IQ: Does it matter any more?

So many of us... just seem to be programmed to want the greatest of everything, even if it's far beyond anything that would ever be useful to them. Resolution, detail and all of that in photography seem to be one of those things. Sure, a huge format offer a technically higher level of performance, but the question is whether it will be useful for the kind of photography that you do.

A friend of mine, who bought a FF Sonny camera against my recommendations (I said that he should start of with smaller, cheaper, more compact gear)
The one thing I think full-frame has going for it for a beginner is that with a reasonably fast lens a person can experiment with shallow depth of field, whether just to obscure a distracting background or for other creative purposes. In the 1970's, when I went from simple box cameras to something with a little more precision and control, a budget-priced SLR with a normal lens of f/1.7 to f/2 was just as capable of this as the more well-heeled users of Rolleiflex and Hasselblad medium format cameras got with their 80mm f/2.8 lenses.

I don't know what your friend bought, but a Sony a7 II with Sony FE 50mm f/1.8 lens goes for about $1100 now, and I could see that as a reasonable choice for a beginner. The Pentax SP1000 with 55mm f/2 lens I got in 1975 was typical of what beginner-level photography instructors wanted students to use in my area, and there is a lot to learn from a similar full-frame DSLR/normal prime combination today, with the bonus of its being hugely more capable for low light and large print sizes, with output quality suitable for some professional/commercial applications you wouldn't dream of attempting with 35mm film.
and decided that with the lenses he might want to use with it that the thing is too big and too expensive,
I'm curious as to what lenses he was wanting, perhaps what types of photography he wanted to pursue.
was talking to me about buying some dedicated video camera that he's read about that's got far better low light performance than the Sony he bought. The thing is that the guy is a newbie to photography and even more so to video... I guess that he just hates the idea of even putting the time into learning photography or video unless he's capturing it on the very best gear that he can get. All of this is of course silly, but it seems that enough folks think this way to sell lots of expensive photo gear. There's a thought process that's more about "what might conceivably be useful to me in the most extreme situation that I haven't yet run into?" rather than the more reasonable thought of "how good do I really need for the kind of thing that I'm most likely to be doing a lot of?
I like that question, "how good do I really need for the kind of thing that I'm most likely to be doing a lot of?" However, I do think it's more easily answered by someone who has been into photography for a while than it would be for a beginner. I think a case can be made for starting a beginner off with such diverse choices as a full-frame starter kit like I described above, a tiny sensor bridge camera like the FZ300 (which more than meets all of my needs), or something in the middle, like a four thirds or APS-C interchangeable lens camera. Depending on what a beginner ends up wanting to do, any of those choices could prove to have been just right, or something that needs to be moved away from.

Your post did remind me of something I read, and largely agree with, in an FZ300 review:

"Does the FZ300 make good enough pictures for you? If you are a professional photographer the answer will be no, as expected. But many amateur and enthusiast photographers buy more and sometimes far more camera capability than they need or can use effectively. The corporations which make cameras and the retailers which sell them . . . will tell you the FULL FRAME model makes better pictures which is true in the technical sense. But most buyers will never need or be able to utilise the full extent of that quality. And the FZ300 provides remarkable versatility in a single, all-in-one package with no need for extra lenses or any other additional equipment." http://cameraergonomics.blogspot.com/2020/01/panasonic-lumix-fz300-re-appraisal-19.html
Well... in the case of my friend who overbought the too, big, too expensive gear, though he certainly didn't know what was likely to be most useful to a beginner like himself, I did offer suggestions, but of course all of those were ignored. I think that he just made the foolish assumption that more expensive gear was going to somehow make him a decent photographer, with little to no effort and that playing with the thing would be pure fun. Once he got the thing (and I really didn't harp on him for ignoring my advice), I tried to explain some basic things to him about the effects of aperture, exposure and how he might deal with focusing on a subject that isn't in the middle of the frame (pretty important, I'd say)... but he seemed to lack the patience even to learn this basic stuff. The guy is decent musician and gaining that kind of knowledge took many years and lots of effort. I think that the difference is that he started with the music thing young and built on that over the years, whereas the photography thing is brand new to him and somehow age has given him far less patience to learn anything from the very begging.
That is a shame about not being able to get into enough photography basics to make better use of his Sony gear. I have a friend who wanted to move past what her FZ35 could do in terms of making better quality large prints while allowing for some cropping. She got a Nikon D3200, but keeping in mind how to set the basics of aperture, shutter speed, etc., to make the pictures turn out how she wanted was distracting her from actually getting the shots she wanted. With her TZ1 and FZ35 compacts, program mode gave her files that were quite workable without having to give camera operation much thought. She ended up shelving the Nikon and getting an FZ1000, which she's been very happy with these past five years.
My advice to a beginner (at least one who seems to be really interested in learning about the medium), would be to get a few year old, cheaper m43 camera with the kit zoom and some kind of faster prime... which isn't too much differnt than what I use (I use m43 gear exclusively). It's a compact kit, easy to carry, still offers very good IQ and is something that one can grow with. I worked briefly selling cameras some years ago and sold a number of Olympus EM-10 mkII cameras to beginner photographers and it's hard to know exactly how they fared, but I remember that a couple of them did come back to the shop and tell me that they were very happy with the gear that they bought.
I have to agree about Four Thirds. My most-used camera at present is the Olympus E-450 DSLR, and all I have is the little 14-42mm kit zoom it came with. That lens has been on the camera since I bought it in 2009, and it has met my simple needs quite well.
I think that the super-zooms sell as well as they do because users like the idea of having that kind of giant zoom. I'm sure that there are some very good photographers who use those cameras and utilize the whole zoom range but for a lot, maybe even most buyers its more about the novelty of having a camera that can go from a roomy wide angle shot to getting the top of the steeple of the church that they're visiting...
I bought the FZ300 when shopping for a replacement for my Pentax W20 weather-proof camera. The FZ300 doesn't have that level of water-proofing, but it holds up in a light rain, and that's all I really need from it. The huge zoom range is not something I've put to use yet (bought mine in 2018). About 95% of my pictures are taken at the 50mm equivalent setting, and I don't think I've gone past 90mm equivalent except for some testing in the backyard. When handling different options at the store, I just liked how the FZ300 felt in hand, it's very usable EVF, and just how easy and pleasant it is to operate. With its f/2.8 lens and effective image stabilization, it's my most capable low light camera; good enough, anyway, for static subjects in dim museums. If and when my E-450 dies, the FZ300 can very easily take its place (though I know I will miss the E-450's optical viewfinder!). Someday I might re-visit a wildlife park, and perhaps then I'll get some use out of the FZ300's telephoto range.
 
Last edited:
I bought the FZ300 when shopping for a replacement for my Pentax W20 weather-proof camera. The FZ300 doesn't have that level of water-proofing, but it holds up in a light rain, and that's all I really need from it. The huge zoom range is not something I've put to use yet (bought mine in 2018). About 95% of my pictures are taken at the 50mm equivalent setting, and I don't think I've gone past 90mm equivalent except for some testing in the backyard. When handling different options at the store, I just liked how the FZ300 felt in hand, it's very usable EVF, and just how easy and pleasant it is to operate. With its f/2.8 lens and effective image stabilization, it's my most capable low light camera; good enough, anyway, for static subjects in dim museums. If and when my E-450 dies, the FZ300 can very easily take its place (though I know I will miss the E-450's optical viewfinder!). Someday I might re-visit a wildlife park, and perhaps then I'll get some use out of the FZ300's telephoto range.
Another factor to consider is how well the latest NR/AI software works with older cameras. As I don't have an FZ300 these are 3 samples from Imaging Resources review { thankfully though the site is closed down they have kept their archive active }. DXO NR, cropped to my favorite 3:2 aspect ration and upscaled to an 18x12" 300 ppi in PS tinkered to taste :-) .

3200 ISO



787ac1307e7a4b339cf6eb23cdfaa5b4.jpg

1000 ISO



70ad57bca6dc44e79510d4bebe85edec.jpg

At its full zoom equiv to a 600mm FF AOV If you are in reasonable light you can get very decent results



aa2f8446510d42cfa426df443285bb40.jpg



--
Jim Stirling:
“It is one thing to show a man that he is in error, and another to put him in possession of truth.” Locke
Feel free to tinker with any photos I post
 
...

The 9yr gap has brought huge advances in the feature set and performance of cameras image quality has received much more modest advances.
They didn't receive and will never receive because there is nothing there left to receive. The read noise is now so low that the pixels can (almost) spot individual photons. 10 years ago the typical input referred read noise was about 5 electrons, now it is about 1.3 electrons. Even if you get read noise to zero, you will not see any improvement in a typical test or landscape shot. Have to test these sensors on deep space photography.
I am not expecting any great steps forward :-) It is just to point out to the folk who claim huge leaps in image quality that things are not quite as rosy as they may imagine . I suspect that the " typical test shot or landscape " is a lot closer to most folks use case than deep space photography
 
...

The 9yr gap has brought huge advances in the feature set and performance of cameras image quality has received much more modest advances.
They didn't receive and will never receive because there is nothing there left to receive. The read noise is now so low that the pixels can (almost) spot individual photons. 10 years ago the typical input referred read noise was about 5 electrons, now it is about 1.3 electrons. Even if you get read noise to zero, you will not see any improvement in a typical test or landscape shot. Have to test these sensors on deep space photography.
Modern RAW processing software is getting so good that even if one camera does have an advantage by the time it's processed the advantage is gone. IMO That's the flaw in comparing unprocessed RAW images.

--
Tom
 
Last edited:
Reading the comments one might conclude that for the majority it actually doesn`t matter since it`s difficult in other than the most controlled tests to see huge or substantial differences in IQ of MFT compared to FF.
 
...

The 9yr gap has brought huge advances in the feature set and performance of cameras image quality has received much more modest advances.
They didn't receive and will never receive because there is nothing there left to receive. The read noise is now so low that the pixels can (almost) spot individual photons. 10 years ago the typical input referred read noise was about 5 electrons, now it is about 1.3 electrons. Even if you get read noise to zero, you will not see any improvement in a typical test or landscape shot. Have to test these sensors on deep space photography.
Modern RAW processing software is getting so good that even if one camera does have an advantage by the time it's processed the advantage is gone. IMO That's the flaw in comparing unprocessed RAW images.
Though the software makers don't restrict their use to oneformat
 
I bought the FZ300 when shopping for a replacement for my Pentax W20 weather-proof camera. The FZ300 doesn't have that level of water-proofing, but it holds up in a light rain, and that's all I really need from it. The huge zoom range is not something I've put to use yet (bought mine in 2018). About 95% of my pictures are taken at the 50mm equivalent setting, and I don't think I've gone past 90mm equivalent except for some testing in the backyard. When handling different options at the store, I just liked how the FZ300 felt in hand, it's very usable EVF, and just how easy and pleasant it is to operate. With its f/2.8 lens and effective image stabilization, it's my most capable low light camera; good enough, anyway, for static subjects in dim museums. If and when my E-450 dies, the FZ300 can very easily take its place (though I know I will miss the E-450's optical viewfinder!). Someday I might re-visit a wildlife park, and perhaps then I'll get some use out of the FZ300's telephoto range.
Another factor to consider is how well the latest NR/AI software works with older cameras. As I don't have an FZ300 these are 3 samples from Imaging Resources review { thankfully though the site is closed down they have kept their archive active }. DXO NR, cropped to my favorite 3:2 aspect ration and upscaled to an 18x12" 300 ppi in PS tinkered to taste :-) .
I'm also thankful that Imaging Resource's content is still available. When DPR's reviews moved away from focusing on JPEG output, Imaging Resource became my favorite review site.

That's interesting information about NR/AI software, but, at least at present, I'm almost entirely a straight-out-of-camera kind of guy. If I don't go above ISO 400 on the FZ300, I'm pretty pleased with the results, but it's nice to see that good results at higher ISO's are possible if a person is willing to do some extra work. The major exception to my SOOC preference is that I do shoot raw with my Sigma DP2s compact. Its JPEG image processor is sorely lacking in comparison to shooting raw, even if the only software I've used has been Sigma Photo Pro (version 4.2, I think), which I've been very pleased with.
 
Reading the comments one might conclude that for the majority it actually doesn`t matter since it`s difficult in other than the most controlled tests to see huge or substantial differences in IQ of MFT compared to FF.
We must be reading different threads. The thread started with a comparison between two m43 cameras, then the OP added a comparison between two FF cameras. A few people said that for a given format one should not expect any more IQ improvement. If you want to potentially access better IQ, then a bigger sensor is the way to go. Sure, there will be people who say "There is no need for me, good enough is already good enough". However, I am pretty sure that for the majority of this forum the m43 is NOT the largest format that they own.
 
...

The 9yr gap has brought huge advances in the feature set and performance of cameras image quality has received much more modest advances.
They didn't receive and will never receive because there is nothing there left to receive. The read noise is now so low that the pixels can (almost) spot individual photons. 10 years ago the typical input referred read noise was about 5 electrons, now it is about 1.3 electrons. Even if you get read noise to zero, you will not see any improvement in a typical test or landscape shot. Have to test these sensors on deep space photography.
Modern RAW processing software is getting so good that even if one camera does have an advantage by the time it's processed the advantage is gone. IMO That's the flaw in comparing unprocessed RAW images.
Though the software makers don't restrict their use to one format
That doesn't matter because the end results look the same to me even if one RAW image has more noise than the other. Of course I am referring to sensors of the same size and from the last 6 or so years.
 
Reading the comments one might conclude that for the majority it actually doesn`t matter since it`s difficult in other than the most controlled tests to see huge or substantial differences in IQ of MFT compared to FF.
It's not that image quality (IQ) doesn't matter. Of course, it matters. It's just that different photographers pursue and optimize IQ in different ways.

A decade ago when I upgraded from an APS-C Nikon D90 to a full-frame D600 for landscape photography, that move came with a 2-stop advantage in dynamic range at base ISO. It changed how I did photography. Over the course of a year, I went from being a dedicated HDR photographer who routinely composited 3-5 bracketed exposures to create a single image to working with single exposures for 80-90% of my photos and two exposures for the balance.

The early 2010s was an era when upgrading from a 5-7 year-old APS-C format camera to full-frame could realize that kind of improvement. Today, the ca. 2016 Nikon 500 I use for wildlife photography gives up less than 1/3-stop of dynamic range at ISO 400 in comparison with a ca. 2020 Nikon Z6II at the same ISO.

The move I made in 2018 from a full-frame D610 to an APS-C D500 allowed me to realize significant gains in IQ as a result of the APS-C body's superior autofocus, burst rate, buffer, controls and user interface. In the last year, advances made by Adobe in the areas of image sharpening and noise reduction, have allowed me to realize further gains in IQ.

Yes, image quality is largely driven by how much light we work with. It's also about getting the shot in focus, having more good options from which to choose, masking noise and enhancing details in a photo.

When I upgrade from the D500 to mirrorless, one of the biggest gains for me will be silent shooting. The ability to remain silent will mean the subjects I'm photographing will be less likely to be aware of my presence and more likely to behave normally. Capturing those behaviors will translate to an improvement in IQ.

IQ matters. People just pursue and achieve improved IQ in different ways.
 
Image quality does not matter, but low-light capability, dynamic range and headroom for editing do matter. However, a larger sensor comes with a hefty price tag and higher weight. I find APS-C to be my balance. I do not find MFT to be cheaper or lighter. Full frame is over my budget as a whole system and a bit too heavy as well (again, as a system with a few bodies).

Image quality matters when I miss the image due to autofocus though. So I will focus more on AF capability in my next purchase.
 
Image quality does not matter, but low-light capability, dynamic range and headroom for editing do matter. However, a larger sensor comes with a hefty price tag and higher weight. I find APS-C to be my balance. I do not find MFT to be cheaper or lighter.
well, full-frame bodies on the used market can be found at attractive prices. And I believe m4/3 beats APS-C regarding the lenses options.
Full frame is over my budget as a whole system and a bit too heavy as well (again, as a system with a few bodies).
+1. Gave up with FF more than ten years ago. Not compatible with hiking, travel,...and my sore back.
Image quality matters when I miss the image due to autofocus though. So I will focus more on AF capability in my next purchase.
+1 on this too!
 
Image quality does not matter, but low-light capability, dynamic range and headroom for editing do matter.
That sounds self-contradictory. Why does dynamic range matter if not for image quality?
 
Of course. Both the buyers and the manufacturers want it. The manufacturers can sell their New Things (with higher profit margins) and the buyers want a bigger jump in performance.
 
... just seem to be programmed to want the greatest of everything, even if it's far beyond anything that would ever be useful to them. Resolution, detail and all of that in photography seem to be one of those things. Sure, a huge format offer a technically higher level of performance, but the question is whether it will be useful for the kind of photography that you do.
This counts for me as well. I consider myself a beginner so I ask questions to learn. Point is, probably for many beginners, I have a limited budget. But for that budget I want to buy the best possible gear. Since I am a beginner it is hard to decide what the best possible gear is for my situation. What I want to prevent is that I am buying something quite expensive, to find out that what I have purchased lacks in image quality. That is why I think newbies like me are focusing on IQ. I do understand I cannot buy the best of the best. I do not have thousands of dollars to spend on a body and a lens. However, the money that I can spend, I want to spend in the best way possible. Then there are other factors that influence my choice. Like for example, I do not want my kit to be very heavy (because then I won't take it with me) and I also want some decent zoom. So all of this together does not make it an easy choice, especially for a beginner.
A friend of mine, who bought a FF Sonny camera against my recommendations (I said that he should start of with smaller, cheaper, more compact gear) and decided that with the lenses he might want to use with it that the thing is too big and too expensive, was talking to me about buying some dedicated video camera that he's read about that's got far better low light performance than the Sony he bought. The thing is that the guy is a newbie to photography and even more so to video... I guess that he just hates the idea of even putting the time into learning photography or video unless he's capturing it on the very best gear that he can get. All of this is of course silly, but it seems that enough folks think this way to sell lots of expensive photo gear. There's a thought process that's more about "what might conceivably be useful to me in the most extreme situation that I haven't yet run into?" rather than the more reasonable thought of "how good do I really need for the kind of thing that I'm most likely to be doing a lot of?
 
you don't need to know nuffing to took a pitcher.
 
Is useful sensor evolution maxed out?
I think it's been established that existing sensor technology is close to maximum theoretical efficiency in terms of light collection. Pixel density is also more than adequate for almost any purpose. What other aspects of IQ would benefit from further sensor advances?
(I exclude shutter technology improvements such as global shutter, and image stabilization which is part of the sensor mount, not the sensor itself).
Funny. I've heard that "maxed out" argument at least a few times a year over the last 20 years. I'm guessing it's not.
Interesting because I have not heard that argument at all until recently. It's only in the last 5 years that we have reached that point.
 
Is useful sensor evolution maxed out?
I think it's been established that existing sensor technology is close to maximum theoretical efficiency in terms of light collection. Pixel density is also more than adequate for almost any purpose. What other aspects of IQ would benefit from further sensor advances?
(I exclude shutter technology improvements such as global shutter, and image stabilization which is part of the sensor mount, not the sensor itself).
Funny. I've heard that "maxed out" argument at least a few times a year over the last 20 years. I'm guessing it's not.
Interesting because I have not heard that argument at all until recently. It's only in the last 5 years that we have reached that point.
Well, I've heard it for quite some time, not just the last five years. There are always those who deride new advances as unnecessary indulgences.
 
Is useful sensor evolution maxed out?
I think it's been established that existing sensor technology is close to maximum theoretical efficiency in terms of light collection. Pixel density is also more than adequate for almost any purpose. What other aspects of IQ would benefit from further sensor advances?
(I exclude shutter technology improvements such as global shutter, and image stabilization which is part of the sensor mount, not the sensor itself).
Funny. I've heard that "maxed out" argument at least a few times a year over the last 20 years. I'm guessing it's not.
Interesting because I have not heard that argument at all until recently. It's only in the last 5 years that we have reached that point.
Well, I've heard it for quite some time, not just the last five years. There are always those who deride new advances as unnecessary indulgences.
While there have been amazing advances in camera technology { Excellent IBIS, AF, multi-image in camera etc } From a pure sensor IQ perspective I don't think things have moved as far as many imagine. Which can be easily demonstrated by looking at for example reasonably controlled raw sample files of cameras going back many years both here and Imaging Resource have a huge selection of camera reviews. You can argue that the advances in camera technology make things easier which is true. I am not sure I want easier

D3s a 14yr old camera 12800 ISO sample from photography blog UK { so anyone can have a go }. With modern processing

686b392e379b45d9b49b3af12c79060f.jpg

I was doing lots of wedding/event stuff back then and moving from Nikon DX { D2x} to the D3s was by far the largest jump in IQ I have experienced , bordering on revolutionary for the time :-)



If you compare the best from each generation to the best from today you can go back quite a way till any differences become glaring. Throw in the genuine advances in AI software and you can just about make a silk purse from a photographic pigs ear :-) Having jumped into the high end high feature set cameras in the formats I shoot { OM-1 & Z9 } I have come to the rather belated conclusion that for me a lot of the "features" actually take the fun out of photography .

Jim Stirling:
“It is one thing to show a man that he is in error, and another to put him in possession of truth.” Locke
Feel free to tinker with any photos I post
 
Last edited:
Is useful sensor evolution maxed out?
I think it's been established that existing sensor technology is close to maximum theoretical efficiency in terms of light collection. Pixel density is also more than adequate for almost any purpose. What other aspects of IQ would benefit from further sensor advances?
(I exclude shutter technology improvements such as global shutter, and image stabilization which is part of the sensor mount, not the sensor itself).
Funny. I've heard that "maxed out" argument at least a few times a year over the last 20 years. I'm guessing it's not.
Interesting because I have not heard that argument at all until recently. It's only in the last 5 years that we have reached that point.
Well, I've heard it for quite some time, not just the last five years. There are always those who deride new advances as unnecessary indulgences.
While there have been amazing advances in camera technology { Excellent IBIS, AF, multi-image in camera etc } From a pure sensor IQ perspective I don't think things have moved as far as many imagine. Which can be easily demonstrated by looking at for example reasonably controlled raw sample files of cameras going back many years both here and Imaging Resource have a huge selection of camera reviews. You can argue that the advances in camera technology make things easier which is true
Actually, I think the advances have been massive. For example:

8e77194bbd9144339d4f98233fbd4e6c.jpg.png

560918ae73664632906fe19dd0451690.jpg.png

To put this in perspective, the IQ of an R5 is so much higher than the IQ of a 5D that we could literally use, say, the Canon RF 35 / 1.8 IS + RF 85 / 2 IS and [in most cases] get IQ as good as a 5D + 24-70 / 2.8L + 70-200 / 2.8L IS via cropping. The R5 would lose the 24-34mm focal range, though, and have a bit more DOF [and a bit worse for wear with IQ] when doing the most extreme cropping. But, you could throw in the RF 24 / 2.8 pancake and regain the 24-34mm focal range, and the whole setup would still be significantly smaller, lighter, and less expensive (that what the 5D setup would have cost at the time).

That's extraordinary, is it not? Well, I think it is. It's just that, well, we all want the "higher IQ", so we don't actually do that. But we could.
If you compare the best from each generation to the best from today you can go back quite a way till any differences become glaring. Throw in the genuine advances in AI software and you can just about make a silk purse from a photographic pigs ear. :-). Having jumped into the high end high feature set cameras in the formats I shoot { OM-1 & Z9 } I have come to the rather belated conclusion that for me a lot of the "features" actually take the fun out of photography.
The operational advantages of modern gear are also amazing. For me, the ability to set the focal point anywhere in the frame and not having to focus and recompose is so nice -- especially combined with servo if the subject is moving.

But, I've yet to take advantage of tracking. For example, when shooting hockey, I have my doubts that it would track the person I lock on to with the other players constantly getting in front of the subject. Maybe it would, though -- I should try it some time.
“It is one thing to show a man that he is in error, and another to put him in possession of truth.” Locke
“A lie is more comfortable than doubt, more useful than love, more lasting than truth.”

― Gabriel Garcia Marquez
Feel free to tinker with any photos I post
You can tinker with the ones I posted above. ;-)
 
Last edited:

Keyboard shortcuts

Back
Top