Question: Full Frame advantages over APS-C in lower light environment?

[herbert7265]

Excellent discussion above, but here is the math that helps put light sensitivity, focal length, aperture together. Remember the exposure triangle - shutter speed, aperture, and ISO, adjusting one, requires adjusting the other for the same "perfect" exposure - but different features such as resulting motion blur or different depth of field. With todays cameras, grainess of higher ISO is not as much a factor as the old film ISO, both native and push processing.

In addition, remember the exposure triangle as you need more light and you can't open the aperture further, so it is slower shutter or higher ISO.

And here we go...!

No offense intended, I am sure you have only the best intentions with your explanations, but as they are fundamentally flawed they do not help overall.

No matter how many articles, webpages, YouTubers, and whatever else, may state it... there is NO "exposure triangle"! This is one of these everlasting misconceptions in photography, but no matter how often it gets repeated and how fiercely it gets defended... it´s simply false.

In most easy terms: The only elements that have direct (!) control over the exposure in your camera are aperture and shutter speed. ISO is NOT the sensitivity of your camera sensor (!), means it does not alter the sensitivity of the sensor when changed. ISO is a scaling coefficient applied to the data / signal obtained as a result of the exposure.

If of interest, potentially to say good-bye to this misconception, maybe read this article. I think it´s one of the better articles to understand why there is NO "exposure triangle", never has been.

The Unbearable Lightness of Mystic "Exposure" Triangle

We should begin by saying that the point of all of this is not just to criticize one of the most repeated photographic myths out there – that's just a tool, because an incorrect understanding of the way that exposure and ISO work, as well as their roles during the shooting process, leads to a...

www.fastrawviewer.com

Another interesting article, again just one of a lot of available articles, is this one about noise:

What's that noise? Part one: Shedding some light on the sources of noise

How would you react if you were told that the aperture and shutter speed you choose make more difference to image noise than the ISO setting? You might be surprised to discover that a lot of the noise in your images doesn't come from your camera at all: it comes from the light you're capturing...

www.dpreview.com

First when we understand these basics in photography we may subsequently understand other impacts like bigger sensors on aspects like low light capabilities, resulting noise in a final image, and so on.

Herbert

 

[r.harris1]

And here we go...!

No offense intended, I am sure you have only the best intentions with your explanations, but as they are fundamentally flawed they do not help overall.

No matter how many articles, webpages, YouTubers, and whatever else, may state it... there is NO "exposure triangle"! This is one of these everlasting misconceptions in photography, but no matter how often it gets repeated and how fiercely it gets defended... it´s simply false.

In most easy terms: The only elements that have direct (!) control over the exposure in your camera are aperture and shutter speed. ISO is NOT the sensitivity of your camera sensor (!), means it does not alter the sensitivity of the sensor when changed. ISO is a scaling coefficient applied to the data / signal obtained as a result of the exposure.

If of interest, potentially to say good-bye to this misconception, maybe read this article. I think it´s one of the better articles to understand why there is NO "exposure triangle", never has been.

The Unbearable Lightness of Mystic "Exposure" Triangle

We should begin by saying that the point of all of this is not just to criticize one of the most repeated photographic myths out there – that's just a tool, because an incorrect understanding of the way that exposure and ISO work, as well as their roles during the shooting process, leads to a...

www.fastrawviewer.com

Another interesting article, again just one of a lot of available articles, is this one about noise:

What's that noise? Part one: Shedding some light on the sources of noise

How would you react if you were told that the aperture and shutter speed you choose make more difference to image noise than the ISO setting? You might be surprised to discover that a lot of the noise in your images doesn't come from your camera at all: it comes from the light you're capturing...

www.dpreview.com

First when we understand these basics in photography we may subsequently understand other impacts like bigger sensors on aspects like low light capabilities, resulting noise in a final image, and so on.

Herbert

I'm a huge fan of Iliah and team on FRV/RawDigger/LibRaw and the article you mentioned from him is fantastic and full of interesting factoids. It may or may not move the needle in thinking. I think the challenge is more the fact that cameras actually still implement and present to users the "triangle" - i.e. expose for lightness more than image quality. In other words, in reality, correct exposure by having clean shadows without blown highlights is often a different set of answers than the results of a classic exposure triangle (as implemented by cameras) gives, to your point. Sometimes though, you do need that shutter speed and/or that aperture and you need to apply that gain. The good thing is that it doesn't have to be a "triangle answer" and having some insight into your camera and how it behaves lets you control your settings appropriately and get the most out of your sensor.

 

[herbert7265]

I'm a huge fan of Iliah and team on FRV/RawDigger/LibRaw and the article you mentioned from him is fantastic and full of interesting factoids. It may or may not move the needle in thinking. I think the challenge is more the fact that cameras actually still implement and present to users the "triangle" - i.e. expose for lightness more than image quality. In other words, in reality, correct exposure by having clean shadows without blown highlights is often a different set of answers than the results of a classic exposure triangle (as implemented by cameras) gives, to your point. Sometimes though, you do need that shutter speed and/or that aperture and you need to apply that gain. The good thing is that it doesn't have to be a "triangle answer" and having some insight into your camera and how it behaves lets you control your settings appropriately and get the most out of your sensor.

Let´s tackle some basic aspects of this discussion about the "famous exposure triangle"...!

Some people may see it as nitpicking, but at the end of the day the correct terminology matters, at least when you want to really understand a topic. Let´s for example take the term lightness, which you used.

Yes, we normally strive to expose for the desired / correct lightness of an image. But lightness of an image is not the same as exposure of an image. Even a "perfect" exposed image may not give you the desired lightness, and at that point ISO may come into play or post processing.

Look, struggling with these fundamental basics, means
- the exposure of an image, given a certain sensitivity of the sensor or film, is only controlled by the aperture and shutter speed, not ISO
- the predominant form of noise is shot noise, which again is connected to the exposure
then easily may lead to wrong conclusions.

As an example: Let´s assume you are shooting a scene with a given aperture, shutter speed and ISO, and with these settings you get the best possible exposure. If you now either use a smaller aperture or faster shutter speed, and you compensate for these changes with a higher ISO setting, you still will get the same image lightness, but your exposure is no longer perfect, means your image is underexposed. On the other hand, shot noise is connected to exposure and increases when the exposure decreases. As a result you will receive an image with the same lightness, but more noise. But this additional noise is not caused by the higher ISO setting, but the lower exposure of the image. (All this in very easy terms).

Now let´s translate this into the original topic of this thread and assume the OP referred also to the aspect of low light capabilities FF vs APS-C: Popular "knowledge" and understanding says you can shoot FF at higher ISO and still get a clean image. Nothing wrong with that, no sense to reflect at this moment to some more detailed aspects in there.
People who understand the basic concepts will still strive for the best possible exposure and use the benefits of FF to do so. But a lot of people who do not understand these basics will potentially just use higher ISO, that way not expose as good as possible, and then barely get the benefits they may have expected.

Finally, there is a lot more involved, as we all know, for example sensor characteristics, using ISO in camera vs post processing, RAW vs JPEG, shooting techniques, just to name a few. But I learned one thing long time ago... first learn to walk before you start to run. And again, terminology matters...!

Herbert

 

[Slartibart]

I think nobody doubts that in a digital camera if one changes the ISO settings, the gain of the system is increased, amplifying the signal from the sensor. AFAIK this is implemented in most cameras at three different steps of the RAW imaging pipeline:

1. In the sensor, by increasing voltage.
2. Via an analog amplifier outside of the sensor.
3. Digitally, after the signal has been digitized, but before storing data in the RAW file.

There are differences in sensor type, but on CMOS, there is an amplifier built into the “photosites“ which combine to pixel, or e.g. each pixel column throughout the sensor. There is a maximum voltage which can be applied in 1.) and the analog amplification in 2.) will max out at some value (in the “past” at ISO 1600, I do not follow this very closely and do not know wether this has changed; AFAIK at least most DSLRs have an amplifier before the analog-digital conversion).

That is why IMHO in 1.) & 2.) it seems that the “classic” relation between aperture, exposure and ISO is still valid.

Above ISO 1600 in 3.) the digital amplification takes place before the RAW files (e.g. 12bit, 14bit, … ) are stored - which is indeed different to the classical A-E-ISO triangular relation. There is a maximum value that can be stored in 12bit &company, so when you e.g. overexposed a photo 4-stops, even though the analog-digital conversion is probably not saturated, the RAW file will probably be clipped.
And yes, due to the analog amplification, RAW files at higher ISO do contain more detail. However, ISO 1600 and e.g. ISO 12800 should contain a very similar if not the same amount of shadow detail - unless there is some additional processing or the analog-digital conversion operates with a higher resolution than in whatever bit depth the RAW files are saved.
And even with 3.) true above ISO 1600, an ISO 1600 RAW may still contain more information about highlights because they can still be clipped through the digital amplification process.
Due to this and other things, like internal buffer size or battery life etc., when shooting RAW it can be good to take photographs at ISO 1600 (if you must) and simply post process later - but if the analog-digital conversion bit-depth is higher than the saved RAW format's bit-depth this obviously would not help.

My guess here is, that all this depends on the individual camera model and one would have to individually test this - e.g. take a RAW photo at ISO 100 and +2 stops (adjusted to an effective ISO 400 <– okay, ISO 400-ish 😀), and compare that to a RAW taken at ISO 1600 and -2 stops adjusted (again ~ISO 400) - of course both taken with same aperture, shutter speed, no noise reduction, sharpening, contrast adjustments, etc..❶ Then continue from there.

And while technically all this impacts on the validity of the A-E-ISO relation in digital photography, the interface presented to the user of a camera, as already mentioned by others, presents the “triangle” and acts “accordingly”.


☕︎☑︎

Otherwise I can’t really wrap my head around the effect on the depth of field/shallowness of a photo when using a FF- compared to e.g. APS-C mentioned here several times. I use Pentax K1 and KS2 and obviously for the same real world scene projected on each sensor size there is a difference in pixels to represent the scene’s x,y,z-coordinates.
I get that there are less y-pixels in APS-C to represent the depth of the real world scene compared to FF. But if I look at a photo at the same display or print size of the same scene taken on FF and APS-C (same aperture, shutter speed, ISO, no crop, etc.❷), up to a certain display/print size I am hard pressed to differentiate the FF from the APS-C based on the DOF - e.g. an A4-print in relation to DOF looks very similar. There might be other visible differences - maybe certain motifs/subjects make this more obvious?


nota bene:
❶ I actually would be surprised if there is no difference in what is stored in a RAW for ISO 100+2stops vs. ISO 400 vs. ISO 1600-2stops on the same camera.

➋ The ISO adjustment is the one which works quite black box-ish IMHO. The RAW stored with “ISO 100” on a Pentax K1 differs from “ISO 100” on a KS2, yes, obviously there is more light captured in the same amount of time because there are simply more sensor pixels, but my - purely anecdotal - impression is that FF-sensors are more sensitive and clearly have a different sensitivity range compared to APS-C.

———————

Probably because I am no native english speaker: @herbert7265 what are you referring to as “image lightness”? The representation of a color's brightness, e.g. defined as “value” or “tone” in some models?

 

[r.harris1]

Terminology can certainly be important, of course, but at the end of the day, it's the outcome that matters. Regardless of whether the concept of a legacy exposure triangle should be replaced by some other easy to memorize terminology (and I'm not sure it should or even can be), camera manufacturers do use those three levers to create what it would call a "properly exposed (or metered)" file, by which is means a nicely shaped histogram. At least in the various "auto modes". Nothing pegged at either end if the situation can be avoided, and certainly no blown highlights. If everything is pegged in the shadows when the camera is reading the scene exposure, an auto-ISO mode might apply gain to spread that signal out better on the histogram. That gain may not create noise but it does amplify any that's there (and there's usually noise).

More and more cameras are metering off of the sensor rather than a separate meter. This is a good thing because this lets people start to think about exposure in a better way (or at least a more controlled way). You have the ability to more easily expose for the highlights, which can help with the shadow end of the histogram too and many modern cameras can stand at least 2 stops of clean shadow recovery. So I'm less concerned with throwing out the concept of exposure triangle (since most cameras provide controls for users to think that way) and more interested in expanding the ways in which we can meter. Exposure - getting the desired amount of light to the sensor to capture the desired scene - then becomes easier.

So taking it back to the OP, understanding what your specific camera can do in a given situation (blue/golden hour) based on the things you can control - aperture, shutter speed and gain (ISO) is very valuable. With automatic settings, changing one thing (aperture) might change shutter speed and/or ISO to generate what the camera manufacturer thinks of as that "properly exposed" histogram. There's a lot more wiggle room in that histogram than folks sometimes realize so manually setting the bounds of what those applied gains can be, the shutter speeds you want to shoot at or the apertures you typically use can go a long way to getting the best out of your camera, whether it's APS-C, FF, MF or other.

 

[Jumpthesnark]

Hallo all you Pros!

Currently I own a Sony A6500 and it does all I need right now.

However, I have noticed that my favourite kind of shooting is very early morning or early evenings… I guess Golden/Blue hours.

I have read that full frame cameras do offer advantages when used in those conditions.

So, my question is, is that correct?

I am not expecting to suddenly get "better" at my photography, but was just curious if there were some advantages vs the downside of a larger and heavier camera.

Thanks for the help! 🙂

Edit:
If I do upgrade I will probably most definitely not be going beyond a used Sony A7 III. Gotta have some spare cash to keep Tim Apple happy. 🤣

You already own a Sony APS-C, so if you're happy with that system then a logical step to full frame would be in the A7 series somewhere.

You mention "lower light," as opposed to "low light." And I think that's accurate for blue hour/golden hour photography, which isn't at all "low light." I think any full frame camera would work under those conditions.

As several have mentioned, light sensitivity with low noise comes down to pixel size more often than sensor size. Sony's most well known full frame low light champ is the A7S series, with a 12MP sensor but a pixel pitch of a whopping 8.4 microns. That's a big bucket to pour light into. So there are fewer pixels, but each individual pixel is larger. Also, the pixels aren't crammed together as tightly, which helps to reduce noise when making long exposures. The other A7 sensors (in A7 & A7R bodies) have pixel pitches of around 3.8-6 pixels. Considerably smaller.

BUT all of that matters more for truly low light photography, where the camera can basically see better than our eyes can. That is not the situation you're describing, with blue hour and golden hour having plenty of ambient light available for an exposure. So depending on what you'll be shooting (you still haven't told us that), how large you want to reproduce your images and how you shoot, most any full frame camera with a fast lens should be fine for the light conditions you describe.

In other words, some cameras are better than others for low light work. But what you've described isn't low light.

 

[herbert7265]

Probably because I am no native english speaker: @herbert7265 what are you referring to as “image lightness”? The representation of a color's brightness, e.g. defined as “value” or “tone” in some models?

Maybe have a look into the link I already posted in my first response:

The Unbearable Lightness of Mystic "Exposure" Triangle

We should begin by saying that the point of all of this is not just to criticize one of the most repeated photographic myths out there – that's just a tool, because an incorrect understanding of the way that exposure and ISO work, as well as their roles during the shooting process, leads to a...

www.fastrawviewer.com

Here the summarized explanation from that link concerning the term “lightness”:

• Lightness* is not the goal of exposure, especially in raw – it’s an artistic element in the image, which is left to individual taste to be addressed during the processing and editing stages. Lightness is often edited not only by and not primarily by scaling, but rather simultaneously with contrast, using curves, channel blending, and sometimes even local intervention (masks, dodge/burn, etc.).

* We are using the term "lightness" while referring to digital values, like it is used in Lab color space; reserving the term "brightness" for where the actual reflected or emitted light is present. Lightness is more like filter transparency, a relative value determining what percentage of light will pass through, and not how bright the image will appear - that depends on the brightness of the light. Lightness is data, while brightness is a visual perception which you can change by simply rotating a knob on your monitor.

Herbert

 

[bunnspecial]

My guess here is, that all this depends on the individual camera model and one would have to individually test this - e.g. take a RAW photo at ISO 100 and +2 stops (adjusted to an effective ISO 400 <– okay, ISO 400-ish 😀), and compare that to a RAW taken at ISO 1600 and -2 stops adjusted (again ~ISO 400) - of course both taken with same aperture, shutter speed, no noise reduction, sharpening, contrast adjustments, etc..❶ Then continue from there.

If I understand what you are saying correctly(and forgive me if I'm not), what you're basically referring to is the concept of an "ISO-less" sensor. In such a sensor, missing a lot of more technical talk, essentially the end result is identical whether you set the ISO in-camera or shoot at base ISO and boost the exposure in post processing.

The behavior seems to have really started being documented/explored in the 36mp Sony-fabbed sensor used in the original A7R, Nikon D800(later D810) and if I'm not mistaken the Pentax K1. The Sony 45mp sensor that Nikon introduced with the D850 and uses a derivative of in pretty much everything now(save for the Z6/Z6II and D6, and D780 in full frame) also demonstrates this behavior well, although the D850 is(splitting hairs) still the best. The Z8/Z9 stacked sensor gives up a bit here too as the read speed adds noise and reduces dynamic range. Also, I have found the 16mp sensor(not sure who makes it) used in the D4/Df to show this behavior relatively well-the D5 sensor much less so.

It's actually a freeing way to shoot. I have been known at night to set my Df to base ISO, pick a shutter speed that will keep motion blur under control, pick my aperture based on desired DOF, and then just go about my business. You can't see the results, but it's kind of amazing when you start working the levels in Lightroom and see an image emerge. Aside from just not having to think about it, the nice thing about this technique too is that night scenes are often very high contrast, and this technique lets you preserve well lit details and still see the full scene. I need to play with my X-T5 and see just how well it handles doing this as it is a camera almost perfectly suited to this style shooting.

 

[herbert7265]

@bunnspecial

You may want to have a look into this website, it may help to understand your X-T5 (and any other camera) better regarding the ISO behavior:

Photons to Photos

Herbert

 

[Slartibart]

@bunnspecial

You may want to have a look into this website, it may help to understand your X-T5 (and any other camera) better regarding the ISO behavior:

Photons to Photos

Herbert

what an excellent source. Thank you!

 

[herbert7265]

Terminology can certainly be important, of course, but at the end of the day, it's the outcome that matters.

Yes, it’s the outcome that matters, but to get a desired outcome it needs understanding, and a correct understanding (!) is also based on a correct terminology. But maybe that’s just me, old style…!

Herbert

 

[r.harris1]

Yes, it’s the outcome that matters, but to get a desired outcome it needs understanding, and a correct understanding (!) is also based on a correct terminology. But maybe that’s just me, old style…!

Herbert

I definitely don't disagree, I just think that any message depends on the audience. I don't expect the exposure triangle concept to go away anytime soon, again (to me) because it's how camera manufacturers project it. We can all watch the ISO, the aperture, the shutter speed change if we're in "all auto mode" (which hopefully we never are!). We're (again, to me) taught that they're all related to what constitutes a "correct exposure" and that when we have low light, we need to crank that ISO if we're going to "capture the shot". We've been taught (intentionally or not) that if we see "noise", it's because we cranked that ISO, and to some degree it's true, because noise gets amplified with signal gain in a proportional way.

While I don't see the exposure triangle going away - for various reasons (dubious or not) - I do think that if we're serious about exposure, learning to read the histogram, understanding what your camera's limits are (all cameras have them) and at least knowing that the in ISO (as in Standard(s)) is very malleable in the digital world, can all be a very good thing. To me a good end goal is: Knowing how to keep as much shadow detail as you can, with minimal noise in those shadow, without blowing the highlights (at least the non-specular ones) in any given situation on any sensor.

At the end of the day, we all want the best images with the least noise, the most flexibility in post and obtained in as straight-forward a way as possible.

 

[Darmok N Jalad]

When it comes to low light photography, it's all about taking in what you have for light. Historically, you did this by having the most surface area available to absorb that light, but today, OIS and IBIS allow you to increase light gathering over a greater interval. If freezing fast motion in low light isn't an objective, then really good IBIS will allow you to use a lower ISO by allowing for much slower shutter speeds. Generally, smaller sensor cameras have better IBIS, simply because there's not as much sensor mass to stabilize. This isn't a fantastically sharp example, but consider the image below, taken on a M43 sensor, using a shutter speed of 3.2s, handheld, while standing in the ocean with the standard ocean breezes. This was basically taken at night. The streaks you see are airplanes. I did use noise reduction in post, which again, is a modern game-changer, but you get the idea that it can be more than just sensor size.

I should also add that low light is not the same as poor light. Poor light means you won't get much detail regardless of what you use. Early morning and late evening really aren't much of a challenge for any decent camera made in the last 5 years or so, IMO. Where you really start to see things differentiate is when attempting to freeze fast motion, with tracking and burst rates and such. If you go seeking the best there, be prepared to drop a lot of money!

 

[Ledgem]

I'm a month late to this topic but it's one of my favorites, so I'm going to reply.

Make the upgrade.

Not because you're going to see a significant difference, especially not for the times of day you're shooting at - although you'll probably think you will, thanks to the placebo effect - but because if you don't experience "full frame" for yourself, you'll always be left wondering what you're missing out on.

I've always been a 4/3 shooter (now µ4/3), which was derided for being a "small, noisy" sensor. Eventually the talk about larger sensors being far superior got to me and I added Fuji's GFX lineup to my bag, first with the GFX 50S and then the GFX 100S. The bigger the better, right? I found that most of what is posted online is overstated. For one week I'd go out shooting random things, choosing either my Fuji 50S or my Olympus E-M1 MkII at random, and at the end of the week I processed all of the photos together. I hid the file names and EXIF data so that I wouldn't know which camera I had used. "Look at how beautiful this is - the power of a larger sensor!" I would think to myself, looking at dynamic range or tonal gradation. Then I would unblind that photo and find that it was taken with the µ4/3 camera, instead. It was rarer that I saw a flaw and felt it must be the µ4/3 camera when it was actually the GFX, but that happened, too. I concluded that we're very good at fooling ourselves with visual things. I also found it amusing how µ4/3 and APS-C were routinely told that the larger "full frame" sensors would be far superior, yet "full frame" shooters commented on the GFX medium format by saying that it didn't seem to be significantly different and wasn't worth the cost. It's illogical and showed that it never really was about sensor size, but about conforming to "full frame" as a standard.

That's not to say there was zero difference, or that I'm running two systems just for the sake of spending money. Dynamic range really is better on the GFX system, although you can still blow highlights. And when it comes to noise performance, the GFX system has about a two-stop advantage over µ4/3 (ISO 6400 on the GFX looks like ISO 1600 on µ4/3). Yet the µ4/3 camera remains my low-light camera, because it has f/1.2 lenses, shooting at f/1.2 still gives me a usable depth of field, and it has a night mode that Fuji has seemingly not yet implemented (refresh rate on the EVF slows down so that you can see even in pitch-black; updates might be slow and it can make manually focusing take a long time, but you can still see). There's more to a camera system than just the sensor size.

How modern a system is makes all the difference, too. The criticisms of 4/3 in the past weren't totally wrong: with what I was shooting about 15 years ago, going above ISO 800 was pretty painful. There probably was a large performance difference with "full frame" systems of the time. Now with µ4/3, I don't hesitate to use ISO 6400 - and performance is likely superior to "full frame" cameras from 15 years ago. The sensors have gotten better, and noise processing has gotten better. The GFX 50S and 100S may not be the ultimate in low-light performance, but it goes to show that the differences based on sensor size are not as great as they used to be - if they really were significantly different in the past, at all.