Exposure puzzle

I like the white balance better on the first one, whites more natural.

D610 is more accurate (digital?) - D700 more of a yellow / magenta tint but I would attribute this to colour profiling on top of the withe balance of course. nevertheless both behaved pretty much similarly under fluorescent light.

You are possibly causing an unnecessary complication by using matrix metering, I'm pretty sure two cameras separated by several years of development will yield different results on occasion.

I am just trying to understand as to why the D700, an otherwise excellent cam and still relevant today, is perceived to overexpose by countless users incl. myself.

As previously mentioned, perhaps the tone curve is different or interpreted differently.

I had two D3 bodies (same sensor) and they never deviated more than 1/3rd of a stop between bodies.

In the case of film, the ISO standard (and previously ASA) gave a somewhat

unambiguous way to rate film. It might depend a little on the developer, though.

In the case of digital cameras, though:

Film speed - Wikipedia

it seems that the ISO gives five different ways to rate them!

So you could have five different cameras, with five different values, all claiming ISO compliance!

In the case of film, there is a curve, which usually is a curve (no straight section) that measures

the sensitivity. There is a little margin on each end, as it starts to flatten out.

With digital, it saturates at either end, no slightly flattening.

Hmmm. I notice the CT and tint settings are different as well. In both examples.

So the sensors would appear to have different 'neutral' white balances in each camera. Therefore at varying ambient light colours they'll require different exposures to keep the RGB channels in balance. But 1 stop difference? Unlikely.

The other variable is the comparative ISO sensitivity of each camera. And since there's no objective test set down by the ISO for digital camera sensors and their associated amplifier circuitry, that leaves the field wide open for differences as well.

Easiest option is to stop trying to figure out Nikon's rationale for making the D700's metering so flakey. Just accept that it is. Set the metering fine (coarse?) tuning to -2/3rds of a stop and put up with it.

With digital, it saturates at either end, no slightly flattening.

But those ends are much further apart than with film - about 11 stops apart on a D700. And the histogram shows end-stopping at least one stop before a RAW file really becomes unrecoverable.

After shooting slides many years, I tend to dial in -1/x exposure compensation almost always without worrying about it too much. I think I would rather have a bit of noise in the shadows than blown highlights. I think this is contrary to "exposure to the right" but I don't care.

My experience is that some lenses just seem to need exposure compensation to match others, especially non Nikon lenses. It is easier to deal with a body that is consistently off a little than occasionally used lenses that need compensation. Guess I should add a label on those lenses to remind me. I check the iris action on these lenses, seems OK, who knows.

 

(snip)

 

The other variable is the comparative ISO sensitivity of each camera. And since there's no objective test set down by the ISO for digital camera sensors and their associated amplifier circuitry, that leaves the field wide open for differences as well.

 

(snip)

There is a quote, which I believe I first heard related to computer network protocols:

"The nice thing about standards is that we have so many to choose from."

As above, it seems that there are five different standards for ISO digital camera ratings.

And five is always better than one. :)

As above, it seems that there are five different standards for ISO digital camera ratings.

I've always thought that standards should be readily available and free to study by any interested party, but since those greedy Swiss gnomes at the ISO have taken over from the ASA, BSI, DIN, etc., they've decided to charge a substantial sum to read any ISO standard. I notice this fee has now risen to over £90 UK / $120 US. Disgusting!

They also obviously scour the net for any infingement of their 'copyright'. But since they displaced our individual national standards organisations, the lazy sods haven't come up with any standard worth a damn anyway.

OK. Rant over.

I think this is contrary to "exposure to the right" but I don't care.

Not really. "Expose To The Right" or ETTR is a bit of a misleading phrase. It should actually be "Expose just short of the right", such that the histogram doesn't exceed the right-hand side of the graph. That way you get the most shadow detail while retaining all highlights.

It should be technically easy to set up any modern digital camera to do this automatically. So why many camera makers don't is a complete mystery.

I've always thought that standards should be readily available and free to study by any interested party, but since those greedy Swiss gnomes at the ISO have taken over from the ASA, BSI, DIN, etc., they've decided to charge a substantial sum to read any ISO standard. I notice this fee has now risen to over £90 UK / $120 US. Disgusting!

 

(snip)

Yes. The IEEE has released some of the Ethernet standards for non-commercial use.

There is access to the NEC. (US National Electrical Code), again free for non-commercial use.

(It generates a watermarked PDF when you access it, so they can track you down if you leak.)

Some computer standards are available from the ECMA (European Computer

Manufacturers Association) for free, even when the ISO or US version is expensive.

In the case of some ISO standards, at least the ones for Fortran and C, the last of

the draft versions before the finalized one, is usually available free. Normally there

are no changes, other than section numbering.

But yes, many IEEE standards are expensive and would be useful to have,

even when we aren't selling anything based on them.

I am just trying to understand as to why the D700, an otherwise excellent cam and still relevant today, is perceived to overexpose by countless users incl. myself.

The measuring of available light in the camera is done by different light-sensors, in a different housing, followed by differnt sensors to register the image, fololwed by different electronics with diferent programs to translate the light on the imaging sensors by different programs, then the program that translates the image information on your computer into colours and lightlevels to show you a recognizable image on your screen even uses a different protocol with its own settings depending on the source camera..

So... how would it be even possible to get exactly the same resulting image? ;):cool:

Gents - truly humbled by your engagement and expert input.

wrt. "true" ISO diff by an exact14! presumably by such reputable testing outlets such as D**M*** - by me extremely rough calcs this would equate to 1/14th of a stop would it not ? (do pardon my ignorance). I am truly stuck and willing to be as desperate as to question the "real" shutter speed readout but that's another story - or - is it?

In the days of film, metering mattered muchly....:)

In the days if digital, with instant critical feedback, not so much....:p

I've always thought that standards should be readily available and free to study by any interested party, but since those greedy Swiss gnomes at the ISO have taken over from the ASA, BSI, DIN, etc., they've decided to charge a substantial sum to read any ISO standard. I notice this fee has now risen to over £90 UK / $120 US. Disgusting!

 

They also obviously scour the net for any infingement of their 'copyright'. But since they displaced our individual national standards organisations, the lazy sods haven't come up with any standard worth a damn anyway.

 

OK. Rant over.

 

Not really. "Expose To The Right" or ETTR is a bit of a misleading phrase. It should actually be "Expose just short of the right", such that the histogram doesn't exceed the right-hand side of the graph. That way you get the most shadow detail while retaining all highlights.

 

It should be technically easy to set up any modern digital camera to do this automatically. So why many camera makers don't is a complete mystery.

The worst thing is that if I were to pay them $120 I can't tell you what's it all about either. They won't allow me to do this. CIPA does list a standard for sensor sensitivity.

Since 5 is better than 1 I decided to define my own standard.

Gents - thanks again and I'll call it a day as your divergence into ISO standards etc. is certainly beyond my primitive comprehension and my reason for initiating this thread. I'll just live with the D700 and appreciate what I could claw back regardless (J.R. below before and after wrongly spot metered in a hurry thinking"tis dumb-dumd again with the blackbox!"). Salutes to all gearheads out there.

I often photograph horses that are a very similar colour to your canine friend. I'm always surprised by just how dark they really are and the meter doesn't agree.

Sometimes I think dark brown somehow appears bright to the camera! Maybe it's their shiny dark fur.....? :D

I often photograph horses that are a very similar colour to your canine friend. I'm always surprised by just how dark they really are and the meter doesn't agree.

 

Sometimes I think dark brown somehow appears bright to the camera! Maybe it's their shiny dark fur.....? :D

he’s actually black&tan and a nightmare to shoot- spot meter and the whole scene horrendously overexposes. Matrix meter and he just seamlessly blends in the shadows and disappears. Must shoot raw.

he’s actually black

Apparently that's the main reason more black cats remain in rescue centres than any other colour.

They're hard to take pictures of for social media posts!

They're hard to take pictures of for social media posts!

Thank goodness for that!

Otherwise the Internet would definitely break from the googleplex of extra cat images posted every half-hour.:cool:

What? Cute little me... evil?

How dare you!?