Theoretical and practical limits of high megapixel cameras.

[Cryve]

Being a high resolution senor enjoyer myself, im wondering where the future might take us.
Will the pixel counts of modern sensors just increase generation to generation forever?

I believe they wont. When i started to photograph, many people were longing for more megapixels, but nowadays the calls for more megapixels got quiter and most people
seem to be content with whats available on the market.

Some people do still want more megapixels tho. And here im asking, how far will it go?

Of course there must be some limits to megapixelcount i imagine. Here some, that i can think of:

* Too limiting for shutterspeeds - Shots of moving subjects will need too fast of a shutter speed to be practial
* Limited by diffraction. Some lenses, like f6.3, will become limited by diffraction very soon, at like 76 megapixels.
* Limited by practicability - who needs it? at what point will there be no more market, because noone prints that big?
* Limited by lenses - is there some point, where it becomes too expensive to design lenses to resolve those high megapixel sensors (even right now most lenses dont resolve the full potential of today 45 - 61 megapixel sensors)?
* Too limited by iso, at some point you will only see all megapixels by iso 100 and anything above will result in not getting the full benefit of the pixel count

Perhaps you can think of more reasons.

What do u think? What will be the most limiting factors and where do you think we will end up, pixel count wise, in the forseeable future?

 

[Cryve]

Some lenses that are coming out, like the new 200-800 f9 lens seem to be limited by difraction already, at about 37 megapixels Full frame, if the calculations of this website are correct:

Diffraction Calculator | PhotoPills

This diffraction calculator will help you assess when the camera is diffraction limited.

www.photopills.com

The beloved 100-500 f7.1 would become limited by 60 megapixels.

 

[neuroanatomist]

It's important to realize that the diffraction 'limit' isn't like a cliff where your image quality falls off and plummets to the (not) sharp rocks below, leaving leaving your precious pictures as broken, tattered fragments of shattered pixels.

 

[Cryve]

It's important to realize that the diffraction 'limit' isn't like a cliff where your image quality falls off and plummets to the (not) sharp rocks below, leaving leaving your precious pictures as broken, tattered fragments of shattered pixels.

Thats an important point of course, its very subtle and you still benefit from more pixels, but the returns of those pixels are diminishing the higher you surpass the difraction limit.

Im personally hoping for the industry to go to around 80-100 megapixels, but i have my doubts of course. maybe we wont go higher than what we are seeing currently. at least the sports bodys seem to have found their sweetspot at 24 mp for ages now and i presume we are in phase thats about finding the sweetspot for higher mp cameras currently, with most manufacturers going for the 45-50 mark.

Canon even took a step back from their 5ds r (50 megapixels) to now not offering anything above 45.

 

[AlanF]

Being a high resolution senor enjoyer myself, im wondering where the future might take us.
Will the pixel counts of modern sensors just increase generation to generation forever?

I believe they wont. When i started to photograph, many people were longing for more megapixels, but nowadays the calls for more megapixels got quiter and most people
seem to be content with whats available on the market.

Some people do still want more megapixels tho. And here im asking, how far will it go?

Of course there must be some limits to megapixelcount i imagine. Here some, that i can think of:

* Too limiting for shutterspeeds - Shots of moving subjects will need too fast of a shutter speed to be practial
* Limited by diffraction. Some lenses, like f6.3, will become limited by diffraction very soon, at like 76 megapixels.
* Limited by practicability - who needs it? at what point will there be no more market, because noone prints that big?
* Limited by lenses - is there some point, where it becomes too expensive to design lenses to resolve those high megapixel sensors (even right now most lenses dont resolve the full potential of today 45 - 61 megapixel sensors)?
* Too limited by iso, at some point you will only see all megapixels by iso 100 and anything above will result in not getting the full benefit of the pixel count

Perhaps you can think of more reasons.

What do u think? What will be the most limiting factors and where do you think we will end up, pixel count wise, in the forseeable future?

We have discussed this many, many times over the years here. When viewed at the same size, DR, S/N movement across the sensor etc do not materially change with decreasing pixel size. In theory, the diffraction limit can be overcome if you know the point spread function. For many, it's not just a question of printing larger but being able crop more.

 

[neuroanatomist]

Canon even took a step back from their 5ds r (50 megapixels) to now not offering anything above 45.

Megapixel count does not tell the whole story. Canon changed the design of their OLPF / AA filter in ~2020 (the new design rolled out with the 1D X III). As a result of that new design, the 24 MP sensor in the R3 outresolves the 30 MP sensor in the 5DIV / EOS R, and the 45 MP sensor in the R5 outresolves the 50 MP self-cancelling AA filter sensor in the 5DsR.

 

[SereneSpeed]

I have a question. My understanding of diffraction is very limited. If I use my R5 in crop mode, will my image have the same diffraction as FF mode, for the same given area of the image.

Asked differently (and hopefully asking the same question) is it solely the pixel density that effects diffraction? Or, does the location of those pixels (closer to the center of the image circle projected by the lens) matter? (Like It does for vignetting).

Sorry if this is way off topic.

 

[neuroanatomist]

…is it solely the pixel density that effects diffraction?

For the sensor, yes. Aperture of the lens affects diffraction as well, of course.

 

[DhlcadR6]

RF 600/800mm f11: So...are we the baddies now?

 

[EricN]

RF 600/800mm f11: So...are we the baddies now?

Only for people who don't understand how things work

 

[Cryve]

Only for people who don't understand how things work

Im not sure what you are adressing here, but if we are talking about diffraction, then the f11 lenses 100% have image degradation because of diffraction with our modnern 45 megapixel cameras.

 

[neuroanatomist]

Im not sure what you are adressing here, but if we are talking about diffraction, then the f11 lenses 100% have image degradation because of diffraction with our modnern 45 megapixel cameras.

When you compare the RF 800/5.6L IS at f/5.6 vs. f/11, it's sharper at f/11 despite '100% having image degradation at f/11'. Most lenses benefit optically from being stopped down a bit from wide open, and if you compare the lens at f/6.3 vs. f/11 the image quality is essentially the same. Comparing f/8 vs. f/11, f/8 is slightly better because the benefit of stopping down a bit outweighs the detriment of the diffraction that starts at the R5's DLA of f/7.1. The 800/11 is not as sharp as the RF 800/5.6L at f/11, but that's not diffraction (and yes, I know it's not really fair to compare a $1000 lens to a $17000 lens...but honestly the 800/11 performs far better than the 17x price differential suggests!).

So while it's technically true that 'f11 lenses 100% have image degradation because of diffraction with our modnern 45 megapixel cameras', the magnitude of that image degradation is relatively small at typically used apertures...small enough that other factors are far more important in determining image quality. I suspect that was the point @EricN was making. Sure, if you stop a lens down to f/22 or f/64 you're going to see a significant effect of diffraction. But at f/11 it's not really meaningful, even with a 45 MP sensor.

 

[Act444]

With Canon not shying away from making lenses slower than f5.6 in the MILC era, IMHO that emphasizes two areas of sensor performance: high ISO performance and high ISO detail retention. Interestingly enough, when using the DPR comparison tool to compare the R5 45MP vs the 5DSR 50MP, at lower ISOs (below 800), I'd still give the 5DSR the edge in detail resolution. However, above ISO 800, the R5 pulls ahead. Similar when comparing the R6 II 24MP to the R 30MP...the R looks crisper to my eye up to 800 - but then once you get to 1600, the R6 II starts to do a better job of retaining finer details (especially color details) despite having less resolution. While this can still be useful if you shoot mostly at f2.8 or faster, when you've got an f8 or f11 setup (or perhaps you just want more DOF), it becomes essential because even in bright daylight, my experience has been that "ISO creep" can easily occur at these apertures, especially once you get under any sort of cover or shade. I've had ISO shots as high as 4,000 shooting at f8 (for DOF purposes) in these conditions!

All this to say, I think the foundation for this may already be a work in progress, with newer sensors potentially resolving fine details better at higher ISOs, and more computational image "correction" technologies (which MILC allows) to attempt to overcome this optical limitation. Who knows?

 

[neuroanatomist]

All this to say, I think the foundation for this may already be a work in progress, with newer sensors potentially resolving fine details better at higher ISOs, and more computational image "correction" technologies (which MILC allows) to attempt to overcome this optical limitation. Who knows?

That's an excellent point. Canon makes that pretty clear, since in their touting of the higher spatial resolutions available with recent cameras, they always mention Digic X (and sometimes RF lenses).

 

[EricN]

When you compare the RF 800/5.6L IS at f/5.6 vs. f/11, it's sharper at f/11 despite '100% having image degradation at f/11'. Most lenses benefit optically from being stopped down a bit from wide open, and if you compare the lens at f/6.3 vs. f/11 the image quality is essentially the same. Comparing f/8 vs. f/11, f/8 is slightly better because the benefit of stopping down a bit outweighs the detriment of the diffraction that starts at the R5's DLA of f/7.1. The 800/11 is not as sharp as the RF 800/5.6L at f/11, but that's not diffraction (and yes, I know it's not really fair to compare a $1000 lens to a $17000 lens...but honestly the 800/11 performs far better than the 17x price differential suggests!).

So while it's technically true that 'f11 lenses 100% have image degradation because of diffraction with our modnern 45 megapixel cameras', the magnitude of that image degradation is relatively small at typically used apertures...small enough that other factors are far more important in determining image quality. I suspect that was the point @EricN was making. Sure, if you stop a lens down to f/22 or f/64 you're going to see a significant effect of diffraction. But at f/11 it's not really meaningful, even with a 45 MP sensor.

Thanks for taking the time to explain all that.

Something I want to add is (and I don't recall if it was stated by Canon) F/7.1 (as in RF100-500mm F4.5-7.1 L IS USM and RF 24–105mm f/4–7.1 IS STM) is the point where diffraction should begin on the R5, while F/11 is the point where (in my and at least some others' estimation) most people will notice the effect of diffraction become visible for average sizes/viewing distances. We can guess the sales data from those four lenses all being introduced in 2020 helped them make the final decisions for RF 100–400mm f/5.6–8 IS USM soon after in the following year and then even more factors regarding the RF 200–800mm f/6.3–9 S USM this year. For me, it's exciting to see what will happen with regards to balancing Image quality, build quality, low price, weight, and size will be solved by (I believe) keeping apertures between F/7.1 and F/11 in these lenses. Yes, I admit I can be wrong and they can make something even smaller than 11.

 

[Quack]

From many discussions regarding diffraction and reading other photographer's experiences, I have always considered diffraction to start being noticeable 2 stops above the actual theoretical start of diffraction. So in practical terms, add 2 or even 3 stops to the theoretical start of diffraction to have a reasonable idea as to diffraction's effects on your photography.
That being said, it will still limit the effectiveness of higher MP sensors at some point. And my guess is that some camera makers will avoid that fact and continue to go higher with MP counts even to the point where IQ decreases, as long as those seeking "more" and "better" will buy such a camera with "bigger number" specs.

 

[Cryve]

When you compare the RF 800/5.6L IS at f/5.6 vs. f/11, it's sharper at f/11 despite '100% having image degradation at f/11'. Most lenses benefit optically from being stopped down a bit from wide open, and if you compare the lens at f/6.3 vs. f/11 the image quality is essentially the same. Comparing f/8 vs. f/11, f/8 is slightly better because the benefit of stopping down a bit outweighs the detriment of the diffraction that starts at the R5's DLA of f/7.1. The 800/11 is not as sharp as the RF 800/5.6L at f/11, but that's not diffraction (and yes, I know it's not really fair to compare a $1000 lens to a $17000 lens...but honestly the 800/11 performs far better than the 17x price differential suggests!).

So while it's technically true that 'f11 lenses 100% have image degradation because of diffraction with our modnern 45 megapixel cameras', the magnitude of that image degradation is relatively small at typically used apertures...small enough that other factors are far more important in determining image quality. I suspect that was the point @EricN was making. Sure, if you stop a lens down to f/22 or f/64 you're going to see a significant effect of diffraction. But at f/11 it's not really meaningful, even with a 45 MP sensor.

i honestly thought difraction would have way more of an impact. thanks for clearing that up.

 

[zarathu]

Sounds like a theoretical discussion. In practice printing to 13 x 19 using editing software, I’ve only run into it when taking macro photos in a small stack.

I use the M1Pro Scottish laptop with only 16 gb ram. If I had a camera with 80-100 mp, I would need to sell my current model and get one with a maxed out Ram of 120 GB for $1500 more than my current model.

 

[vikingar]

For those interested in optics the channel "Huygens Optics" has a very technical yet approachable video "How a Lens Creates an Image" on Youtube:

The very cool animation at 4:45 visually shows the effects of "diffraction" between aperture sizes, and right after that there's a short part on critical dimension / smallest resolvable feature size of a lens. The rest of the video goes into more details and has some neat real-world images from a microscope and various experiments.

Note that the channel is not focused on photography, but on optics in general. But Huygens Optics has some great videos on topics sich as lens grinding and optical testing of lenses, and they're all very approachable while being technically accurate.

 

[Cryve]

With more extremely resolving lenses becoming available now, i actually do think that difraction is a practically limiting factor.

Altough its true, that for most lenses stopping down the lens above the difraction limit will still improve sharpness.

But if we take a theoretical "perfectly sharp" lens, then we actually are limited by difraction. The r5 with 45 has a difraction limit of about 7.1. We can see this already with the extremly sharp RF 135mm f1.8 here: Hover with your mouse over the picture to see f8 and take the mouse back away from the picture to see f5.6. f8 is noticably softer, less contrasty.

Canon RF 135mm F1.8 L IS USM Lens Image Quality

View the image quality delivered by the Canon RF 135mm F1.8 L IS USM Lens using ISO 12233 Resolution Chart lab test results. Compare the image quality of this lens with other lenses.

www.the-digital-picture.com

So while difraction might not be relevant right now, because the lenses arent perfectly resoling the sensors, the 45 mp cameras of today are absolutly theoretically limited by lenses like the 100-500 f7.1, 200-800 f9 and 600 and 800 f11. It just isnt apparent because the lenses arent perfectly resolving the sensors, so stopping down still improves the image more, than diffraction degrades it.

This might mean, that there is not much reason to increase the megapixel count with the current market trend of preference for small and lightweight lenses, that need small aperatures in order to achieve this form factor for some of the lenses. I wouldnt be suprised if the r5 ii stays in the 45 mp range. Same for the sony r range, i dont think it will go much beyond 61mp.

This leads me to question, where camera makers will head next to? We already have really good ergonomic camera bodies. We have more than enough frames per second for most applications, we have a practically good amount of megapixels. I think the r5 ii will be a really big tell for where we are heading with the camera industry, as i think we wont see much more meaningful improvement for most people. I hope it will be a great camera, but im unsure what could meaningfully be improved, except for making it a stacked sensor, maybe a bit more dynamic range.