't a "processing" period.
It looks horrible. When you zoom in you can see the chroma noise is all over, plus the luminance noise that is also quite noticeable. I recently got the A7RIII, which most people absolutely love, coming from a Sony A7II.

I am used to having no discernible noise, and no chroma noise at base ISOs. It's so bad that the default detail enhancer (in RawTherapee), which worked perfectly on shots made with my A7II, now greatly amply the luminance noise everywhere in the picture, making it unusable. It's so bad that even increasing the contrast threshold doesn't make it acceptable.

a7bc78bb5bc5401491afca0955fbba7a.jpg

Random area. All the out of focus areas have this patterns and chroma.

08e37936bdca411e889073afe792fce0.jpg

If this is normal, it's great that I noticed as I have the camera box and would go back to where it came from. Unfortunately, the A7II is no longer with me. This has me really baffled, and I guess would be the last Sony camera I'd own, other than jumping back to the A7II. I wouldn't shoot with either this camera, nor the A7RIV if the shot you did is indeed the standard of quality for the camera.

I wish more people had responded, either accepting that level of noise at base ISO or helping troubleshoot. Maybe it's normal for these cameras? Really, the A7II was not at all like this. It was crystal clear regardless if I zoomed in to 10%, 100% or 1000%. Smooth things where smooth much less a grainy blotchy rainbow.
 
Last edited:
The noise levels don't look unexpected here.

LENR does affect the raw data. I always shoot with LENR *on* whenever I do long exposures; and I also shoot uncompressed raw.

My suggestion is that you test the effects for yourself at home though; just do a few test shots in a somewhat darkened room with LENR on, and LENR off. Have a look and see what you think?

I would be tempted to change your raw converter though; DxO PhotoLab v4 Elite's DeepPRIME NR really brings out the best in A7R4 files. There's a free 30 day trial on the DxO website, and it's available for both PC and Mac. Their latest NR is amazing :)
That's good to know. The example photo I posted was only a 3.2 second exposure though. Even with such a short exposure, this is an expected amount of noise? I turned on LENR for my last shoot, but it only seems to activate with bulb mode. I just shot some night photos with bracketing and some of the bright shots were 15-20 seconds, but there wasn't a "processing" period.
It looks horrible. When you zoom in you can see the chroma noise is all over, plus the luminance noise that is also quite noticeable. I recently got the A7RIII, which most people absolutely love, coming from a Sony A7II.

I am used to having no discernible noise, and no chroma noise at base ISOs. It's so bad that the default detail enhancer (in RawTherapee), which worked perfectly on shots made with my A7II, now greatly amply the luminance noise everywhere in the picture, making it unusable. It's so bad that even increasing the contrast threshold doesn't make it acceptable.

a7bc78bb5bc5401491afca0955fbba7a.jpg

Random area. All the out of focus areas have this patterns and chroma.

08e37936bdca411e889073afe792fce0.jpg

If this is normal, it's great that I noticed as I have the camera box and would go back to where it came from. Unfortunately, the A7II is no longer with me. This has me really baffled, and I guess would be the last Sony camera I'd own, other than jumping back to the A7II. I wouldn't shoot with either this camera, nor the A7RIV if the shot you did is indeed the standard of quality for the camera.

I wish more people had responded, either accepting that level of noise at base ISO or helping troubleshoot. Maybe it's normal for these cameras? Really, the A7II was not at all like this. It was crystal clear regardless if I zoomed in to 10%, 100% or 1000%. Smooth things where smooth much less a grainy blotchy rainbow.
I was comparing on my 4K monitor some of my true long exposure shots (30+ seconds) from my Lumix S1 (which I no longer have) with similar shots from my Sony A7Riv. The A7Riv shots have more pronounced ugly green and magenta blotches in the water when viewed at 100%. I also noticed this when compared with other long exposure shots I found in these forums. I was trying to figure out if this could be because I used LENR for the S1 shots, but not the A7Riv shots. However, it's unclear to me exactly what LENR does, if anything, beyond eliminating hot pixels, which don't appear to be a problem. I plan to investigate this more when I go to St. John by taking the same photo with and without LENR on my A7Riv.
 
The noise levels don't look unexpected here.

LENR does affect the raw data. I always shoot with LENR *on* whenever I do long exposures; and I also shoot uncompressed raw.

My suggestion is that you test the effects for yourself at home though; just do a few test shots in a somewhat darkened room with LENR on, and LENR off. Have a look and see what you think?

I would be tempted to change your raw converter though; DxO PhotoLab v4 Elite's DeepPRIME NR really brings out the best in A7R4 files. There's a free 30 day trial on the DxO website, and it's available for both PC and Mac. Their latest NR is amazing :)
That's good to know. The example photo I posted was only a 3.2 second exposure though. Even with such a short exposure, this is an expected amount of noise? I turned on LENR for my last shoot, but it only seems to activate with bulb mode. I just shot some night photos with bracketing and some of the bright shots were 15-20 seconds, but there wasn't a "processing" period.
It looks horrible. When you zoom in you can see the chroma noise is all over, plus the luminance noise that is also quite noticeable. I recently got the A7RIII, which most people absolutely love, coming from a Sony A7II.

I am used to having no discernible noise, and no chroma noise at base ISOs. It's so bad that the default detail enhancer (in RawTherapee), which worked perfectly on shots made with my A7II, now greatly amply the luminance noise everywhere in the picture, making it unusable. It's so bad that even increasing the contrast threshold doesn't make it acceptable.

a7bc78bb5bc5401491afca0955fbba7a.jpg

Random area. All the out of focus areas have this patterns and chroma.

08e37936bdca411e889073afe792fce0.jpg

If this is normal, it's great that I noticed as I have the camera box and would go back to where it came from. Unfortunately, the A7II is no longer with me. This has me really baffled, and I guess would be the last Sony camera I'd own, other than jumping back to the A7II. I wouldn't shoot with either this camera, nor the A7RIV if the shot you did is indeed the standard of quality for the camera.

I wish more people had responded, either accepting that level of noise at base ISO or helping troubleshoot. Maybe it's normal for these cameras? Really, the A7II was not at all like this. It was crystal clear regardless if I zoomed in to 10%, 100% or 1000%. Smooth things where smooth much less a grainy blotchy rainbow.
I was comparing on my 4K monitor some of my true long exposure shots (30+ seconds) from my Lumix S1 (which I no longer have) with similar shots from my Sony A7Riv. The A7Riv shots have more pronounced ugly green and magenta blotches in the water when viewed at 100%. I also noticed this when compared with other long exposure shots I found in these forums. I was trying to figure out if this could be because I used LENR for the S1 shots, but not the A7Riv shots. However, it's unclear to me exactly what LENR does, if anything, beyond eliminating hot pixels, which don't appear to be a problem. I plan to investigate this more when I go to St. John by taking the same photo with and without LENR on my A7Riv.
The signature "fferreres" for surese do not understand the nature of digital noise, which is always present in any exposure. If he expect better results from other cameras, he will get a more plastic look with smoothed out finest detail.

No areas on a subject is completely clean, without color shift, tonal differences, light and shadow effects, etc., and our vision can't even differ between some colors that look identical, but might be made from a narrow vawelength or combinations of vawelengths. Or color can be caused by interference.

Long exposure noise reduction will record termic noise and hot pixels, which is subtracted from the light frame.

Long exposure noise level will highly depent on temperature. In cold climate, with themperstures lower than about the freezing point, this can be turned off. In hotter climate, be sure to use LENR (or make your own dark frames). Note that Sony cameras also have a filtering that is efficient for long exposures, except for very faint stars. This doesn't make the cameras bad for astrophotography, though. They have very high quantum efficiency and will record just as faint stars as midified Nikon and Canon cameras, real world tests have shown.

If you want to compare noise level between cameras, be sure to use exact the same camera and raw converter settings (manual everything, also WB), photograph the same subject, and use exactly the same framing.
 
Last edited:
The noise levels don't look unexpected here.

LENR does affect the raw data. I always shoot with LENR *on* whenever I do long exposures; and I also shoot uncompressed raw.

My suggestion is that you test the effects for yourself at home though; just do a few test shots in a somewhat darkened room with LENR on, and LENR off. Have a look and see what you think?

I would be tempted to change your raw converter though; DxO PhotoLab v4 Elite's DeepPRIME NR really brings out the best in A7R4 files. There's a free 30 day trial on the DxO website, and it's available for both PC and Mac. Their latest NR is amazing :)
That's good to know. The example photo I posted was only a 3.2 second exposure though. Even with such a short exposure, this is an expected amount of noise? I turned on LENR for my last shoot, but it only seems to activate with bulb mode. I just shot some night photos with bracketing and some of the bright shots were 15-20 seconds, but there wasn't a "processing" period.
It looks horrible. When you zoom in you can see the chroma noise is all over, plus the luminance noise that is also quite noticeable. I recently got the A7RIII, which most people absolutely love, coming from a Sony A7II.

I am used to having no discernible noise, and no chroma noise at base ISOs. It's so bad that the default detail enhancer (in RawTherapee), which worked perfectly on shots made with my A7II, now greatly amply the luminance noise everywhere in the picture, making it unusable. It's so bad that even increasing the contrast threshold doesn't make it acceptable.

a7bc78bb5bc5401491afca0955fbba7a.jpg

Random area. All the out of focus areas have this patterns and chroma.

08e37936bdca411e889073afe792fce0.jpg

If this is normal, it's great that I noticed as I have the camera box and would go back to where it came from. Unfortunately, the A7II is no longer with me. This has me really baffled, and I guess would be the last Sony camera I'd own, other than jumping back to the A7II. I wouldn't shoot with either this camera, nor the A7RIV if the shot you did is indeed the standard of quality for the camera.

I wish more people had responded, either accepting that level of noise at base ISO or helping troubleshoot. Maybe it's normal for these cameras? Really, the A7II was not at all like this. It was crystal clear regardless if I zoomed in to 10%, 100% or 1000%. Smooth things where smooth much less a grainy blotchy rainbow.
I was comparing on my 4K monitor some of my true long exposure shots (30+ seconds) from my Lumix S1 (which I no longer have) with similar shots from my Sony A7Riv. The A7Riv shots have more pronounced ugly green and magenta blotches in the water when viewed at 100%. I also noticed this when compared with other long exposure shots I found in these forums. I was trying to figure out if this could be because I used LENR for the S1 shots, but not the A7Riv shots. However, it's unclear to me exactly what LENR does, if anything, beyond eliminating hot pixels, which don't appear to be a problem. I plan to investigate this more when I go to St. John by taking the same photo with and without LENR on my A7Riv.
Before jumping to A7II I used a CCD based sensor, Canon 5Dc. That camera rapidly picks noise above base ISO and is also 12MP. But long exposure of all sorts at ISO 100 come out with little to no noise at all. I gave up the “clean” expectation of long exposures with A7II.

The Sony NR Long Exposure related setting does a dark frame, and removes hot pixels. There is also some other processing, which some baptized “Star Eater”. The camera cannot tell many 1 pixel stars from noise, and will therefore remove these details from the RAW.



Hope you post how the comparison goes. My preferred camera is one that has a very clean base ISO without any grainy look at ISO 100. The low contrast detail is important to me, and so far the A7II was great in that regard, provided it was not a long exposure.
 
Below are two crops from OOC jpegs of water taken with a long exposure at the same location. The first one is from the Sony A7Riv and the second one is from the Panasonic S1. One is a 55-second exposure; the other is a 50-second exposure. To my eye, it seems that the A7Riv photo has a lot more noise and splotches in the water. Is this normal? What causes this? Does LENR fix this rather than just hot pixels (which don't appear to be an issue in either photo). The Sony A7riv photo also has that band of tan/green running through it. What is that and why is it there?



Sony A7Riv 55-second exposure.
Sony A7Riv 55-second exposure.



Panasonic S1 50-second exposure.
Panasonic S1 50-second exposure.
 
Below are two crops from OOC jpegs of water taken with a long exposure at the same location. The first one is from the Sony A7Riv and the second one is from the Panasonic S1. One is a 55-second exposure; the other is a 50-second exposure. To my eye, it seems that the A7Riv photo has a lot more noise and splotches in the water. Is this normal? What causes this? Does LENR fix this rather than just hot pixels (which don't appear to be an issue in either photo). The Sony A7riv photo also has that band of tan/green running through it. What is that and why is it there?

Sony A7Riv 55-second exposure.
Sony A7Riv 55-second exposure.

Panasonic S1 50-second exposure.
Panasonic S1 50-second exposure.
I see a lot of artifacts in the first, some in the second but way more contained. The lighting is different though. Do you have another sample? As suggested, can you try Uncompressed? Seems these kinds of images, maybe, a good candidate for compression artifacts.
 
Below are versions of the images cropped from the raw files and exported full size at 100% quality to jpegs from Lightroom with low screen sharpening. Before export, I set sharpening to 0, color noise removal to 30, and noise removal to 20. Panasonic does not have a choice between compressed or uncompressed raw files, but the Sony raw file I used is uncompressed. What do you think of these? Does the A7riv photo have an unusual amount of blotchiness in the water? What causes that?



Sony A7Riv 61-second exposure
Sony A7Riv 61-second exposure



Panasonic S1 50-second exposure.
Panasonic S1 50-second exposure.
 
Below are versions of the images cropped from the raw files and exported full size at 100% quality to jpegs from Lightroom with low screen sharpening. Before export, I set sharpening to 0, color noise removal to 30, and noise removal to 20. Panasonic does not have a choice between compressed or uncompressed raw files, but the Sony raw file I used is uncompressed. What do you think of these? Does the A7riv photo have an unusual amount of blotchiness in the water? What causes that?

Sony A7Riv 61-second exposure
Sony A7Riv 61-second exposure

Panasonic S1 50-second exposure.
Panasonic S1 50-second exposure.
The A7RIV seems to have less artifacts and posterization in the water near the peer, but at the same time higher resolution. It's 61MP and a long exposure, so I get a very good impression from both camera. I have no idea if long exposures get a higher read noise penalty or what limitations they pose (other than non static camera or scene, which does not depend on the camera)
 
Why is the Lumix image som much darker? Something wrong with the sensor? Artifacts is much more hidden in the dark here.
 
Last edited:
Why is the Lumix image som much darker? Something wrong with the sensor? Artifacts is much more hidden in the dark here.
It was just a bit underexposed. The sensor was fine as far as I know. Are the splotches in the A7Riv image normal? I'm trying to figure out if I did something wrong so that I don't make the same mistake in the USVI.
 

Keyboard shortcuts

Back
Top