bernardl wrote:
I still prefer Nikon colors comparing a9III to Z8 but the gap is now much closer as well.
But that is not the sensor, it is the combination of the processor and the editor profile automatically applied on importing the RAW file, and your monitor color biases. You can just create and set the default profile to your own taste.
There really is no such thing as SOOC color. It is another internet forums myth.
Now, there is such a thing as lens color tuning in order to provide consustant color impacts on lenses. Nikon has the best discipline on matching lenses in the industry I have found.
Feb 05, 2024 at 07:23 AM
Steve Spencer Offline • • • • • • • Upload & Sell: On
1bwana1 wrote:
But that is not the sensor, it is the combination of the processor and the editor profile automatically applied on importing the RAW file, and your monitor color biases. You can just create and set the default profile to your own taste.
There really is no such thing as SOOC color. It is another internet forums myth.
Now there is such a thing as lens color tuning in order to provide consustant color impacts od lenses. Nikon has the best discipline on matching lenses in the industry I have found.
While that is true, DXO tries to measure the color sensitivity of sensors and reports a small observable color sensitivity advantage for the Z9 over the A1 at base ISO that seems to be attributable to the native ISO 64 on the Z9. The A1 has a slight advantage at higher ISOs which seems about as big as the resolution advantage. In their measurements the Z9 also does a bit better at signal to noise ratio (but not DR) at base ISO that seems again attributable to the native ISO 64. Here are those results:
Steve Spencer wrote:
While that is true, DXO tries to measure the color sensitivity of sensors and reports a small observable color sensitivity advantage for the Z9 over the A1 at base ISO that seems to be attributable to the native ISO 64 on the Z9. The A1 has a slight advantage at higher ISOs which seems about as big as the resolution advantage. In their measurements the Z9 also does a bit better at signal to noise ratio (but not DR) at base ISO that seems again attributable to the native ISO 64. Here are those results:
None of that changes my position that each implementation may have a slight advantage or disadvantage in a certain part of the range. But they are very small and in the end are capable of producing equal image quality. That includes DR, noise, and color. They are both based on the same core architecture, which is the property of the same company, built in the same facility, using the same processes. It should surprise no one that results are so similar.
Sensors are the film of the digital age. Back in the film days which film one used had more impact on results than which camera was used to expose it. At the base level all the cameras using the Same core Sony architecture produce graphs and results that are very close. If you view Sony's influence on sensors in the context of film, in results, and number of images being produced by it, Sony's influence on photography is about what Kodak's was at its peak.
Examining minute differences is an exercise in making distinctions in a failed search for differences in result.
Feb 05, 2024 at 09:12 AM
Steve Spencer Offline • • • • • • • Upload & Sell: On
1bwana1 wrote:
None of that changes my position that each implementation may have a slight advantage or disadvantage in a certain part of the range. But they are very small and in the end are capable of producing equal image quality. That includes DR, noise, and color. They are both based on the same core architecture, which is the property of the same company, built in the same facility, using the same processes. It should surprise no one that results are so similar.
Sensors are the film of the digital age. Back in the film days which film one used had more impact on results than which camera was used to expose it. At the base level all the cameras using the Same core Sony architecture produce graphs and results that are very close. If you view Sony's influence on sensors in the context of film, in results, and number of images being produced by it, Sony's influence on photography is about what Kodak's was at its peak.
Examining minute differences is an exercise in making distinctions in a failed search for differences in result....Show more →
I totally agree. The differences between brands that use similar sensors from the same manufacturer (Sony semiconductor in many instances) are small. Sony sometimes keeps a technology to themselves and has a unique sensor (as they do now with the A9 III) but typically we see similar sensors across brands and then very similar performance. Subtle differences remain.
Nikon has been choosing lately to have a lower base ISO which improves performance when using that setting just a little at the cost of just a little performance at high ISO, but the differences are not large at either base ISO or high ISO and in practice the differences would be very hard to observe.
For colors I prefer Nikon, especially for outdoor photos. For noise and dynamic range I feel like Nikon has a slight edge, but it's close enough I can't think of a time where it was a determining factor on which camera I was going to use. I know raw workflows allow for a lot of flexibility for color in post, I went down the rabbit whole with Sony of different camera profiles in LR, using the color calibration panel in LR, and even messing with DXO profiles. They can get me closer, but I have yet to get them to the point I can say they truly match Nikon colors. The best thing I found for getting closer to Nikon colors outside with Sony was adding Zeiss UV filters to my lenses. They have a hard UV cut off that definitely helps color issues I noticed with my Sony gear. The UV color shift was almost impossible to correct in post (at least for my post skills) because it did not effect the image evenly, it was much more noticeable towards the corners. (What I am talking about is not a deal breaker or Sony would not be where it is today, but it's one of those picky things that once you notice, it's hard to ignore)
As for MP, for photos it would be hard for me to go under 30MP today, ideally I would like to stay over 40MP. I don't necessarily need 60MP, but I can't deny it is nice for landscapes, or cropping. I would like to see the Z7iii stay with 45MP or more. If so I will probably end up picking one up for landscapes / stiller subjects and use my Z8 for action photos. My dream would be to see a ZFr with the 45mp sensor, but I doubt Nikon will do that any time soon. If Nikon was to really kill the Z7 line I may end up picking up a second Z8, but I doubt they will do that looking at their competitors, and I prefer not to have two of the same exact cameras.
When viewed on a 4k monitor, my pictures from the 61mp Sony sensor are markedly more detailed than those from the 45mp Z8/Z9, even with objectively inferior lenses. Even the slight bump to the 50mp GFX sensor blows the Nikon detail away.
That said, the colors of the Nikon are much easier to work with for me, and those details would only be viewable by me, on a high resolution screen. In practical terms they are roughly equal for all other applications. It's easier and faster to craft a finished photo with a Nikon file than a Sony for me.
Steve Spencer wrote:
I totally agree. The differences between brands that use similar sensors from the same manufacturer (Sony semiconductor in many instances) are small. Sony sometimes keeps a technology to themselves and has a unique sensor (as they do now with the A9 III) but typically we see similar sensors across brands and then very similar performance. Subtle differences remain.
Nikon has been choosing lately to have a lower base ISO which improves performance when using that setting just a little at the cost of just a little performance at high ISO, but the differences are not large at either base ISO or high ISO and in practice the differences would be very hard to observe....Show more →
Yes, very small differences. I have the A1 for AF and action photography. It gives up a very little bit below ISO 100 for good gians up higher.
For landscapes portraits and other super high IQ uses I have the base ISO 64 61mpx Sony Sensor in my Leica M11. This is the highest performing full frame sensor available only being bested by a couple of the medium format sensors.
Pick your tool based on the job you want to do. Best thing is that today there are really very few bad choices.
1bwana1 wrote:
Yes, very small differences. I have the A1 for AF and action photography. It gives up a very little bit below ISO 100 for good gians up higher.
For landscapes portraits and other super high IQ uses I have the base ISO 64 61mpx Sony Sensor in my Leica M11. This is the highest performing full frame sensor available only being bested by a couple of the medium format sensors.
Pick your tool based on the job you want to do. Best thing is that today there are really very few bad choices.
Yes Steve, as you are saying, performance is very close from similar Sony sensors. For example, Bill Claff’s tests show the M11’s highest dynamic range of 11.82 at ISO 64 edges out the A7RV’s best DR of 11.69 at ISO 100 for a difference of 0.13, and the Sony is besting the Leica at ISO 100 by 0.50 where the M11 is lower at 11.19. As you and @Steve Spencer have noted, it would be virtually impossible for these minor DR differences to come into play in real-world shooting. These days there are so many more meaningful reasons to chose one camera over another. Like you have noted many times, we photographers have it so good these days with all of the excellent options to choose from.
RoamingScott wrote:
When viewed on a 4k monitor, my pictures from the 61mp Sony sensor are markedly more detailed than those from the 45mp Z8/Z9, even with objectively inferior lenses. Even the slight bump to the 50mp GFX sensor blows the Nikon detail away.
That said, the colors of the Nikon are much easier to work with for me, and those details would only be viewable by me, on a high resolution screen. In practical terms they are roughly equal for all other applications. It's easier and faster to craft a finished photo with a Nikon file than a Sony for me.
I respect that is true for you regarding the colors in processing, and it has been noted by other Nikon shooters also. But for me personally, I have to say that the files from my A7RV are processing more beautifully and easily in my last year of using compared to my previous years with various Nikon bodies (mostly landscape, some portraits, birds, night walkarounds). I use Lightroom and I choose individually the profile that I like best for each particular image, usually not the default Adobe Color. The Adobe Landscape is frequently what I prefer and it plays nicer with the colors then that same profile name did on my Nikons and my GFX100S. I also get good results on some images with the camera matching Standard profile, and another excellent choice many times is the Huelight profiles made by Colin at Color Fidelity for specific camera models.
Also, I agree with what you’re seeing on the detail differences. The 61mp sensor is providing me with meaningfully more real-world detail of the landscapes I love to experience and make images of compared to the 45mp sensors I used extensively before. Of course as you see from yours also, the GFX100/S files are a leap above the Sony detail, but for a variety of reasons I get far more joy and productivity out of my fulframe system than I did with my [sold] GFX. Like most things we argue about on forums, so much personal preferwence is involved
Ross Martin wrote:
Yes Steve, as you are saying, performance is very close from similar Sony sensors. For example, Bill Claff’s tests show the M11’s highest dynamic range of 11.82 at ISO 64 edges out the A7RV’s best DR of 11.69 at ISO 100 for a difference of 0.13, and the Sony is besting the Leica at ISO 100 by 0.50 where the M11 is lower at 11.19. As you and @Steve Spencer@ have noted, it would be virtually impossible for these minor DR differences to come into play in real-world shooting. These days there are so many more meaningful reasons to chose one camera over another. Like you have noted many times, we photographers have it so good these days with all of the excellent options to choose from.
I prefer to use the Nikon profiles in Lightroom over anything Adobe has. I would assume with the improved AWB that the A7R5 is largely as easy to color correct as Nikon. Every brand has its particular color it prefers, Nikon is definitely green and Sony is definitely red.
I also use different profiles depending on light/shadow conditions, as you mentioned. There’s no one size fits all solution, sadly!
Ross Martin wrote:
I respect that is true for you regarding the colors in processing, and it has been noted by other Nikon shooters also. But for me personally, I have to say that the files from my A7RV are processing more beautifully and easily in my last year of using compared to my previous years with various Nikon bodies (mostly landscape, some portraits, birds, night walkarounds). I use Lightroom and I choose individually the profile that I like best for each particular image, usually not the default Adobe Color. The Adobe Landscape is frequently what I prefer and it plays nicer with the colors then that same profile name did on my Nikons and my GFX100S. I also get good results on some images with the camera matching Standard profile, and another excellent choice many times is the Huelight profiles made by Colin at Color Fidelity for specific camera models. ...Show more →
1bwana1 wrote:
But that is not the sensor, it is the combination of the processor and the editor profile automatically applied on importing the RAW file, and your monitor color biases. You can just create and set the default profile to your own taste.
There really is no such thing as SOOC color. It is another internet forums myth.
Now, there is such a thing as lens color tuning in order to provide consustant color impacts on lenses. Nikon has the best discipline on matching lenses in the industry I have found.
Apologies, I don’t see how you could be right.
In the Bayer sensors most cameras use the information reaching the sensor is impacted by the custom color filters each manufacturers applies on top of the silicon. They get to decide the exact spectrum reaching each RGB pixel, how transparent these filters are (which impacts base ISO), how sharp the transition is between the spectrum of R vs those of G, the amount of cross talk resulting from the presence of micro lenses, how the analog signal coming out of the pixels is mapped to RGB values. All that happens before WB and a profile is applied and can result is large differences of the ability of the sensor to tell apart colors.
If that were not the case profiles and WB could result in different cameras having exactly the same colors and we all know it is not the case. If you own Sony and Nikon cameras and use C1 Pro just try to use the best profiles in the industry, the pro standard ones and compare.
The best proof of this being that P1 released the Trichromatic backs whose only difference vs its predecessor was a different color filtration. I used to own that back.
This is well documented and isn’t an internet myth. It’s easily explained and the impacts are important and clear.
It’s the same reason why audio lovers understand why the source is an essential component in a system and not just the speakers as some beginners in the hobby think.
Btw Sony did change the specs of their color filters from the a73s onwards. This also is well documented and was needed. The color filtration of the previous generation (a9II,…) was subpar, especially under neon light.
bernardl wrote:
Apologies, I don’t see how you could be right.
In the Bayer sensors most cameras use the information reaching the sensor is impacted by the custom color filters each manufacturers applies on top of the silicon. They get to decide the exact spectrum reaching each RGB pixel, how transparent these filters are (which impacts base ISO), how sharp the transition is between the spectrum of R vs those of G, the amount of cross talk resulting from the presence of micro lenses, how the analog signal coming out of the pixels is mapped to RGB values. All that happens before WB and a profile is applied and can result is large differences of the ability of the sensor to tell apart colors.
If that were not the case profiles and WB could result in different cameras having exactly the same colors and we all know it is not the case. If you own Sony and Nikon cameras and use C1 Pro just try to use the best profiles in the industry, the pro standard ones and compare.
The best proof of this being that P1 released the Trichromatic backs whose only difference vs its predecessor was a different color filtration. I used to own that back.
This is well documented and isn’t an internet myth. It’s easily explained and the impacts are important and clear.
It’s the same reason why audio lovers understand why the source is an essential component in a system and not just the speakers as some beginners in the hobby think.
Btw Sony did change the specs of their color filters from the a73s onwards. This also is well documented and was needed. The color filtration of the previous generation (a9II,…) was subpar, especially under neon light.
All this being true my point remains the same. You are not seeing the colors straight off the sensor. There are still processors, profiles, monitors and all the rest I brought up in between. You cant control the in camera RAW processors. But you can create/select a profile to get output from different sensor to match to the point of irrelevance. You can also Choose a monitor, color balance it and pick the monitor's gamut which will have a huge effect on color. None of this is SOOC.
1bwana1 wrote:
All this being true my point remains the same. You are not seeing the colors straight off the sensor. There are still processors, profiles, monitors and all the rest I brought up in between. You cant control the in camera RAW processors. But you can create/select a profile to get output from different sensor to match to the point of irrelevance. You can also Choose a monitor, color balance it and pick the monitor's gamut which will have a huge effect on color. None of this is SOOC.
In fact not always, garbage in garbage out. Profiles can not recover lost information.
It's a basic entropic issue, one the information is lost about a difference of color nuances, it's lost.
bernardl wrote:
Sony and Nikon have been very close in DR for years. Until the a7s3/a1/... the main difference per my own experience was in color filtration but Sony has improved a lot on this too.
I still prefer Nikon colors comparing a9III to Z8 but the gap is now much closer as well.
The big loser throughout the years in sensor quality was Canon as has been well demonstrated.
I don't know if it's due to the CFA differences or what not. If anything increasing colour filtration should come at the expense of some DR, at least the engineering DR. Perhaps you still get slightly better colour fidelity in the recovered shadow areas but it all seems quite close.
Two characteristics I find quite a nuisance are colour cast in recovered shadows and patterned noise. I guess I'll add chromo noise to that list too then the rest is much more difficult to quantify like the 'quality' of the noise.
Which is just to say we have our own thresholds as well as preferences of what's pleasing/unpleasing and how to deal with it. And it seems many people are quite pleased with Nikon's handling of these issues with their files, which I also find to be very malleable and easy to work with.
swifty168 wrote:
I don't know if it's due to the CFA differences or what not. If anything increasing colour filtration should come at the expense of some DR, at least the engineering DR. Perhaps you still get slightly better colour fidelity in the recovered shadow areas but it all seems quite close.
Two characteristics I find quite a nuisance are colour cast in recovered shadows and patterned noise. I guess I'll add chromo noise to that list too then the rest is much more difficult to quantify like the 'quality' of the noise.
Which is just to say we have our own thresholds as well as preferences of what's pleasing/unpleasing and how to deal with it. And it seems many people are quite pleased with Nikon's handling of these issues with their files, which I also find to be very malleable and easy to work with. ...Show more →
I agree with your characterization. I too find Nikon, among the many files very nice to work with. No one is arguing that in this thread.
1bwana1 wrote:
I agree with your characterization. I too find Nikon, among the many files very nice to work with. No one is arguing that in this thread.
Not saying anyone's arguing with that point.
Just pointing out it's too difficult to discuss such a multi-faceted topic as DR with just numerical numbers. So opinions about what people find superior/inferior are all valid even if it doesn't necessarily correlate with a specific tests since any one test can't tell the whole story.
What I do find objectionable is the characterisation that 'Nikon guys' are going around deluding one another in some big lie.
