[Wheatfield]

The point of that part of the exercise was to illustrate, anyway, that megapixels sometimes *do* matter. If you have a 12MP sensor, and I have a 16mp sensor, we have the same FOV and good lenses, and we both light our scene to sufficiently exceed the noise floor, then print them both full-frame on 8x10, the 16mp sensor will RECORD MORE DETAIL, even if it's a 16x24mm.

A friend and I were having this discussion today (we laugh about a lot of what comes up on forums).
Came to the conclusion that what mostly affects IQ these days is megapixels, plain and simple.
Pretty much everything else is either window dressing or finding some small advantage that one format or the other does better and holding it up as the holy grail.

[jsherman999]

A friend and I were having this discussion today (we laugh about a lot of what comes up on forums).
Came to the conclusion that what mostly affects IQ these days is megapixels, plain and simple.
Pretty much everything else is either window dressing or finding some small advantage that one format or the other does better and holding it up as the holy grail.

Let me guess: By the fourth beer, dynamic range was out the window, by the eighth, high-ISO performance was for weenies.


.

[Wheatfield]

Let me guess: By the fourth beer, dynamic range was out the window, by the eighth, high-ISO performance was for weenies.


.

Coffee, actually. And we were discussing the present state of the art, not what was available two years ago.
The technology is to the point now that dynamic range and high ISO performance is almost a non issue.
Seriously, we have over 14 stops of dynamic range available to us, and usable high ISO performance in excess of ISO 12,800.
They will improve upon this stuff because the manufacturers have to do it lest we stop buying their equipment, but we are at the point now where we have to start looking fairly hard to find stuff that our cameras can't take a good picture of.
The thing we can always use is more pixels. Pixels are what resolve fine detail.

[jsherman999]

Coffee, actually. And we were discussing the present state of the art, not what was available two years ago.
The technology is to the point now that dynamic range and high ISO performance is almost a non issue.
Seriously, we have over 14 stops of dynamic range available to us, and usable high ISO performance in excess of ISO 12,800.
They will improve upon this stuff because the manufacturers have to do it lest we stop buying their equipment, but we are at the point now where we have to start looking fairly hard to find stuff that our cameras can't take a good picture of.
The thing we can always use is more pixels. Pixels are what resolve fine detail.

Ah, I see, makes sense. I don't agree on ISO performance, though. Ever-more-extreme ISO ceilings brings extremely clean ISO performance down at the levels you're most likely to be shooting at. Current 100 ISO noise levels at ISO 1600, ISO 400 noise at ISO 6400? I'll take it. No more handshake blur, no more subject-movement blur, under almost any typical low-light shooing situation.

Interestingly this would make VR and SR less crucial.

It also would go in-step with the increasing MP you wish for - those sensors will be even more unforgiving to handshake blur.


.

[Wheatfield]

Ah, I see, makes sense. I don't agree on ISO performance, though. Ever-more-extreme ISO ceilings brings extremely clean ISO performance down at the levels you're most likely to be shooting at. Current 100 ISO noise levels at ISO 1600, ISO 400 noise at ISO 6400? I'll take it. No more handshake blur, no more subject-movement blur, under almost any typical low-light shooing situation.

Interestingly this would make VR and SR less crucial.

It also would go in-step with the increasing MP you wish for - those sensors will be even more unforgiving to handshake blur.


.

Absolutely, I'll take it too, for the very few occasions where I find myself shooting in almost total darkness.
However, if you have been paying any attention, you will have noted that people are spending a lot of effort on finding ways to make the equipment trip up at the extremes rather than using said extremes for making meaningful pictures.

[jstevewhite]

Absolutely, I'll take it too, for the very few occasions where I find myself shooting in almost total darkness.
However, if you have been paying any attention, you will have noted that people are spending a lot of effort on finding ways to make the equipment trip up at the extremes rather than using said extremes for making meaningful pictures.

I've been shooting at the edges. Well, using ISO 800-1200 for macro is pretty edgy for me. I'm used to 100-400 w/ my K20D. I can't say how MEANINGFUL the images are, though...

I get involved in these long arguments because I work in front of a computer and I type really fast...

[jsherman999]

Absolutely, I'll take it too, for the very few occasions where I find myself shooting in almost total darkness.

.

It doesn't have to be total darkness at all.

(Rupert, if you're reading this, I'm going to borrow a link to one of your squirrel shots.)

In the shot below, Rupert did a pretty good job handheld at 500mm, 1/125s, ISO 3200. Noise looks OK, like a modern ISO 3200 in fairly good light.



One wonders how his keeper ratio is at 500mm, 1/125s, handheld, though. I'm a pretty steady shooter, and that would be a problem for me. I imagine I'd blur a good percentage of shots, maybe missing a real great one here and there.

But if he could take that same shot at 1/500s, with the equivalent noise of (current) ISO 400, that would represent a 5 stop improvement in ISO. And that's not a shot of a black cat in a coal mine, it's not a science-fiction scenario, it's just an appropriate shutter speed and a good clean image in a slightly-lower light situation, outdoors.

I don't know about you, but I'm shooting handheld telephoto at those modest EV levels all the time, and I'm also shooting shorter FLs at EVs 3, 2, and 1 quite a bit too - with lots of subject movement. A 4 or 5 stop improvement over current isn't something I'd have any trouble finding a use for on day one.

.

---

Last edited by jsherman999; 05-18-2011 at 10:48 PM.

[jstevewhite]

DOF test... stitched from six or eight images, ~50mpixels after merge and crop. About nine minutes of PP (stitching and adjusting) in this image. With practice I could get that down to three or four minutes, I'm pretty sure... It's dull but sharp, sharp, sharp - a result of clouds and poor PP adjustment due to lack of practice. I'm pretty sure if I give it a little zing, I could print a 24x24 @ 300 DPI.

Daughter looking hard at me for making her guinea pig for me.



This isn't a "rebuttal", so much as a "more than one way to skin a cat".

Obviously, if you're looking for that razor thin DOF with action, you're going to have to go with FF, at least until someone breaks down and makes a 75mm f1.0 for our APS-c machines.

[Kerrowdown]

there is no significant difference besides DOF between FF and APS-C

Oh yes, size matters.

[Laurentiu Cristofor]

Absolutely, but to achieve that you have to drop back to a lower effective pixel count; you might as well use the larger sensor sites and a lower MP sensor.

For the purpose of the equivalence comparison, it doesn't matter whether you have more pixels or larger ones, because you'll compare images of the same size. You may find this article on equivalence useful - I got the gist of the theory from falconeye, but I never had the patience to continue by reading this article.

But I agree that in practice this theory of equivalence doesn't feel that useful to me. It's not IQ issues that cause me to throw away shots, but a bunch of other issues that I'd have on FF as well as on APS. And I'm fine with APS being able to produce images that look as good as crops from a FF camera - it's not like FF users never crop their shots because DR is slipping through their fingers.

[peasant]

And that's not a shot of a black cat in a coal mine, it's not a science-fiction scenario, it's just an appropriate shutter speed and a good clean image in a slightly-lower light situation, outdoors.

And for casual shooters like me this a total boon.Btw don't write off the cat/coal mine scenario just yet.

[jsherman999]

DOF test... stitched from six or eight images, ~50mpixels after merge and crop. About nine minutes of PP (stitching and adjusting) in this image. With practice I could get that down to three or four minutes, I'm pretty sure... It's dull but sharp, sharp, sharp - a result of clouds and poor PP adjustment due to lack of practice. I'm pretty sure if I give it a little zing, I could print a 24x24 @ 300 DPI.

Daughter looking hard at me for making her guinea pig for me.



This isn't a "rebuttal", so much as a "more than one way to skin a cat".

Obviously, if you're looking for that razor thin DOF with action, you're going to have to go with FF, at least until someone breaks down and makes a 75mm f1.0 for our APS-c machines.

.


Excellent job, OOF transition and background look very 'real' too, not shopped.

Now, if they only provided that feature in camera as bracketing option...


.

[Rondec]

For the purpose of the equivalence comparison, it doesn't matter whether you have more pixels or larger ones, because you'll compare images of the same size. You may find this article on equivalence useful - I got the gist of the theory from falconeye, but I never had the patience to continue by reading this article.

But I agree that in practice this theory of equivalence doesn't feel that useful to me. It's not IQ issues that cause me to throw away shots, but a bunch of other issues that I'd have on FF as well as on APS. And I'm fine with APS being able to produce images that look as good as crops from a FF camera - it's not like FF users never crop their shots because DR is slipping through their fingers.

That is a very good point. I don't know anyone who makes their decision to crop (or not) based on whether or not their depth of field will increase.

[Blue]

I'd have no problem shooting weddings or portraits again with the old *ist D, but for poster-sized prints and faster low-light focusing I'm happier with the K-5.

I use a *istD quite a bit (for other things) but its buffer is s-l-o-w. It makes the K20d seem fast. I Pentax releases a ff the size of the *istD, some people will poop their pants.

[jstevewhite]

I use a *istD quite a bit (for other things) but its buffer is s-l-o-w. It makes the K20d seem fast. I Pentax releases a ff the size of the *istD, some people will poop their pants.

I loved my *istDL, and I got some shots with it that I love, but I can tell the difference between the K-5 cropped to 8x10 and the *ist DL cropped to 8x10 (in subjects with a lot of detail). At 11x14 I can look at a single high-detail image (like one of my bug or flower macros) and tell you which camera it came from. That's my measure of "important difference" in IQ.

If I print full-frame on 8x10, there's no difference, because the *ist DL just makes 300 dpi for 10" - but I can crop the daylights out of the K-5 image and get the same quality of result.