[jsherman999]

Can we at least agree that Bill & Ted is an awesome movie?

If you like Bill & Ted then you are probably an awesome person and I should apologize for any snark directed your way. Have a good day, dude.

[clackers]

Overall image noise can be broken down into photon shot noise + read noise.

JS, you're obviously an okay guy because you also love Bill and Ted (I even have time for the sequel), but if you're getting your stuff from Clarkvision.com, he's unambiguous:




His words are: "The ultimate in high signal-to-noise ratio, high dynamic range, and high ISO performance would be one large pixel ... When choosing between cameras with the same sized sensor but differing pixel counts, the one with larger pixels (and fewer total pixels) will have better high ISO and low light performance (assuming read noise and fixed pattern noise are similar, which may not be the case)"

[clackers]

I'm not sure what set of assumptions and requirements you're using, so you'll have to spell it out to me for me to be able to explain where your misconception is.

Such a comedian, EJ!

You seem to have the same misconceptions of Marcus (and JS briefly) that were laid to rest in the https://www.pentaxforums.com/forums/169-pentax-full-frame/275243-pentax-dslr-...ull-frame.html thread.

We'll work out exactly how much of the Total Light Kool-Aid you've swallowed.

You and I are shoulder to shoulder.

You've got a 645Z, I've got a tiny, tiny Q.

I meter at 1/125s, ASA200, f8.

You set to 1/125s, ASA200.

Now, what should your aperture be?

(The Q for your Total Light purposes has a crop factor of 5.6, the 645Z of 0.79. Let's say I have a zoom currently at 10mm, and you have a 55mm prime.)

Can't be fairer on your theory than this.

[ElJamoquio]

Your question shows a fundamental misunderstanding of the people you're responding to.

[clackers]

Your question shows a fundamental misunderstanding of the people you're responding to.

That's not an answer to the scenario outlined - which, remember, was something you offered to do.

[Fogel70]

You've got a 645Z, I've got a tiny, tiny Q.

I meter at 1/125s, ASA200, f8.

You set to 1/125s, ASA200.

Now, what should your aperture be?

(The Q for your Total Light purposes has a crop factor of 5.6, the 645Z of 0.79. Let's say I have a zoom currently at 10mm, and you have a 55mm prime.)

Can't be fairer on your theory than this.

What do you want to prove with this?
You can set f8 on both, then the Q will use about 2% of the total light the 645Z use for the exposure.
To be able to use same amount of light for both cameras with different sensor size, you need to use different exposure.

[dakight]

JS, you're obviously an okay guy because you also love Bill and Ted (I even have time for the sequel), but if you're getting your stuff from Clarkvision.com, he's unambiguous:

His words are: "The ultimate in high signal-to-noise ratio, high dynamic range, and high ISO performance would be one large pixel ... When choosing between cameras with the same sized sensor but differing pixel counts, the one with larger pixels (and fewer total pixels) will have better high ISO and low light performance (assuming read noise and fixed pattern noise are similar, which may not be the case)"

Well that's a little extreme, and actually untrue. A pixel represents one discrete digital packet of the image and it has zero dynamic range. I understand what he's trying to say, but taken to that extreme it's nonsense.


The other factor not being considered is magnification. The original image might in fact have a stellar signal to noise ratio but noise is magnified by the same factor as the enlargement and will become very noticeable at high magnification... just like grain in a film image. A sensor with larger and fewer pixels may well have less noise but at high enlargement factors that noise will be very evident.


At extreme enlargement the image will appear patchy and grainy, and digital artifacts will be more evident. A sensor of 16 or 24 megapixels will produce images that will stand up better than one of 10 or 12 megapixels if the final image is to be greatly enlarged. For that reason a FF or even medium format sensor of equivalent pixel density will always give better results.


Now, I don't know all the theory and underlying equations regarding total light, noise, etc., etc., but I know from practical experience that all else being equal a larger captured image will maintain it's integrity better at higher enlargement factors. Whether that is significant depends on the user and what the user is trying to accomplish.


The whole argument is moot if all you're interested in is digital images on an electronic display medium or 4x6 prints from the local WalMart, but if you're doing commercial work with large final output or if fine detail at high magnification is critical then it will be significant and you will have to chose your equipment accordingly.

[Rondec]

I don't understand all this ado about nothing. People tend to argue extremes. What happens on a pixel level is mostly unimportant. Pixel sizes are often pretty close between formats. In most photos, where you are printing/viewing really large, you won't see individual pixels. Size of pixels should have more to do with dynamic range, than with visible noise. Bigger pixels should just have deeper wells.

As I said before, the biggest thing is how much you are magnifying the image you get from a sensor. The more you magnify it (never mind how many megapixels you have), the more you will see noise, lens flaws, etc.

[jsherman999]

JS, you're obviously an okay guy because you also love Bill and Ted (I even have time for the sequel), but if you're getting your stuff from Clarkvision.com, he's unambiguous:

I've referred to him before, he seems to explain things very clearly.

His words are: "The ultimate in high signal-to-noise ratio, high dynamic range, and high ISO performance would be one large pixel ...

I've said this exact thing before, in the previous thread. Take the SOA pixel tech, apply it to one pixel, you get the best SNR possible. And you have a 1-pixel image to share!

When choosing between cameras with the same sized sensor but differing pixel counts, the one with larger pixels (and fewer total pixels) will have better high ISO and low light performance (assuming read noise and fixed pattern noise are similar, which may not be the case)"

Note the "assuming read noise and fixed patter noise are similar, which may not be the case." He's not claiming a generic "larger pixels automatically mean lower noise", he's saying hat given the exact same pixel tech, you will probably see less image noise as you lower the pixel count.

There's a trade-off. I think Nikon found that taking their D800 pixel tech down to 12MP would gain them about 1/4 to 1/3 of a stop (thogan) - not enough to make it worthwhile from a marketing perspective.

The noise performance of the D800 already beats the D700 - I wouldn't buy a 12MP D800, losing all that resolution, just to get that extra 1/4 - 1/3 stop... would you? If I were *that* concerned about noise alone, to the point I'd be wiling to give up a ton of resolution I'd buy something like a D4s. And we see that going just from 36MP to 24MP gained *no* SNR in that comparison.

What happened between say 2004 and 2012 was people saw different things happening with sensor MP and pixel pitch and attributed gains directly to those changes, when really it was a combination of engineering things that those companies did that can't be attributed to pixel size alone - yet that's what people saw, so that's what they attributed it to. As in for example, "Canon 5D II has 21MP and worse ISO performance tha D700 which has 12MP, thus lower pixels are the causal link - less MP = always less noise!" when really the underlying pixel technology was better in that D700 as well.

Back to Dr Clark - he's a good source I think. He uses "True Exposure" in place of "Total Light", but he's describing the exact same thing here.. I like Total Light better, because from past experience a term like "True Exposure" is going to make some people cranky, because it will suggest to them that they're using "fake exposure" and what they do is being discounted,. Kinda lie the "Full Frame" label makes some people cranky. So I like Total Light, but True Exposure works.



(quoted below in case the above is hard to read)


"Exposure, Light Meters, and Digital Cameras

Discussions I have observed and participated in on the internet shows that many photographers do not actually understand exposure, including amateur, advanced amateur, and even pro photographers. Along with exposure confusion, there also seems to be confusion about light meters. With this article, I'll try and clarify exposure, and make a very important distinction between actual light collected during an exposure, versus camera exposure. Metered light, and camera exposure versus actual light (true exposure) recorded are usually quite different. It is different between camera formats (sensor size), pixel size, and even different focal length lenses of the same f/ratio.

Why is this important? In making choices between lenses and cameras understanding these concepts may allow one to make better choices, especially when pushing photographic limits, as in fast action+low light photography, including baby's first steps in a dimly lit room, to wildlife action near sunrise/sunset, to things like nightscape photography.

Definitions

Camera exposure: The relative exposure recorded by a camera. In a digital camera, this is how full a pixel is to a given reference level. Smaller pixels fill with fewer photoelectrons than do larger pixels. As ISO is raised, the reference level is reduced. For example, at ISO 200, the reference level is half the photoelectrons as at ISO 100. Thus, a "properly" exposed digital camera image records half the light at ISO 200 as at ISO 100.

True exposure: the actual amount of light recorded, e.g. expressed in photons (or photoelectrons) on the subject. Remember, the subject is what we are concerned about in our images, not pixels.

There is a significant distinction between the total light recorded (true exposure) from an object and the camera exposure on an object. Image quality is directly related to the amount of light collected by a camera and its lens. Exposure is about relative density of signal (density on film or relative signal level in a digital camera pixel). With film, the relative density is controlled by the film speed and the amount of development. On a digital camera, the exposure is a relative level determined by a post sensor gain level. Actual light recorded is dependent on aperture diameter and the angular size of SUBJECT on the sensor (film or digital).

There are digital cameras of many sizes, with different sized sensors and different sized pixels, and each can be set to a large range of ISO. Camera exposure may appear to be the same, but true exposure is different in these cameras with differing sensor and pixel sizes and even at the same ISO."

[Dartmoor Dave]

Thus, a "properly" exposed digital camera image records half the light at ISO 200 as at ISO 100.

I know I said yesterday that I was going to bow out of this thread. . . but c'mon! You can't get away with that nonsense. Innocent bystanders might get hurt.

A digital camera does NOT record half the light at ISO200 than it does at ISO100. It applies double the gain to the analog signal output from the sensor. The amount of light hitting the sensor is the same, given the same aperture and shutter speed.

I think I finally understand the notion of "Total Light" now. You guys think that the ISO selected in a digital camera is something to do with exposure, but it isn't. Aperture and shutter speed determine exposure in a digital camera, in the same way they've always done. The ISO rating that you select merely adjusts the analog gain stage downstream from the sensor.

[Kunzite]

That article is a nightmare and it's mixing everything in an indistinguishable mess. MHO (just to be clear, I don't claim I could write clearer articles!).
Exposure is light flux per unit of area, measured in lux second. His "Camera exposure" appears to be something different... why? is he trying to talk about "light per pixel" and "light per full sensor's area"? An interesting approach; but what's the purpose?

And I cannot agree with his definition of "true exposure". Exposure has a precise definition; multiplying it with area does not make it "true" - you're getting something different. Just like multiplying speed with time, you won't get "true speed" - but distance.

Be careful, very careful about what you're reading on the Internet.

[ElJamoquio]

You and I are shoulder to shoulder.

You've got a 645Z, I've got a tiny, tiny Q.

I meter at 1/125s, ASA200, f8.

You set to 1/125s, ASA200.

No, I don't, presuming (big assumption) that I want to take a similar picture as you.

I set to 1/125s, ISO~1600 or so.

[Dartmoor Dave]

No, I don't, presuming (big assumption) that I want to take a similar picture as you. I set to 1/125s, ISO~1600 or so.

You set that at f/8?

I guess you don't want to take the same picture as Clackers. You want to take one that's three stops overexposed.

[ElJamoquio]

You set that at f/8?

I guess you don't want to take the same picture as Clackers. You want to take one that's three stops overexposed.

No/Reread post/No.

[ChristianRock]

Apparently EJ wants to take the pic at f22. Why, I have no idea. Because he loves equivalence, I guess. And diffraction.