Light Gathering, noise and diffraction: an XH2 question

[silversportsman]

Without quoting any one thread in particular , let me paraphrase what I think I have gathered in multiple threads regarding the X-H2 vs the X-H2s. An XH2 needs good light to give the best results. Using the XH2 with a long lens at apertures of F8 and smaller may lead to diffraction and noisy images that even when cropped my not give you the best results, largely due to a poor signal to noise ratio. If this is generally accurate, and I am not saying it is exactly so, when purchasing a lens whether prime or zoom would it not be to one's advantage (again generally speaking) to use the fastest lens that meets your use case? I am contemplating purchasing a couple more lenses to use on my Xh2 and I want to get the very best results in terms of image quality. I am contemplating lenses such as the XF 33/1.4 and the Sigma 17-40/1.8. My use cases do not include long telephoto applications or BIF. Would be interested in your thoughts.

 

[Morris0]

Without quoting any one thread in particular , let me paraphrase what I think I have gathered in multiple threads regarding the X-H2 vs the X-H2s. An XH2 needs good light to give the best results. Using the XH2 with a long lens at apertures of F8 and smaller may lead to diffraction and noisy images that even when cropped my not give you the best results, largely due to a poor signal to noise ratio. If this is generally accurate, and I am not saying it is exactly so, when purchasing a lens whether prime or zoom would it not be to one's advantage (again generally speaking) to use the fastest lens that meets your use case? I am contemplating purchasing a couple more lenses to use on my Xh2 and I want to get the very best results in terms of image quality. I am contemplating lenses such as the XF 33/1.4 and the Sigma 17-40/1.8. My use cases do not include long telephoto applications or BIF. Would be interested in your thoughts.

Subject and use are important for this discussion. You already have:

XF 18-55mm F2.8-4 R LM OIS

Very good

XF 16mm F1.4 R WR

Excellent

XF 90mm F2

Superb

XF 70-300 F4-5.6 R LM OIS WR

Very good

Unless you need bokeh and/or light gathering of the Sigma 17-40 f1.8 you are not going to gain much from that lens. As you are also considering the XF 33 f1.4, possibly you want to bokeh and light.

I personally really enjoy fast lenses and the bokeh they produce as they can make your photos look a lot nicer.

Morris

 

[Erik Baumgartner]

Image noise is really about the same for both sensors. The 40MP sensor will look a bit noisier at 100% on your computer screen (pixel noise), but only because you’re looking closer at it. Diffraction will never look worse at 40 vs. 26MP, but it can limit the maximum resolution available from the 40MP sensor, but no more than what you’ll see at 26MP. With more image data to work with, A.I. noise reduction is typically more effective at 40MP as it can better distinguish between image detail and noise for a sharper/cleaner result. You will get a bit more out of the best primes with the 40MP sensor if you are careful to optimize your shooting discipline.

 

[MrAndreas]

Thank you. I agree with every word you said, bookmarked this post for future reference.

--
https://www.viewbug.com/member/stesinou

 

[Earl Goodson]

Using the XH2 with a long lens at apertures of F8 and smaller may lead to diffraction and noisy images that even when cropped my not give you the best results, largely due to a poor signal to noise ratio. If this is generally accurate, and I am not saying it is exactly so, when purchasing a lens whether prime or zoom would it not be to one's advantage (again generally speaking) to use the fastest lens that meets your use case?

Not necessarily. People often treat diffraction as some magic aperture value where an image is unusable. When it's really a slow, generally imperceptible decline in sharpness.

I've read comments from folks saying they never go past f/4-5.6, etc, for maximum sharpness. That's absurd. Because one, it's not that big a deal. And two, depth of field gains will often increase perceived resolution, despite diffraction losses. Meaning, if I shoot a telephoto landscape at f/4 to avoid diffraction, the image foreground may still look soft because sections are out of focus. If I use f/8-11 instead, it will look better, despite hitting diffraction territory.

Diffraction isn't that big a deal; use whatever aperture value best fits your creative intention, knowing it comes at the cost of max resolution. And compensate in post with added sharpness, if needed. Capture One has a Diffraction Correction tool that's very handy to apply at the start.

 

[Bill Ferris]

An XH2 needs good light to give the best results.

All cameras benefit from working with good light.

Using the XH2 with a long lens at apertures of F8 and smaller may lead to diffraction and noisy images that even when cropped my not give you the best results, largely due to a poor signal to noise ratio.

The above suggests you may be making a couple of incorrect assumptions.

The first is that smaller pixels reduce a sensors ability to detect and process light. However, it's not pixel size that determines the light-collection ability of a sensor. It's sensor surface area that does. Since the H2 and H2S have same size sensors, both collect the same total light when working with the same exposure, and their photos display the same amount of shot noise.

Folks who complain that the H2 is noticeably noisier than the H2S are usually responding to a comparison of 100% views. A 100% view of a 40MP photo will be zoomed in farther than a 100% view of a 26MP photo. That increased magnification makes noise more prominent.

A comparison of photos displayed at the same size makes the read noise differences between those sensors quite subtle. It's not an issue: https://www.dpreview.com/reviews/im...7&x=0.5757840228378028&y=-0.19987511201612654

The second incorrect assumption possibly being made is that higher megapixel sensors are more susceptible to diffraction. This notion also comes from viewing and comparing photos at 100%. The more magnification used when studying an image, the less sharp fine details will look.

However, when photos are compared at the same size, the increased resolution of the H2 (and other 40MP sensors) is on display. It's photos show slightly more detail than shown in 26MP photos. It's not a game changing difference but it is noticeable.

If this is generally accurate, and I am not saying it is exactly so, when purchasing a lens whether prime or zoom would it not be to one's advantage (again generally speaking) to use the fastest lens that meets your use case? I am contemplating purchasing a couple more lenses to use on my Xh2 and I want to get the very best results in terms of image quality. I am contemplating lenses such as the XF 33/1.4 and the Sigma 17-40/1.8. My use cases do not include long telephoto applications or BIF. Would be interested in your thoughts.

It would be easy to encourage you to buy the most expensive, highest quality lenses you can get. That's a safe answer. I'll advise that your lens choice depends on the kind of photography you enjoy doing. What subjects? Do you prefer deep depths of field or shallow? Do you work in low light or bright environments? With or without a tripod?

Whichever lenses you choose, my advice is to set diffraction aside as a concern. It's not going to be an issue unless you're in the f/11 and higher f-stop range. All APS-C systems are near the threshold of where diffraction becomes an issue in that range.

Good luck.

 

[MrAndreas]

However, when photos are compared at the same size,

... downscaled for viewing on your display device...

the increased resolution of the H2 (and other 40MP sensors) is on display.

No. Your display device is rarely an 8 K display. I'm not aware of any display device capable of reproducing more than "8k" (that's ~33MP). But it is at a pixel level. At a human viewing level (not aliens), from a realistic viewing distance, we are bound to be around 8MP - 12MP perceivable resolution.

A 40MP sensor offers benefits here because downsampling from 40MP to the display device's MP reduces noise and improves overall picture IQ. This is true.

It's photos show slightly more detail than shown in 26MP photos. It's not a game changing difference but it is noticeable.

I agree. At the pixel level, we have an ~f/8 boundary with the 26MP APS-C sensor, where diffraction starts to interfere, and with 40MP, it starts earlier. It doesn't matter for real-world viewing on real-world devices with human (not alien) eyes. We are viewing downscaled images on our real-world devices, anyway. Do you agree?

Whichever lenses you choose, my advice is to set diffraction aside as a concern.

I second this.

 

[Bill Ferris]

However, when photos are compared at the same size,

... downscaled for viewing on your display device...

Or upscale. The DPR Studio Scene upscales photos made with smaller format cameras.

the increased resolution of the H2 (and other 40MP sensors) is on display.

No. Your display device is rarely an 8 K display.

An 8K display isn't necessary. I see the difference on 4K and 1080 monitors. I can see the difference in my phone screen.

A 40MP sensor offers benefits here because downsampling from 40MP to the display device's MP reduces noise and improves overall picture IQ. This is true.

Downscaling isn't necessary.

It's photos show slightly more detail than shown in 26MP photos. It's not a game changing difference but it is noticeable.

I agree. At the pixel level, we have an ~f/8 boundary with the 26MP APS-C sensor, where diffraction starts to interfere, and with 40MP, it starts earlier. It doesn't matter for real-world viewing on real-world devices with human (not alien) eyes. We are viewing downscaled images on our real-world devices, anyway. Do you agree?

Downscaling the 40MP image isn't necessary to do a fair comparison. Upscaling the smaller image, as fine in the DPR Studio Scene, is also an option.

Whichever lenses you choose, my advice is to set diffraction aside as a concern.

I second this.

 

[Erik Baumgartner]

An XH2 needs good light to give the best results.

All cameras benefit from working with good light.

Using the XH2 with a long lens at apertures of F8 and smaller may lead to diffraction and noisy images that even when cropped my not give you the best results, largely due to a poor signal to noise ratio.

The above suggests you may be making a couple of incorrect assumptions.

The first is that smaller pixels reduce a sensors ability to detect and process light. However, it's not pixel size that determines the light-collection ability of a sensor. It's sensor surface area that does. Since the H2 and H2S have same size sensors, both collect the same total light when working with the same exposure, and their photos display the same amount of shot noise.

Folks who complain that the H2 is noticeably noisier than the H2S are usually responding to a comparison of 100% views. A 100% view of a 40MP photo will be zoomed in farther than a 100% view of a 26MP photo. That increased magnification makes noise more prominent.

A comparison of photos displayed at the same size makes the read noise differences between those sensors quite subtle. It's not an issue: https://www.dpreview.com/reviews/im...7&x=0.5757840228378028&y=-0.19987511201612654

The second incorrect assumption possibly being made is that higher megapixel sensors are more susceptible to diffraction. This notion also comes from viewing and comparing photos at 100%. The more magnification used when studying an image, the less sharp fine details will look.

However, when photos are compared at the same size, the increased resolution of the H2 (and other 40MP sensors) is on display.

No, it really isn’t. Now you’ve thrown away 14MP of the 40MP sensor's resolution advantage.

Downscaled, the resolution is now the same. Only 26MP of resolution is on display. However, the finer rendering of detail that comes with more original image data to work with and fewer resulting demosaicing errors is on display. It still looks better despite there being no longer being any resolution advantage.

It's photos show slightly more detail than shown in 26MP photos. It's not a game changing difference but it is noticeable.

If this is generally accurate, and I am not saying it is exactly so, when purchasing a lens whether prime or zoom would it not be to one's advantage (again generally speaking) to use the fastest lens that meets your use case? I am contemplating purchasing a couple more lenses to use on my Xh2 and I want to get the very best results in terms of image quality. I am contemplating lenses such as the XF 33/1.4 and the Sigma 17-40/1.8. My use cases do not include long telephoto applications or BIF. Would be interested in your thoughts.

It would be easy to encourage you to buy the most expensive, highest quality lenses you can get. That's a safe answer. I'll advise that your lens choice depends on the kind of photography you enjoy doing. What subjects? Do you prefer deep depths of field or shallow? Do you work in low light or bright environments? With or without a tripod?

Whichever lenses you choose, my advice is to set diffraction aside as a concern. It's not going to be an issue unless you're in the f/11 and higher f-stop range. All APS-C systems are near the threshold of where diffraction becomes an issue in that range.

Good luck.

--
Bill Ferris Photography
https://billferrisphotography.pixieset.com/arizonaslittleserengeti/

 

[Bill Ferris]

An XH2 needs good light to give the best results.

All cameras benefit from working with good light.

Using the XH2 with a long lens at apertures of F8 and smaller may lead to diffraction and noisy images that even when cropped my not give you the best results, largely due to a poor signal to noise ratio.

The above suggests you may be making a couple of incorrect assumptions.

The first is that smaller pixels reduce a sensors ability to detect and process light. However, it's not pixel size that determines the light-collection ability of a sensor. It's sensor surface area that does. Since the H2 and H2S have same size sensors, both collect the same total light when working with the same exposure, and their photos display the same amount of shot noise.

Folks who complain that the H2 is noticeably noisier than the H2S are usually responding to a comparison of 100% views. A 100% view of a 40MP photo will be zoomed in farther than a 100% view of a 26MP photo. That increased magnification makes noise more prominent.

A comparison of photos displayed at the same size makes the read noise differences between those sensors quite subtle. It's not an issue: https://www.dpreview.com/reviews/im...7&x=0.5757840228378028&y=-0.19987511201612654

The second incorrect assumption possibly being made is that higher megapixel sensors are more susceptible to diffraction. This notion also comes from viewing and comparing photos at 100%. The more magnification used when studying an image, the less sharp fine details will look.

However, when photos are compared at the same size, the increased resolution of the H2 (and other 40MP sensors) is on display.

No, it really isn’t. Now you’ve thrown away 14MP of the 40MP sensor's resolution advantage.

Perhaps, you'd care to explain how it is that we can see the resolution advantage when comparing same-sized images in the DPR Studio Scene.

Downscaled, the resolution is now the same.

You're making an unnecessary assumption. The smaller image can be upscaled so both images are viewed at the same size. That's what the Studio Scene does.

Even if the H2 image is downscaled, the difference is visible. We've had this exchange before and I've posted the photos demonstrating this.

Only 26MP of resolution is on display. However, the finer rendering of detail that comes with more original image data to work with and fewer resulting demosaicing errors is on display. It still looks better despite there being no longer being any resolution advantage.

If the H2 image looks sharper, it's sharper.

It's photos show slightly more detail than shown in 26MP photos. It's not a game changing difference but it is noticeable.

If this is generally accurate, and I am not saying it is exactly so, when purchasing a lens whether prime or zoom would it not be to one's advantage (again generally speaking) to use the fastest lens that meets your use case? I am contemplating purchasing a couple more lenses to use on my Xh2 and I want to get the very best results in terms of image quality. I am contemplating lenses such as the XF 33/1.4 and the Sigma 17-40/1.8. My use cases do not include long telephoto applications or BIF. Would be interested in your thoughts.

It would be easy to encourage you to buy the most expensive, highest quality lenses you can get. That's a safe answer. I'll advise that your lens choice depends on the kind of photography you enjoy doing. What subjects? Do you prefer deep depths of field or shallow? Do you work in low light or bright environments? With or without a tripod?

Whichever lenses you choose, my advice is to set diffraction aside as a concern. It's not going to be an issue unless you're in the f/11 and higher f-stop range. All APS-C systems are near the threshold of where diffraction becomes an issue in that range.

Good luck.

 

[Erik Baumgartner]

An XH2 needs good light to give the best results.

All cameras benefit from working with good light.

Using the XH2 with a long lens at apertures of F8 and smaller may lead to diffraction and noisy images that even when cropped my not give you the best results, largely due to a poor signal to noise ratio.

The above suggests you may be making a couple of incorrect assumptions.

The first is that smaller pixels reduce a sensors ability to detect and process light. However, it's not pixel size that determines the light-collection ability of a sensor. It's sensor surface area that does. Since the H2 and H2S have same size sensors, both collect the same total light when working with the same exposure, and their photos display the same amount of shot noise.

Folks who complain that the H2 is noticeably noisier than the H2S are usually responding to a comparison of 100% views. A 100% view of a 40MP photo will be zoomed in farther than a 100% view of a 26MP photo. That increased magnification makes noise more prominent.

A comparison of photos displayed at the same size makes the read noise differences between those sensors quite subtle. It's not an issue: https://www.dpreview.com/reviews/im...7&x=0.5757840228378028&y=-0.19987511201612654

The second incorrect assumption possibly being made is that higher megapixel sensors are more susceptible to diffraction. This notion also comes from viewing and comparing photos at 100%. The more magnification used when studying an image, the less sharp fine details will look.

However, when photos are compared at the same size, the increased resolution of the H2 (and other 40MP sensors) is on display.

No, it really isn’t. Now you’ve thrown away 14MP of the 40MP sensor's resolution advantage.

Perhaps, you'd care to explain how it is that we can see the resolution advantage when comparing same-sized images in the DPR Studio Scene.

As I’ve already explained, you aren’t really seeing a resolution advantage (you cannot see 40MP of resolution at 26MP), you’re seeing what detail is being resolved detail being rendered more faithfully.

Downscaled, the resolution is now the same.

You're making an unnecessary assumption. The smaller image can be upscaled so both images are viewed at the same size. That's what the Studio Scene does.

No, that doesn’t appear to be the case. When the COMP button is applied, the higher resolution image is reduced to the size of the smaller image, not the other way around.

Even if the H2 image is downscaled, the difference is visible. We've had this exchange before and I've posted the photos demonstrating this.

Yes, the difference is visible, but if you could view them both at the same size and at full resolution the difference would be greater still - but this not possible on a single screen.

Only 26MP of resolution is on display. However, the finer rendering of detail that comes with more original image data to work with and fewer resulting demosaicing errors is on display. It still looks better despite there being no longer being any resolution advantage.

If the H2 image looks sharper, it's sharper.

”Sharper”, or even “better” doesn’t necessarily mean the resolution is greater. Two images from the same camera can look significantly sharper or softer/better or worse depending on the processing etc. A well processed 40MP image will always have better rendered fine detail (greater pixel density = fewer errors and artifacts in the demosaicing process). That more faithfully rendered detail typically translates to a better looking image when downscaled to 26MP than you’d get with a 26MP sensor as there are errors and artifacts that are unavoidable when there is less data for the processing to work with (more guesswork, more errors), especially with an X-Trans sensor. So yes, 40MP looks better downscaled to 26MP than a native 26MP sensor, but you’ve still lost 14MP of resolution potential that can never be displayed at 26MP. That difference might not equate to much (if anything) with a mediocre or excessively stopped down lens (or just poor processing), but a lens capable of resolving 40MP will definitely be losing image quality potential when limited to only 26MP, and that goes for the DPR test scene in “COMP” mode too

It's photos show slightly more detail than shown in 26MP photos. It's not a game changing difference but it is noticeable.

If this is generally accurate, and I am not saying it is exactly so, when purchasing a lens whether prime or zoom would it not be to one's advantage (again generally speaking) to use the fastest lens that meets your use case? I am contemplating purchasing a couple more lenses to use on my Xh2 and I want to get the very best results in terms of image quality. I am contemplating lenses such as the XF 33/1.4 and the Sigma 17-40/1.8. My use cases do not include long telephoto applications or BIF. Would be interested in your thoughts.

It would be easy to encourage you to buy the most expensive, highest quality lenses you can get. That's a safe answer. I'll advise that your lens choice depends on the kind of photography you enjoy doing. What subjects? Do you prefer deep depths of field or shallow? Do you work in low light or bright environments? With or without a tripod?

Whichever lenses you choose, my advice is to set diffraction aside as a concern. It's not going to be an issue unless you're in the f/11 and higher f-stop range. All APS-C systems are near the threshold of where diffraction becomes an issue in that range.

Good luck.

--
Bill Ferris Photography
https://billferrisphotography.pixieset.com/arizonaslittleserengeti/

 

[MrAndreas]

However, when photos are compared at the same size,

... downscaled for viewing on your display device...

Or upscale. The DPR Studio Scene upscales photos made with smaller format cameras.

This case is so niche and unique. And DPR test chart photos are downscaled to the user device resolution for viewing, anyway.

the increased resolution of the H2 (and other 40MP sensors) is on display.

No. Your display device is rarely an 8 K display.

An 8K display isn't necessary. I see the difference on 4K and 1080 monitors. I can see the difference in my phone screen.

A 40MP sensor offers benefits here because downsampling from 40MP to the display device's MP reduces noise and improves overall picture IQ. This is true.

Downscaling isn't necessary.

How do you view the 40MP image rather than downscaled to the display device (or print) resolution, I wonder.

It's photos show slightly more detail than shown in 26MP photos. It's not a game changing difference but it is noticeable.

I agree. At the pixel level, we have an ~f/8 boundary with the 26MP APS-C sensor, where diffraction starts to interfere, and with 40MP, it starts earlier. It doesn't matter for real-world viewing on real-world devices with human (not alien) eyes. We are viewing downscaled images on our real-world devices, anyway. Do you agree?

Downscaling the 40MP image isn't necessary to do a fair comparison. Upscaling the smaller image, as fine in the DPR Studio Scene, is also an option.

... and view it downscaled, anyway.

Whichever lenses you choose, my advice is to set diffraction aside as a concern.

I second this.

 

[KariP]

Interesting and long discussion - perhaps centered in pixel peeping .

I know it is possible to see what diffraction can do and it is possible to compare the amount of noise 26MP versus 40MP. In 100-200% view on a 5K screen

This discussion is just about the invisible world ( or something we see only if we look things with a "microscope" )

I have photographed with various digital cameras during the last 20 years. After 2007 and my first DSLR (Canon 40D) the image quality in practice - and in the visible world has been surprisingly good. X-E1 gave more dynamic range. The differences in visible image quality are hard to see especially when I have looked at big prints. And there is also editing and skillful printing and better papers - as part of the things creating visible images. Who sees the diffraction ?

My favourite cameras at the moment are Canon R5 and Fuji X-E3 .... IMO both have really enough of MP, but if I'm looking back at the collection of my greatest and best images of all times ( not so many) ... Better cameras/lenses give a slight visible advantage in some cases. BUT ! we all should look at our photographs and think what makes our images better. And will they get better if I by a new camera with more resolution and more pixels and and a f 1,4 lens instead of f 2 and so on. What makes me to make better pictures ?

 

[MrAndreas]

...

However, when photos are compared at the same size, the increased resolution of the H2 (and other 40MP sensors) is on display.

No, it really isn’t. Now you’ve thrown away 14MP of the 40MP sensor's resolution advantage.

Downscaled, the resolution is now the same. Only 26MP of resolution is on display. However, the finer rendering of detail that comes with more original image data to work with and fewer resulting demosaicing errors is on display.

Bravo, you nailed it perfectly. Thank you for the clarity of your formulation.

A remark on demosaicing. Even the in-camera demosaicing part of the imaging pipeline is good enough, despite being constrained by real-time processing and available computing resources (so it trades off the algorithm sophistication and result quality for speed and effectiveness). I can't recall any artifacts in my Fuji SOOC JPEGs that I'd relate specifically to demosaicing.

The only case of wrong demosaicing I recall is the infamous "X-Trans worms" epic, which turned out to be specific to Adobe products and to Adobe's malicious negligence in the demosaicing algorithm choice and under-the-hood implementation, superposed by the dubious USM nonswitchable default (over)sharpening.

Free darktable software has supported a choice of X-Trans demosaicing algorithms since its early days (now at version 5.2); it never performs unsolicited sharpening, and thanks to this, I have never had any problems with RAFs. Even more, because desktop computers have "unlimited" computational resources, it is easy to demosaic the RAF at a quality level far beyond the in-camera CPU's capabilities. Doing so is my standard practice.

It still looks better despite there being no longer being any resolution advantage.

Yes, exactly. Stronger downsampling performs well here.

BTW, this reminds me of the smartphone-specific wiles and tricks that are common nowadays. Let's take the Samsung S23 as an example. It uses Samsung’s 50MP ISOCELL GN3 sensor (~10.19 mm diagonal, 4:3 aspect, ~4.2 crop, and a 1.00 μm pixel size), for ~1.0 megapixels/mm². Crazy dense, yeah?

Apple iPhone 14/15 Pro is a bit less extreme: 48MP, 1.22 μm pixel size, ~0.66 megapixels/mm², ~3.5 crop.

But have you ever seen the 50MP RAW file from S23? I did, out of curiosity. It looks completely horrible, even if taken in good light. Postprocessing this mess by hand is far beyond my own skills and desire.

So what did these smart guys do? They both implemented 4-to-1 pixel binning (a 2x2 quad maps to 1 pixel in the resulting image) in hardware under different trade names (“Quad” / Tetrapixel binning). Add some supplementary computational imaging tricks, such as invisibly merging multiple exposures (or multiple gains), and we effectively get a 12MP image that is quite good, and 12-bit, AFAIK.

For comparison, the Fuji 40MP sensor is ≈0.11 MP/mm² only, so Fujifilm has ample room for future improvements on its way to approaching the smartphone market with its cameras. Imagine a 360MP physical APS-C sensor, made by Samsung with all its underlying smart tech, tetrabinned to 90MP pseudo-RAW? That's what will make a real difference for the 4k or 8k screen viewing.

--
https://www.viewbug.com/member/stesinou

 

[silversportsman]

...

I personally really enjoy fast lenses and the bokeh they produce as they can make your photos look a lot nicer.

Morris

As do I. Maybe that is one of my underlying motivating factors.

 

[silversportsman]

An XH2 needs good light to give the best results.

All cameras benefit from working with good light.

Good point.

Using the XH2 with a long lens at apertures of F8 and smaller may lead to diffraction and noisy images that even when cropped my not give you the best results, largely due to a poor signal to noise ratio.

The above suggests you may be making a couple of incorrect assumptions.

The first is that smaller pixels reduce a sensors ability to detect and process light. However, it's not pixel size that determines the light-collection ability of a sensor. It's sensor surface area that does. Since the H2 and H2S have same size sensors, both collect the same total light when working with the same exposure, and their photos display the same amount of shot noise.

Folks who complain that the H2 is noticeably noisier than the H2S are usually responding to a comparison of 100% views. A 100% view of a 40MP photo will be zoomed in farther than a 100% view of a 26MP photo. That increased magnification makes noise more prominent.

That makes sense.

A comparison of photos displayed at the same size makes the read noise differences between those sensors quite subtle. It's not an issue: https://www.dpreview.com/reviews/im...7&x=0.5757840228378028&y=-0.19987511201612654

The second incorrect assumption possibly being made is that higher megapixel sensors are more susceptible to diffraction.

This assumption probably largely came from comments I have read rather than real world experience. I am glad you have called out this assumption and set me straight.

This notion also comes from viewing and comparing photos at 100%. The more magnification used when studying an image, the less sharp fine details will look.

However, when photos are compared at the same size, the increased resolution of the H2 (and other 40MP sensors) is on display. It's photos show slightly more detail than shown in 26MP photos. It's not a game changing difference but it is noticeable.

If this is generally accurate, and I am not saying it is exactly so, when purchasing a lens whether prime or zoom would it not be to one's advantage (again generally speaking) to use the fastest lens that meets your use case? I am contemplating purchasing a couple more lenses to use on my Xh2 and I want to get the very best results in terms of image quality. I am contemplating lenses such as the XF 33/1.4 and the Sigma 17-40/1.8. My use cases do not include long telephoto applications or BIF. Would be interested in your thoughts.

It would be easy to encourage you to buy the most expensive, highest quality lenses you can get. That's a safe answer. I'll advise that your lens choice depends on the kind of photography you enjoy doing. What subjects? Do you prefer deep depths of field or shallow? Do you work in low light or bright environments? With or without a tripod?

I am a huge fan of shallow depths of field and do often find myself in low light without a tripod.

Whichever lenses you choose, my advice is to set diffraction aside as a concern. It's not going to be an issue unless you're in the f/11 and higher f-stop range. All APS-C systems are near the threshold of where diffraction becomes an issue in that range.

Good luck.

 

[silversportsman]

Image noise is really about the same for both sensors. The 40MP sensor will look a bit noisier at 100% on your computer screen (pixel noise), but only because you’re looking closer at it. Diffraction will never look worse at 40 vs. 26MP, but it can limit the maximum resolution available from the 40MP sensor, but no more than what you’ll see at 26MP. With more image data to work with, A.I. noise reduction is typically more effective at 40MP as it can better distinguish between image detail and noise for a sharper/cleaner result. You will get a bit more out of the best primes with the 40MP sensor if you are careful to optimize your shooting discipline.

Thanks Erik. Could you summarize your definition of optimizing shooting discipline?

 

[silversportsman]

However, when photos are compared at the same size,

... downscaled for viewing on your display device...

the increased resolution of the H2 (and other 40MP sensors) is on display.

No. Your display device is rarely an 8 K display. I'm not aware of any display device capable of reproducing more than "8k" (that's ~33MP). But it is at a pixel level. At a human viewing level (not aliens), from a realistic viewing distance, we are bound to be around 8MP - 12MP perceivable resolution.

A 40MP sensor offers benefits here because downsampling from 40MP to the display device's MP reduces noise and improves overall picture IQ. This is true.

It's photos show slightly more detail than shown in 26MP photos. It's not a game changing difference but it is noticeable.

I agree. At the pixel level, we have an ~f/8 boundary with the 26MP APS-C sensor, where diffraction starts to interfere, and with 40MP, it starts earlier. It doesn't matter for real-world viewing on real-world devices with human (not alien) eyes. We are viewing downscaled images on our real-world devices, anyway. Do you agree?

Whichever lenses you choose, my advice is to set diffraction aside as a concern.

I second this.

After reading through the replies, I agree as well.

 

[silversportsman]

Using the XH2 with a long lens at apertures of F8 and smaller may lead to diffraction and noisy images that even when cropped my not give you the best results, largely due to a poor signal to noise ratio. If this is generally accurate, and I am not saying it is exactly so, when purchasing a lens whether prime or zoom would it not be to one's advantage (again generally speaking) to use the fastest lens that meets your use case?

Not necessarily. People often treat diffraction as some magic aperture value where an image is unusable. When it's really a slow, generally imperceptible decline in sharpness.

I've read comments from folks saying they never go past f/4-5.6, etc, for maximum sharpness. That's absurd. Because one, it's not that big a deal. And two, depth of field gains will often increase perceived resolution, despite diffraction losses. Meaning, if I shoot a telephoto landscape at f/4 to avoid diffraction, the image foreground may still look soft because sections are out of focus. If I use f/8-11 instead, it will look better, despite hitting diffraction territory.

Diffraction isn't that big a deal; use whatever aperture value best fits your creative intention, knowing it comes at the cost of max resolution. And compensate in post with added sharpness, if needed. Capture One has a Diffraction Correction tool that's very handy to apply at the start.

Thanks, Earl. Good advice, and I am a C1 user but have not used that tool. I will check it out.

 

[Erik Baumgartner]

...

However, when photos are compared at the same size, the increased resolution of the H2 (and other 40MP sensors) is on display.

No, it really isn’t. Now you’ve thrown away 14MP of the 40MP sensor's resolution advantage.

Downscaled, the resolution is now the same. Only 26MP of resolution is on display. However, the finer rendering of detail that comes with more original image data to work with and fewer resulting demosaicing errors is on display.

Bravo, you nailed it perfectly. Thank you for the clarity of your formulation.

A remark on demosaicing. Even the in-camera demosaicing part of the imaging pipeline is good enough, despite being constrained by real-time processing and available computing resources (so it trades off the algorithm sophistication and result quality for speed and effectiveness). I can't recall any artifacts in my Fuji SOOC JPEGs that I'd relate specifically to demosaicing.

If the RAW processing pipeline includes a color filter array (CFA), there has to be interpolation (guesswork) in reconstructing the image. The more image data there is to work with, the fewer errors there will be. Those errors are typically so small that they aren’t immediately apparent, but upon closer inspection various errors and artifacts become obvious - both in color and detail.

Below is a comparison of the 16MP X-T1 sensor vs. the (downscaled to 16MP) X-T5 sensor and then both viewed at 200% to exaggerate the situation - all the obvious false detail/artifacts that can easily be seen in the X-T1 image text is due to demosaicing errors - processing guesswork that failed. With 40MP of image detail to work with, the X-T5 image contains fewer errors, and while there is a loss of fine detail (resolution) when downscaled, the more faithful detail rendering is still retained, producing a better looking image at 16MP despite its resolution being significantly reduced. Some will say that, hey, you’re pixel peeping to see those errors, and they would be right, but it’s those low level errors that get exaggerated with sloppy sharpening …“worms” and ugly foliage especially, that can easily be noticed at normal viewing sizes, and that are largely absent with the 40MP sensor - even with Adobe’s less than stellar standard X-Trans demosaicing.