And they claim 5D Mark III has a poor dynamic range.

RTPVid wrote in post #14694618
Unmanipulated?

As I understand it, yes...

Different makes respond differently to highlight recovery and ability to push the shadows so your real DR is how much is there at the time of capture.

I mean, you don't use GND filters to get certain levels for post processing. You use GND filters to put the DR of a given scene within the specs of your camera.

To put it another way, if Camera A requires a .6 GND to avoid blowing the highlights and Camera B only needs a .3 to do the same, then, IMO, Camera B has better (native) dynamic range.

fogboundturtle wrote in post #14694475
But it would be out of focus

But that's only if you're using the left side of the AF sensor. (allegedly)

I'm hoping a lot of people jump ship from Canon to Nikon actually. Because only then, will it light a fire under Canon's bums to really try and improve their sensor design.

In the meantime, I'm more than happy shooting with my substandard DR cameras as is. It's very safe to say that *I* am the weakest link, and not my gear. That extra 2-stops at iso100 isn't going to do *me* any good. It'd be like putting lipstick on the pig, the pig being me.

jwcdds wrote in post #14693970
But you see...

Had the same shot been taken with a D800... instead of resolving a silly little bird, you'd have resolved a bald eagle.

and the D800E would have given that bald eagle magnificent flowing locks of hair

If it were a black bird on a white cloudy background and you could pull back the shadows and not blow the sky that would be great dynamic range, which is what the D800 can do. This is just changing exposure in post. A dark scene made brighter. Not a dark and bright scene wihtout clipping.

Nice test though, any NR applied?

Nice result.

FlyingPhotog wrote in post #14694639
As I understand it, yes...

Different makes respond differently to highlight recovery and ability to push the shadows so your real DR is how much is there at the time of capture.

I mean, you don't use GND filters to get certain levels for post processing. You use GND filters to put the DR of a given scene within the specs of your camera.

To put it another way, if Camera A requires a .6 GND to avoid blowing the highlights and Camera B only needs a .3 to do the same, then, IMO, Camera B has better (native) dynamic range.

But, here's the rub... define "native".

How much of the in-camera manipulation is considered part of "native"? Given the power of the in-camera image processors, what, exactly IS "native"? And, in the digital imaging context, does it even matter?

DXO (supposedly) tries to measure the "bare" sensor, but is that a test that anyone should care about outside of the vendor's camera design team? Which is better for the photographer, a camera with a sensor with some superior characteristics "natively" or a camera with an image processor with some superior characteristics "natively"? Should a photographer even care?

In the film days, it was possible to independently test the "contributions" of the lens, the camera body (such issues as film plane flatness, for example), film processing, print darkroom technique, print paper characteristics, etc. and make judgments about lens resolving power (compare all lenses from all manufacturers using the same subject and the same film with the same film processing with the negative examined with a lupe). That cannot be done anymore. Even lens comparisons across manufacturers are dicey, since nothing behind the lens remains constant across the tests.

In that context, why is image manipulation within the camera ("native" performance) "good" and image manipulation in post "bad"? Can't you even make the argument that minimal in-camera manipulation is better (or, at least, the manufacturer gives the photographer the option of turning it off)?

In the film days, the DR of transparency film was a big deal since there was virtually nothing to be done to recover lost detail. As the OP apparently demonstrated, such is not the case with the 5DIII's sensor. In the end, does it really matter if the gazillion stop DR was achieved in post or in-camera?

Neilyb wrote in post #14695378
Nice test though, any NR applied?

none

"The dynamic range measurement shows improved performance behavior at high ISO, but is disappointing for the limited dynamic range at low ISO."

Above quote from DXOMark. But I guess "disappointing" is better than "poor." And from pictures posted here I disagree with either word; saying that the D800 is better in low-ISO DR does not mean the III is poor.

Recovering from that exposure won't look good with any camera.

I like the first photo

FlyingPhotog wrote in post #14693983
The ability to "save" an image is not the textbook definition of "Dynamic Range"

DR = The camera's ability to resolve as many shades between 255,255,255 (featureless white) and 0,0,0 (featureless black) as possible in one frame...

Exactly, what the rescued image shows us is NOT dynamic range in any shape or form.

Canon are fond of quoting "11 plus stops", which in theory they can achieve. However the top third and bottom third are almost horizontal and so contribute nothing, only the middle third has a useful gradient and its around 5 stops maximum.

Another one of these threads that's beating a dead horse. Who would print that crop anyway?

jdizzle wrote in post #14697116
Another one of these threads that's beating a dead horse. Who would print that crop anyway?

Print is so pre-circa-2000.

jwcdds wrote in post #14697169
Print is so pre-circa-2000.

Yeah, most people store images on hard drives nowadays. Who wants to waste paper?

jthomps123 wrote in post #14693786
Ok it's settled. Guess 'they' were wrong... Who are 'they' again? And when did they say what you claim?

OTHERS