[Anvh]

I've never read or heard about 12%. I just tried by aiming at a maximized Notepad on the monitor. Compensated with 1/2 stop and had the peak in the histogram rest in the exact middle.

Damn you internets

Edit: but now I'm confused - why doesn't the cam do this on it's own?

Edit2: Strike that - I'm getting very random results. Could it be my 3rd party focusingscreen?

Try live view, it should roughly give the same metering although it works differently.

I have a K-5. I didn't understand what matrix metering is. I just now read up on it and will try again with multisegment metering and setting the AE to the AF point. Just going to multisegment doesn't really tell me what the camera is going to do if it has 77 different solutions. At least I can't get that from manual.

Multi segment???

Any way.
- matirx metering looks at the whole frame and takes into account the focus point, camera orientation, light distribution and probably also things like focal length and maybe a lot more.
It has these "presets" so to say, so it knows how a landscape photo looks like and it tries to get a correct exposure for that to explain it in a simple way.
- centre weighted meters in the centre the most and the value goes down to further you get to the edge.
- spot metering meters one spot, you can set it to follow the focus point if you want.

Centre weighted is probably the best mode if you want to know what the camera is doing since the spot metering point is quite small so the value can jump up and down quite a bit.

[twitch]

In this situation, white birds against darker (and changing?) background I'd be using manual mode. Any semi auto mode is changing exposure on you depending on background.

[Zafar Iqbal]

Im sure there is a reason for this conflicting bit of info. Im not really intrested enough to delve into it if there is though lol

I did quick googling before and found a usercomment on some forum stating dedicated light meters were calibrated to meassure 12% while cam meters were calibrated for 18%. I tried googling on this (very quick again) but didn't find further info.

I was using my 10 year old sigma lens when I gave it a shot few monents ago. Could be the lens itself underexposes because I now recall I used it a lot during summer in Av mode and had the EV dialed down to -2 for most of the time, ranging between -1 to -3 depending on location and direction to the sun. I used matrix metering then.

I tried again with my latest 17-50 2.8 Sigma and shot different white surfaces in my kitchen. I had to compensate between 1/2 to 1 stop this time (in manual and using spot metering). The ratio seemed to be 50/50.

[Marc Sabatella]

I use M mode and set my own exposure. The green button gives me the same starting point (I always use center weighted meering) you'd get in other mdoes, then I just use "common sense" (ie, knowledge plus experience) to get an idea of whether I might need to bump or down. I also shoot RAW so I don't sweat a small amount of overexposure - it's usually recoverable.

[Zafar Iqbal]

Try live view, it should roughly give the same metering although it works differently.

It works as described. 1/2 stop compensation and I hit the middle of the histogram, but I had to be in AV mode. The histogram didn't change according to the shutter speed changes in manual mode.

Then tried manual again and same results as before. Worst cases (few of them) deviated by 1 1/2 stop.

Then tried manual but using with center weighted metering and nailed it every single time without any compensation. But I was aiming at flat bright surfaces so this conclusion, or finding rather, may prove not to be as exact in real work scenarios.

Marc: I find this subject to be very interesting. I used to dial in based on what I saw in LV, and only little based on the histogram. An shot can be very underexposed but still look OK if only looking at the LV.

I recently purchased a DVD were there's a talk about metering and how to learn to recognize various shades (in colors especially) and thereby be able to pre-dial more precisely without having to do as many test shots.

Before this I wouldn't "uh, white surface. Meter it and open up by 2 stops". This gets me in the ballpark and ultimately I'm aiming for what you do, but I only just recently started to *really* pay attention to the histogram.

[Ash]

Marc's on the mark with this point (sorry for the pun). It's getting to the point where you're comfortable making the exposure triad adjustments to get the exposure you want in your images. Of course, nothing will change the scene in front of you (high contrast scenes are always a challenge) but a simple measure such as a graduated ND filter might limit the need for HDR for very high contrast nature scenes with bright skies and dark foregrounds. But most scenes are good for the K-5 to be able to capture most of the detail within the K-5's dynamic range.

[Anvh]

I did quick googling before and found a usercomment on some forum stating dedicated light meters were calibrated to meassure 12% while cam meters were calibrated for 18%. I tried googling on this (very quick again) but didn't find further info.

I've read that Nikon engineer has stated that they are confirm with the ANSI standard, so that is 12% or at least around that number.
I've read that for example the sekonic light meters are around the 14% give or take.

How are you checking this btw?
Just take a photo of a white paper and make sure it fills the whole frame and set the metering to zero.
Then in photoshop or something like that take a RGB reading and translate that to a grey value, that way you see where it metered for.

or use the histogram in the camera or liveview, you see that the white paper will peak slightly to the left and not in the center of the histogram.
The camera meters for 12% so it wants to make something that kind of grey so it turns your white paper into 12% grey, if it was 18% it would be in ther center.

---

Last edited by Anvh; 01-03-2012 at 03:04 PM.

[Anvh]

and a sekonic 535 for the Mamiya RZ67. Both meters proved to be extrordinarly reliable in the exposure department even with slide film.
Which Im sure would have shown a half stop under exposure lol

Ah yes but those meters meter correctly it's when you use a grey card when you need to compensate.

If you've a light meter and a grey card you can try it out, use the camera to meter from the grey card and use the light meter, wanna bet the exposure of the geycard is slightly darker.


Ansel Adams has always been promoting 18% grey, that's propably why we hear it so much and read it in books.


incidental, shooting whit a grey card and not compensating will help you for not clipping the highlights

[westmill]

Ah yes but those meters meter correctly it's when you use a grey card when you need to compensate.

If you've a light meter and a grey card you can try it out, use the camera to meter from the grey card and use the light meter, wanna bet the exposure of the geycard is slightly darker.


Ansel Adams has always been promoting 18% grey, that's propably why we hear it so much and read it in books.


incidental, shooting whit a grey card and not compensating will help you for not clipping the highlights

I will stick to the correct 18% thanks lol, and Ansel adams developed and used the zone system !

[Anvh]

What do you mean correct 18%?

[Markbrumbaugh]

Dudes...I am just trying to get a picture of a white duck. I did not ask if he was 12 or 18 percent grey.
Just looking for practical experience.
Very interesting stuff here on this thread, but it has gone off to the milky way.

[westmill]

What do you mean correct 18%?

i mean i beleive 18% is correct ? Why the question mark ? You beleive its watever you beleive. im happy with that.
Ive already stated im not that intrested. I never asked ! its not my thread ! I tried to help the person whose thread this is !
I offered sound advice ! Thats it lol. If i wanted a lenghy discussion on how a meter works i would post a thread.
We have you might say... gone off track lol

[westmill]

Dudes...I am just trying to get a picture of a white duck. I did not ask if he was 12 or 18 percent grey.
Just looking for practical experience.
Very interesting stuff here on this thread, but it has gone off to the milky way.

So sorry... was thinking the same thing lol

[joe.penn]

Dudes...I am just trying to get a picture of a white duck. I did not ask if he was 12 or 18 percent grey.
Just looking for practical experience.
Very interesting stuff here on this thread, but it has gone off to the milky way.

How to avoid blown highlights?

Negative EV Comp is probably the quickest and easiest solution. The more correct way would be to properly meter - one thing to remember is that ALL automated metering on ALL digital slr's/cameras suck, period! As others have noted, "Hyper Manual (M Mode)" is probably best to work with, "GRN button" the exposure and dial up shutter speed [snap] dial up shutter speed again [snap] - this will give you 2 shots underexposed different stops to work with and should save your highlights.

My personal preference is to use "Spot Metering" with "A/E Lock" - you can give that a shot also but it can get a bit tricky...



..

[Anvh]

Dudes...I am just trying to get a picture of a white duck. I did not ask if he was 12 or 18 percent grey.
Just looking for practical experience.
Very interesting stuff here on this thread, but it has gone off to the milky way.

It's still about light metering though.
You did got your answer i hope. =]

i mean i beleive 18% is correct ? Why the question mark ? You beleive its watever you beleive. im happy with that.
Ive already stated im not that intrested. I never asked ! its not my thread ! I tried to help the person whose thread this is !
I offered sound advice ! Thats it lol. If i wanted a lenghy discussion on how a meter works i would post a thread.
We have you might say... gone off track lol

The question mark is there because i didn't understand what you mean and so i asked you, that's what question marks are for right?

You can believe what you want but the standard is 12.3%, i'm very sorry but you nor i can change that, you've to live with.
The good news is it's like this already for years so just keep doing what you always do.

Nothing to do with the duck but everything to do with lightmetering though and i believe we were blaming the lightmetering right so we aren't that far off from track