[Class A]

In regards to 1 & 2, because the bigger sensor in the Nikon can utilize more of the light coming in and has bigger photosites makes it "cleaner" at higher iso values. #3 becomes moot because of #1 and #2. That's doesn't make it a myth, that's just physics.

A bigger sensor can only utilise "more of the light coming in", if there is more light to utilise. In other words, it is the lens providing more light to utilise that should be given the credit, not the larger sensor.

If you simply use a larger sensor, but not a faster lens, then the same total amount of light is used; it is just spread out over a larger area with a corresponding loss of light flux (photons per square mm). Projecting the same image onto a larger area and then capturing it with a larger sensor does not give you any noise advantage whatsoever. The noise advantage comes from projecting onto a larger area while maintaining the same flux which means you are using a higher amount of total light. The latter is available only if the lens can provide it.

Ergo, what the lens delivers counts, not what the size of the image forming area is.

There is no larger-sensor advantage if you always must keep DOF the same for the same FOV. But no-one ever mandates that, and in fact 'accepting less DOF for the same FOV' often turns into 'welcoming less DOF for the same FOV' in a lot of cases, once you shoot for a while.

Of course you can change the DOF.

But if you "accept less DOF" then it is no wonder that noise performance will be better as more total light will be used. This obviously happens if you don't change the sensor size at all. Open up the aperture and you get a brighter (less noisy) image, on FF, APS-C, any format.

You don't need to switch to FF to "accept less DOF". Just open up the aperture a bit more.

With one exception: If you would like to open up the aperture some more but there is no APS-C lens that allows you to. Then you may want to switch to FF because there are FF lenses that open up their apertures some more.

The Sigma 18-35/1.8 is an example for a lens that suddenly addressed a low-light shortcoming of APS-C with respect to zoom lenses. Prior to its introduction you could not take certain images with a zoom on APS-C. You had to switch to FF to get these images (because there are respective, faster, FF zooms). Now you don't have to switch anymore (but your sensor size did not need to change for this to happen).

So... 'myth' is misleading...

I disagree.

To claim that a larger image forming area is responsible for a reduction of noise is factually wrong.

I absolutely do not object to someone stating "I bought an FF camera so that I can utilise the fast lenses that are only available for the FF format (as in they have no equivalent in APS-C mount) to support my low-light photography".

But people often state (or imply) "I bought an FF camera because the larger sensor gives me better low-light performance ".

The latter is a "a widely held but false belief or idea", i.e. a "myth".

---

Last edited by Class A; 04-05-2014 at 08:56 AM.

[normhead]

Fortunately you can just look at the images, you have no need to listen to someone going on and on about what's true and what's not, in theory, with total ignorance of the obvious practical reality, not that we have anyone like that here..

Where is the part where I say... what you think you know is not verified by the images I'm looking at. ?

[Class A]

Fortunately you can just look at the images, ...

I'd like to rephrase that as "Unfortunately, you can just look at the images ...".

Looking without thinking is not helpful.

Say someone takes a photo of a runner with a Pentax at 1/10s shutter speed and f/28.
Someone else takes a photo of the same runner with a Nikon at 1/1000s and f/2.8.

The exposures are the same, but on the Pentax photo the runner is blurred, on the Nikon photo the runner is sharp.

Obviously this makes the Nikon the better camera for sports.
To know that, "fortunately you can just look at the images".

Nevermind the inappropriate comparison.

[jsherman999]

Of course you can change the DOF.

But if you "accept less DOF" then it is no wonder that noise performance will be better as more total light will be used. This obviously happens if you don't change the sensor size at all. Open up the aperture and you get a brighter (less noisy) image, on FF, APS-C, any format.

You don't need to switch to FF to "accept less DOF". Just open up the aperture a bit more.

With one exception: If you would like to open up the aperture some more but there is no APS-C lens that allows you to. Then you may want to switch to FF because there are FF lenses that open up their apertures some more.

But your 'one exception' is a rather large one. It applies to all f/2.8 zooms (recent Sigma 18-35 1.8 not-withstanding - it has more limited FL range than a 24-70,) and many other common lenses you would buy. For example, a 50 1.8 on FF to replace a 35 1.8 on aps-c. There are no real aps-c equivalents to a 14-24 2.8, 20mm 2/8, 24mm 1.4, 24mm 2.8, 35mm 1.4, 35mm 1.8, 50mm 1.4, 85mm 1.4, or even really my 180 2.8. (& there's reportedly a new Sigma 24-50 f/2 coming!)

I disagree. (about 'myth' being misleading)

To claim that a larger image forming area is responsible for a reduction of noise is factually wrong.

I absolutely do not object to someone stating "I bought an FF camera so that I can utilise the fast lenses that are only available for the FF format (as in they have no equivalent in APS-C mount) to support my low-light photography".

But people often state (or imply) "I bought an FF camera because the larger sensor gives me better low-light performance ".

Here's a better way to state it, IMO: The reasons some people think FF gives them a low-light advantage are often wrong, or misconceptions.

That's different than labeling it a 'myth'. because that 'myth' adjective always has to carry the caveats you gave, otherwise it could easily be misconstrued into meaning "you will never see a low light advantage to shooting a larger sensor." The fact is you do see it with many common lens combinations you might choose. In my experience, when people move up a format, they generally frame the same (same FOV,) use the same f-stop to get the same shutter speeds, and then see the noise advantage, while accepting the change in DOF. And I think this is exactly why DXOmark represents things that way - they recognize that shooting reality.

I know exactly what you're saying and there's nothing I really disagree with besides what I consider a too-loosely applied adjective

.

[Class A]

But your 'one exception' is a rather large one.

Granted, but the exception still pertains to lenses rather than sensor size.

The reasons some people think FF gives them a low-light advantage are often wrong, or misconceptions.

I agree with this statement, but not that it is "a better way to state it ".

As you do, I think we are actually in agreement, but have a different view on what "FF' refers to.

If "FF" means "the whole FF system, including current lens offerings" then you are correct. The whole system with the current FF lens offerings indeed has an advantage over the current lens APS-C offerings.

If "FF" means "full frame sensor" then my characterisation of a myth is correct.

I believe that the vast majority of people decide to move to "FF" because of the larger sensor and that they attribute the "low-light advantage" to the larger sensor. They are acting upon a myth, notwithstanding the fact that when they then actually use fast FF lenses (as opposed to a 70-200/4 zoom, for instance) that they'll actually get a low-light advantage, if they are happy to accept less DOF.

[SteveM]

I've had a few months with both cameras and still pull out the K5 and K7 from time to time. Does the pixel size on FF equate to more efficient light absorption? Is the better availability of fast lenses for FF the difference?

I don't know....does it matter?

After a few months with these cameras, I have realized cameras are like boats....you're never happy with one (inside joke for anyone that has gone from an aluminum you can run up on the beach to a gelcoat...or one for the family...or one cheap on gas....). Anyways;

Df - My go-to camera for indoor shots where I don't want a strobe. Regardless of the reason, far best for noise. I have since picked up the Zeiss 21/2.8 and the Voigtlander 125/2.5. Nice camera and I enjoy using it, but I am finding I am not drawn to it outside of low light shooting....which puts these lenses in question. Presumably the 16M sensor, it doesn't pull in the detail offered by the K3.

K3 - Amazing detail, and it really extends your ability for telephoto work, but more difficult to use. Camera shake will show up where it won't with my K5 or Df....not a forgiving camera, but that is not meant as a bad thing. Doesn't have the issue with reds like the K5. Depending on the light, noise becomes an issue even at low ISO. It can be equal or possibly better than the K5 under certain light, but drops considerably under many light conditions. It is not implied to be a large issue....if I was to choose one out of all 4 cameras, the K3 would still be it because of what it does offer. Side note......damn that Voigtlander 125 I bought for the Df!...I had to pick one up in Pentax mount. What a ridiculously good lens.

K5 - Good handling of noise at high ISO and a very forgiving camera. Becomes flat in detail like the Df at 16MP. I don't take it out much now, but I still love it. After 30 years of buying cameras.....this is one of about 4 I would consider a true flagship.

K7 - Processed from the pixels/sensors to "raw" is one of the better qualities of this camera. That was the one complaint with the K5, which sometimes suited certain types of photography, but I didn't like it for landscape. The K7 shines in this area, but I suspect the K3 may be the same in this regard (haven't spent enough time using the K3 for landscape yet)

[jsherman999]

If "FF" means "the whole FF system, including current lens offerings" then you are correct. The whole system with the current FF lens offerings indeed has an advantage over the current lens APS-C offerings.

Perfectly well stated IMO. And it might be appropriate to point out that you see this advantage immediately with something like a $100 50 1.8, or if you stick with zooms, a $300 Tamron 28-75 f/2.8. You don't need uber-lenses with FF.

If "FF" means "full frame sensor" then my characterisation of a myth is correct.

Yes, in exactly the same way "The aps-c sensor allows you better noise performance than point n' shoot" is actually incorrect, or a myth. It is incorrect, but the practical realities of the typical aps-c system (ie you have lenses available that open up wider than f/13 ) make it effectively true, and something a buyer would enjoy right away.

I believe that the vast majority of people decide to move to "FF" because of the larger sensor and that they attribute the "low-light advantage" to the larger sensor. They are acting upon a myth, notwithstanding the fact that when they then actually use fast FF lenses (as opposed to a 70-200/4 zoom, for instance) that they'll actually get a low-light advantage, if they are happy to accept less DOF.

Well, I don't think that describes the vast majority of buyers. It might be the slight majority of all forum readers who are perhaps simply commenting on things who think that.... I could be wrong.

Part of the reason 'myth' is misleading is that I think you would have to almost make an effort to give up the noise advantage with FF. For example. if you bought the f/4 zooms only, or variable-aperture consumer zooms only.** The F4 zooms would probably be excellent optically and a bit smaller and less expensive than the f/2.8 zooms, but you would be giving up the FF noise/DOF advantage in buying those (if they were all you bought.) Or if you for some reason were only shoooting the fast primes wide-open on aps-c and moved to FF to shoot f/2.8 zooms - not gaining anything that way in terms of noise control. I just don't think that describes many real people.

I will say this - if you are right about the % of people who 1) don't understand why larger sensors give you the noise advantage and 2) intend to buy only variable aperture or f4 zooms with their FF purchase while 3) still thinking they will see a low-light advantage, strong language like 'myth' might be appropriate to get their attention and serve as a consumer advocacy. Otherwise, IMO, 'myth' has legs the same way a provocative thread title does, and causes more confusion than it's worth. IMO.

** EDIT: or if you bought fast-enough lenses, but were continually stopping down to match the same DOF you had before with aps-c or m/43. That might describe more people than the other scenario...

(OK, back to K3 vs. Df or whatever this thread is really supposed to be about)

.

---

Last edited by jsherman999; 04-06-2014 at 06:55 AM.

[Class A]

(OK, back to K3 vs. Df or whatever this thread is really supposed to be about)

I don't think our discussion was off-topic.

The thread started with the idea to compare two cameras and one of the cameras was bought in order to address low-light situations. The OP wrote:

" So why the Nikon? Half of what I shoot is in low light, so there is definitely an allure from having the high ISO performance that full frame offers. "

Well,

1. "high ISO performance that full frame offers" definitely is a myth (or "false belief" if you feel that is more appropriate).
2. a comparison is only instructive if the playing field is level.

Hence, I believe a discussion on what a level playing field is and whether some buying motivations are based on myths is on-topic.

I think we very much agree on everything except that we don't know how large the percentage of people is that believe in the incorrect "larger sensor -> better low light performance" idea. I think that there are sufficiently many to call this false belief a "myth". Maybe there aren't, in which case it is just a "false belief". Not sure the difference matters that much. Surely in the context of this forum, I believe it is warranted to speak of a myth since it pops up time and again with reliable frequency.

Part of the reason 'myth' is misleading is that I think you would have to almost make an effort to give up the noise advantage with FF.

I don't want to debate for the sake of it and believe we are in agreement in principle, but I really don't think this particular statement is true.

My assumption is that a photographer wants to make a certain image with a certain DOF. That's why Av mode is very popular. If you know what DOF you would like, you simply choose the f-ratio accordingly. On an FF-camera, the f-ratio will simply be higher than on an APS-C camera. You essentially get the same image with no noise advantage for the FF-camera.

Your statement would be true if people always shot with certain f-ratios, say f/2.8 and f/8, and would not change their habits when switching systems. But I'd be seriously wondering about such people because they'd suddenly get images with less DOF but wouldn't worry about that and instead just be happy about less noise. That doesn't make sense for me.

Another way in which your statement would be (trivially) true is if APS-C shooters were shooting wide open all the time. If they continued that practice on FF then obviously they'd get less noise (and less DOF). However, I don't think that shooting wide open frequently is that common among APS-C users. Note that some of the comparison shots by the OP did not fully exploit the capabilities of the APS-C lens, so it doesn't really make sense to open up the aperture on FF even more.

I know that often people stop down slightly to get better IQ and there is less need to do that on FF, so surely FF invites you to open up the aperture more than you might have done on APS-C, but then I don't think that many people feel constrained by the DOF / IQ APS-C is currently giving them. They may feel constrained by the noise, but switching formats won't help unless they are prepared to accept less DOF (which they apparently were not, unless they were shooting wide open already).

[Class A]

Since a message can get lost in a string of arguments and nested sentences, here's what I would like to express in a nutshell:

If you are not yet using the widest possible aperture on APS-C that still provides you with an IQ that you are happy with, don't think that moving to FF will buy you a low-light/noise advantage.

Simply "sacrifice" more DOF and only if you cannot do the latter anymore, it is necessary to move a system which has faster lenses.

[tuco]

Why are most people concerned with noise between APS-C and FF? IMHO, that is not as important as how much light the camera can capture at the higher ISOs. For example, between ISO 800-1600 you capture about two more stops of light with the Df over say the K3. I'd rather have that.

[normhead]

And why would anyone care about that? What's more important, how much light it captures, or what it does with the light it captures?

And how do you figure it captures 2 more stops?
The sensor is double the size, that's one stop... where does the second one come from?

[tuco]

And why would anyone care about that? What's more important, how much light it captures, or what it does with the light it captures?

And how do you figure it captures 2 more stops?
The sensor is double the size, that's one stop... where does the second one come from?

Because taking candid shots of people and similar shots exposure blending and all that stuff is not very practical. And around ISO 800+ is where the majority of my candid and casual pictures are taken. I'll take all the DR I can get in those situations. I'm trusting the DR measurements vs ISO range that DXO posts for that two stops biz. I know you think DXO are frauds but let's hope that chart is one of their more valid ones.

[normhead]

Because taking candid shots of people and similar shots exposure blending and all that stuff is not very practical. And around ISO 800+ is where the majority of my candid and casual pictures are taken. I'll take all the DR I can get in those situations. I'm trusting the DR measurements vs ISO range that DXO posts for that two stops biz. I know you think DXO are frauds but let's hope that chart is one of their more valid ones.

All I know is if I shot low light and 800-1600 ISO I'd already have a D600 or 5D or something like it.... but it's less that 2% of my images so I don't even care to investigate. For what I do, it's one stop.

[tuco]

All I know is if I shot low light and 800-1600 ISO I'd already have a D600 or 5D or something like it.... but it's less that 2% of my images so I don't even care to investigate. For what I do, it's one stop.

In the digital camera part of my photography, that's why I have two cameras. One for resolution/tripod work and one for candid and casual type work. I use to have a mirrorless for the casual role the but I decided to share a common mount to economize my glass and the fact I'm old-school and prefer FF 35mm because that is what I'm use to.

[dosdan]

I know that often people stop down slightly to get better IQ and there is less need to do that on FF, so surely FF invites you to open up the aperture more than you might have done on APS-C,

I wonder if this is the case. Vignetting is usually present in most lenses used on a FF. So if you take your FA 50/F1.7 from your APS-C camera and put it on your FF, even if you can accept the very shallow DOF & softness wide open, you may not appreciate the obvious vignetting.

Dan