Evaluative Metering: The Wrong Way

[boxf1]

Evaluative metering is nothing more than a marketing scheme. It averages the entire scene. Translation: Jack of all trades, master of none. When you use matrix, multi-segment, or whatever your camera calls evaluative metering, it tries to figure out the best overall exposure. The problem is, it will not hit the exact exposure correctly on most of the subjects since the algorithm is designed to find the best compromise. For that reason, spot, partial, or center-weighted metering are better options, depending on your situation.

The relatively new "Face-detection technology" is nothing more than spot metering on specific points (faces) and that is why faces are accurately exposed and everything else in the scene is ignored (and thus inaccurately exposed). This is acceptable because in a scene with people, the things we want to expose accurately are the people, and especially their faces.

Several cameras on the market today, like the Nik*n D80 and Ca*on Rebel XTI (400D) have problematic "matrix/evaluative" metering. Users are finding center-weighted metering (among the others I mentioned above - again, dependent on the situation) addresses the varying exposure issues. So my message is simple: turn off the multi-everything metering and use the more traditional, tried and true method that was perfectly fine until marketers told you it wasn't. Expose the subject close up, use exposure lock, or, if your camera doesn't have exposure lock, meter the scene up close and then use manual mode to regain the proper exposure setting after you recompose.

 

[Booga_roo]

Evaluative metering is nothing more than a marketing scheme. It
averages the entire scene. Translation: Jack of all trades, master of
none. When you use matrix, multi-segment, or whatever your camera
calls evaluative metering, it tries to figure out the best overall
exposure. The problem is, it will not hit the exact exposure
correctly on most of the subjects since the algorithm is designed to
find the best compromise. For that reason, spot, partial, or
center-weighted metering are better options, depending on your
situation.

It's not a marketing scheme, it's a valid metering choice. The marketing is in how they try to say "our metering is smarter" and that's certainly a valid thing to market. Every single feature on the camera has a marketing point to go with it.

There's nothing wrong with evaluative if you have a wide range of tones in diverse spots. It's easier than metering all the important bits and averaging them to preserve the best you can get.

It does well enough for most people, but there are plenty of situations that can fool the "best guess" Evaluative metering algorithms. It's up to the user to be smarter than the camera and recognize those situations and choose the proper metering mode. It's a disservice to tell others that they shouldn't use Evaluative metering if it will get them what they expect from the picture.

One doesn't always have time to go meter the scene properly and then go back to take the shot.

Exposure problems reported with certain models of camera does not mean that Evaluative metering is the problem, even if the other metering modes seems to help the problem. It certainly could be, but how about the other 200 models of camera that don't have these issues?
--
'Be right, fearless, faithfull, and true to others...'--T.S. Elliott

 

[Barry Fitzgerald]

Evaluative metering is nothing more than a marketing scheme. It
averages the entire scene. Translation: Jack of all trades, master of
none. When you use matrix, multi-segment, or whatever your camera
calls evaluative metering, it tries to figure out the best overall
exposure. The problem is, it will not hit the exact exposure
correctly on most of the subjects since the algorithm is designed to
find the best compromise. For that reason, spot, partial, or
center-weighted metering are better options, depending on your
situation.

You have a choice of metering patterns..for different situations..evaluative should work well 90% of the time..if not more.

There is no such thing as "exact" exposure...only the effect you want.

To understand metering, you must ist grasp "how" it works.....

The relatively new "Face-detection technology" is nothing more than
spot metering on specific points (faces) and that is why faces are
accurately exposed and everything else in the scene is ignored (and
thus inaccurately exposed). This is acceptable because in a scene
with people, the things we want to expose accurately are the people,
and especially their faces.

Its of use to some, of little use to experienced photographers...

Several cameras on the market today, like the Nik*n D80 and Ca*on
Rebel XTI (400D) have problematic "matrix/evaluative" metering. Users
are finding center-weighted metering (among the others I mentioned
above - again, dependent on the situation) addresses the varying
exposure issues. So my message is simple: turn off the
multi-everything metering and use the more traditional, tried and
true method that was perfectly fine until marketers told you it
wasn't. Expose the subject close up, use exposure lock, or, if your
camera doesn't have exposure lock, meter the scene up close and then
use manual mode to regain the proper exposure setting after you
recompose.

I dont agree...sure some are better than others, what exactly Nikon did to their D80 matrix metering, is a mystery...yes that isnt as good as it should be..some perform better than others in different scenarios, but most should give good results.

Some 400D's went back..seems a QC problem with metering or somethig..

But you missed the whole point of multizone metering..

Read this:

http://en.wikipedia.org/wiki/Metering_mode

Once you grasp multizone is "AF based" you will get it ;-)

--



Clint is on holiday! Soon to return! ;-)

 

[boxf1]

Evaluative metering is nothing more than a marketing scheme. It
averages the entire scene. Translation: Jack of all trades, master of
none. When you use matrix, multi-segment, or whatever your camera
calls evaluative metering, it tries to figure out the best overall
exposure. The problem is, it will not hit the exact exposure
correctly on most of the subjects since the algorithm is designed to
find the best compromise. For that reason, spot, partial, or
center-weighted metering are better options, depending on your
situation.

You have a choice of metering patterns..for different
situations..evaluative should work well 90% of the time..if not more.

90%? I know better, and so must you.

There is no such thing as "exact" exposure...only the effect you want.

Yes, there is. Read "Exposure." Exposure is a science when you understand latitude (film) or dynamic range (digital sensor).

To understand metering, you must ist grasp "how" it works.....

And I do.

The relatively new "Face-detection technology" is nothing more than
spot metering on specific points (faces) and that is why faces are
accurately exposed and everything else in the scene is ignored (and
thus inaccurately exposed). This is acceptable because in a scene
with people, the things we want to expose accurately are the people,
and especially their faces.

Its of use to some, of little use to experienced photographers...

Point and shoots are not for experienced photographers, that was clearly not the point of the example. Rather, I used FD metering as an example of what matrix metering is not.

Several cameras on the market today, like the Nik*n D80 and Ca*on
Rebel XTI (400D) have problematic "matrix/evaluative" metering. Users
are finding center-weighted metering (among the others I mentioned
above - again, dependent on the situation) addresses the varying
exposure issues. So my message is simple: turn off the
multi-everything metering and use the more traditional, tried and
true method that was perfectly fine until marketers told you it
wasn't. Expose the subject close up, use exposure lock, or, if your
camera doesn't have exposure lock, meter the scene up close and then
use manual mode to regain the proper exposure setting after you
recompose.

I dont agree...sure some are better than others, what exactly Nikon
did to their D80 matrix metering, is a mystery...yes that isnt as
good as it should be..some perform better than others in different
scenarios, but most should give good results.

But when they don't, and one learns about exposure, you can work around it, but not with evaluative metering!

Some 400D's went back..seems a QC problem with metering or somethig..

But you missed the whole point of multizone metering..

No, it seems you did.

Read this:

http://en.wikipedia.org/wiki/Metering_mode

If you use wiki to learn about photography, you don't know very much at al.

Once you grasp multizone is "AF based" you will get it ;-)

Nonsense. Take a basic photo class and learn about exposure. It is about reflected light.

--



Clint is on holiday! Soon to return! ;-)

 

[boxf1]

Evaluative metering is nothing more than a marketing scheme. It
averages the entire scene. Translation: Jack of all trades, master of
none. When you use matrix, multi-segment, or whatever your camera
calls evaluative metering, it tries to figure out the best overall
exposure. The problem is, it will not hit the exact exposure
correctly on most of the subjects since the algorithm is designed to
find the best compromise. For that reason, spot, partial, or
center-weighted metering are better options, depending on your
situation.

It's not a marketing scheme, it's a valid metering choice. The
marketing is in how they try to say "our metering is smarter" and
that's certainly a valid thing to market. Every single feature on the
camera has a marketing point to go with it.

Artifacts, slow lenses, high ISO noise, and a myriad of other elements in badly designed cameras are not marketable.

There's nothing wrong with evaluative if you have a wide range of
tones in diverse spots.

A wide range of tones in diverse spots is not where evaluative metering shines! Similar tones/reflected light, which is not very common in most scenes, is where evaluative metering does best.

It's easier than metering all the important

bits and averaging them to preserve the best you can get.

Sure it is, but if the subject is under/over exposed, what good does easy do if the picture is spoiled?

It does well enough for most people, but there are plenty of
situations that can fool the "best guess" Evaluative metering
algorithms.

We agree here! ;-)

It's up to the user to be smarter than the camera and
recognize those situations and choose the proper metering mode.

Read my post. I said this.

It's a disservice to tell others that they shouldn't use Evaluative
metering if it will get them what they expect from the picture.

It isn't a disservice. It's a disservice to not teach people about exposing the right way.

One doesn't always have time to go meter the scene properly and then
go back to take the shot.

Who does? Photography is an art that takes time.

Exposure problems reported with certain models of camera does not
mean that Evaluative metering is the problem, even if the other
metering modes seems to help the problem.

I didn't make that connection. You are creating a false argument. Read my original post.

It certainly could be, but how about the other 200 models of camera that > don't have these issues?

What about the other 200 cameras? Their EM works, but not as well as taking the time to learn how to expose properly - unless of course you have the perfect scene with near perfectly distributed light reflections.

--
'Be right, fearless, faithfull, and true to others...'--T.S. Elliott

 

[becksi]

No, it's a tool that is quite useful if you know how and when to apply it. A camera or its meter can't think and it certainly can't predict or read your mind about what kind of picture with what kind of exposure you are trying to take/achieve. There is no "right or wrong" in the classical sense, there is only knowledge, skill, talent and experience that allow you to achieve desired result on a consistant basis with any given equipment. The desired result might be different from reality or how it might have been seen in real life, that doesn't make it false though.

And I agree with you that partial or spot or center-weighted metering ARE better options/modes, BUT - and this is where I disagree - only depending on you situation. They are not superior in any general form, but there are particular situtations where one should be choosen over another ... depending on the desired effect one wants to achieve.

Is it (ab)used for marketing purposes? Sure, but if you now better, why bother complaining about it?

Evaluative metering is nothing more than a marketing scheme. It
averages the entire scene. Translation: Jack of all trades, master of
none. When you use matrix, multi-segment, or whatever your camera
calls evaluative metering, it tries to figure out the best overall
exposure. The problem is, it will not hit the exact exposure
correctly on most of the subjects since the algorithm is designed to
find the best compromise. For that reason, spot, partial, or
center-weighted metering are better options, depending on your
situation.

The relatively new "Face-detection technology" is nothing more than
spot metering on specific points (faces) and that is why faces are
accurately exposed and everything else in the scene is ignored (and
thus inaccurately exposed). This is acceptable because in a scene
with people, the things we want to expose accurately are the people,
and especially their faces.

Several cameras on the market today, like the Nik*n D80 and Ca*on
Rebel XTI (400D) have problematic "matrix/evaluative" metering. Users
are finding center-weighted metering (among the others I mentioned
above - again, dependent on the situation) addresses the varying
exposure issues. So my message is simple: turn off the
multi-everything metering and use the more traditional, tried and
true method that was perfectly fine until marketers told you it
wasn't. Expose the subject close up, use exposure lock, or, if your
camera doesn't have exposure lock, meter the scene up close and then
use manual mode to regain the proper exposure setting after you
recompose.

 

[boxf1]

No, it's a tool that is quite useful if you know how and when to
apply it.

Sure it's a tool, but usually not the best choice. If you really read what I wrote and my responses, you will see that I don't completely discount EM; my point is that there are usually better choices.

A camera or its meter can't think and it certainly can't
predict or read your mind about what kind of picture with what kind
of exposure you are trying to take/achieve. There is no "right or
wrong" in the classical sense, there is only knowledge, skill, talent
and experience that allow you to achieve desired result on a
consistant basis with any given equipment.

Yes, but knowledge comes through learning, and we are living in a time when people are overwhelmed by marketing and not knowledge. Again, that was the thrust of my post. I am not against knowledge and skill; I am trying to promote it by getting people too understand and switch metering modes!

The desired result might be different from reality or how it might have been seen in real life, that doesn't make it false though.

Sure, one might want to create a silhouette or similar effect, in which case evaluative in a back lit situation might work OK.

And I agree with you that partial or spot or center-weighted metering
ARE better options/modes, BUT - and this is where I disagree - only
depending on you situation.

Please read my first post before you disagree with what we agree on. ;-)

They are not superior in any general
form, but there are particular situtations where one should be
choosen over another ... depending on the desired effect one wants to
achieve.

Well, they are superior in that they give more accurate readings in most situations, especially when used properly with a knowledge of a particular cameras abilities (sensor or film used and what range of dynamic range/latitude)

Is it (ab)used for marketing purposes? Sure, but if you now better,
why bother complaining about it?

I'm not complaining - not all all. I am just informing people because this is a forum where people come to exchange ideas and knowledge.

Evaluative metering is nothing more than a marketing scheme. It
averages the entire scene. Translation: Jack of all trades, master of
none. When you use matrix, multi-segment, or whatever your camera
calls evaluative metering, it tries to figure out the best overall
exposure. The problem is, it will not hit the exact exposure
correctly on most of the subjects since the algorithm is designed to
find the best compromise. For that reason, spot, partial, or
center-weighted metering are better options, depending on your
situation.

The relatively new "Face-detection technology" is nothing more than
spot metering on specific points (faces) and that is why faces are
accurately exposed and everything else in the scene is ignored (and
thus inaccurately exposed). This is acceptable because in a scene
with people, the things we want to expose accurately are the people,
and especially their faces.

Several cameras on the market today, like the Nik*n D80 and Ca*on
Rebel XTI (400D) have problematic "matrix/evaluative" metering. Users
are finding center-weighted metering (among the others I mentioned
above - again, dependent on the situation) addresses the varying
exposure issues. So my message is simple: turn off the
multi-everything metering and use the more traditional, tried and
true method that was perfectly fine until marketers told you it
wasn't. Expose the subject close up, use exposure lock, or, if your
camera doesn't have exposure lock, meter the scene up close and then
use manual mode to regain the proper exposure setting after you
recompose.

 

[Booga_roo]

Artifacts, slow lenses, high ISO noise, and a myriad of other
elements in badly designed cameras are not marketable.

Sure they're marketable. Look at how they market the useless high ISO binned output that looks like watercolors. It's junk and we know it, but it makes for great marketing. Slow lenses are perfectly marketable as long as you market the zoom range.

Similar tones/reflected light, which is not very common in most scenes, is
where evaluative metering does best.

It's almost all reflected light unless you're shooting pictures of light sources. Evaluative does it's best with the wide range of tones based on a set of algorithms. Those algorithms may not be perfect, but neither is any other metering mode. They all expose to the same exposure for the area selected.

It's easier than metering all the important

bits and averaging them to preserve the best you can get.

Sure it is, but if the subject is under/over exposed, what good does
easy do if the picture is spoiled?

Which is why you need to make a choice.
Are you capturing a scene without a definitive subject of focus? Evaluative.
Are you capturing a subject? Perhaps another metering mode will be better.

What good does having your subject properly exposed do if your picture is spoiled with overexposed/underexposed environments. You wouldn't want a beach volleyball player in black cloths with no beach because it blew out the sand and sky would you?

It's up to the user to be smarter than the camera and
recognize those situations and choose the proper metering mode.

Read my post. I said this.

Except you discounted one of the choices: Evaluative.

I know I can name situations where Evaluative is the wrong choice. However, Evaluative DOES work well for me when trying to capture your average street scene. Center weighted, partial, and spot will often overexpose the sky and lighter buildings. Evaluative will of course lower the exposure, but not usually to the point where people are underexposed. I'd rather not have to recolor all the skies.

It's a disservice to tell others that they shouldn't use Evaluative
metering if it will get them what they expect from the picture.

It isn't a disservice. It's a disservice to not teach people about
exposing the right way.

Yes, teaching them to expose properly is good. However, telling people not to use one of the tools on their camera is not "the right way." It's "your way."

If you refuse to use a feature, that's fine. I rarely use evaluative myself because of what I take pictures of. That doesn't mean I never use it though.

When I'm under the forest canopy in the evening, I have to adjust the exposure compensation down to keep the exposure realistic. None of the metering modes will give me what I want because they're not smart, they're just tools on the camera.

One doesn't always have time to go meter the scene properly and then
go back to take the shot.

Who does? Photography is an art that takes time.

You told people to take time they don't have by stating: "Expose the subject close up, use exposure lock, or, if your camera doesn't have exposure lock, meter the scene up close and then use manual mode to regain the proper exposure setting after you recompose. "

This makes sense in some situations, but none for your average snapshot. Not everyone is trying to make art with every shot. There's no need for people to do this 99% of the time.

Exposure problems reported with certain models of camera does not
mean that Evaluative metering is the problem, even if the other
metering modes seems to help the problem.

I didn't make that connection. You are creating a false argument.
Read my original post.

Yes, you did. You specifically mentioned the Nikon and Canon(you don't have to use 's they aren't cursewords.) You stated that using another metering mode improved exposure problems for these people. Your message insinuates Evaluative metering as the cause, even if you didn't mean to.

What about the other 200 cameras? Their EM works, but not as well as
taking the time to learn how to expose properly - unless of course
you have the perfect scene with near perfectly distributed light
reflections.

Proper exposure is subjective. It's always a choice. If Evaluative gives you what you want, then it is the proper exposure. Trying to make it seem like you can't get a proper exposure without a "perfect scene" makes no sense.
--
'Be right, fearless, faithfull, and true to others...'--T.S. Elliott

 

[Yuzuha]

My old Canon T90 could take up to 5 spot meter readings and average them (Yes, overkill, I usually only used 2 or 3). It worked just great, provided you had the time to use it (which was almost never for quick grab shots). Center weighted worked fine too but it was really easy to fool it. I think matrix is sort of a high speed, scatter-gun approach that tries to get a decent exposure of something, especially when the camera is used in "point and blast" mode. Sometimes it works well, other times it doesn't.

 

[boxf1]

Artifacts, slow lenses, high ISO noise, and a myriad of other
elements in badly designed cameras are not marketable.

Sure they're marketable. Look at how they market the useless high ISO
binned output that looks like watercolors. It's junk and we know it,
but it makes for great marketing. Slow lenses are perfectly
marketable as long as you market the zoom range.

They are marketable as long as people are in the dark about their uselessness. Again, this is a forum where we should be, and are, free to discuss this.

Similar tones/reflected light, which is not very common in most scenes, is
where evaluative metering does best.

It's almost all reflected light unless you're shooting pictures of
light sources.

In fact, it is all reflected light.

Evaluative does it's best with the wide range of tones
based on a set of algorithms.

No, the opposite is true!

Those algorithms may not be perfect,
but neither is any other metering mode.

I never said any exposure mode was perfect. You are trying to create a straw-man argument. I'm not biting, sorry. ;-)

They all expose to the same exposure for the area selected.

No they don't. EM averages the scene!

It's easier than metering all the important

bits and averaging them to preserve the best you can get.

But again, as I said, it is not as good as being more precise by metering the subject you want properly exposed.

Sure it is, but if the subject is under/over exposed, what good does
easy do if the picture is spoiled?

Which is why you need to make a choice.

And this is what I am saying one should do - make a choice - not just go with the default marketing mode!

Are you capturing a scene without a definitive subject of focus?
Evaluative.

Ok. I'm not arguing against that - but good photos generally have a subject of some sort.

con't next post

 

[boxf1]

Are you capturing a subject? Perhaps another metering mode will be
better.

Right!

What good does having your subject properly exposed do if your
picture is spoiled with overexposed/underexposed environments. You
wouldn't want a beach volleyball player in black cloths with no beach
because it blew out the sand and sky would you?

Uh.... no. Did I say I would? No.

It's up to the user to be smarter than the camera and
recognize those situations and choose the proper metering mode.

Again, yes. You write this as if we disagree. ???

Read my post. I said this.

Except you discounted one of the choices: Evaluative.

Nope, read again. I did not discount it. The point of the post was to say that evaluative is a marketing scheme that doesn't usually make for the best choice. Rad my follow up posts... I provide more detail where I agree evaluative metering is in fact effective.

I know I can name situations where Evaluative is the wrong choice.
However, Evaluative DOES work well for me when trying to capture your
average street scene.

I agree, and I'm glad it works for you. But the average person who doesn't understand metering and blows highlights or clips shadows doesn't shoot street scenes regularly.

Center weighted, partial, and spot will often
overexpose the sky and lighter buildings. Evaluative will of course
lower the exposure, but not usually to the point where people are
underexposed. I'd rather not have to recolor all the skies.

If my subject is a street/buildings, I'd rather partial or center weight them and have the buildings in proper exposure, rather that use evaluative/matrix and underexpose them because the camera is trying not to blow out the sky. I discussed this principle in my original post when I discussed averaging and missing the proper exposure on the subject.

It's a disservice to tell others that they shouldn't use Evaluative
metering if it will get them what they expect from the picture.

It isn't a disservice. It's a disservice to not teach people about
exposing the right way.

Yes, teaching them to expose properly is good. However, telling
people not to use one of the tools on their camera is not "the right
way." It's "your way."

Spot, partial, and center-weighted metering are hardly my way. Evaluative metering is a tool, yes, but often it is one used for lack of knowing the others, and a reason why many photos posted here are poorly exposed. In the truest sense, exposing for anything but the subject is improper exposure. If you interpret that as "my way" then I suggest you read up on the science of exposure.

 

[boxf1]

If you refuse to use a feature, that's fine. I rarely use evaluative
myself because of what I take pictures of. That doesn't mean I never
use it though.

And I am not implying that I never use it either. But those who claim to use it 90% of the time either do not understand exposure, or are too lazy to switch to a mode that requires more thought. I would say that the former is true most of the time and that's ok - that is why we have a forum to teach, learn, and be taught.

When I'm under the forest canopy in the evening, I have to adjust the
exposure compensation down to keep the exposure realistic.

No, you really don't. You never have to use exposure compensatrion. It is a tool, a a shortcut/convenience to take the place of setting your exposure manually. There's nothing wrong with it, but it can be useful to know the DR of your sensor, assuming you shoot digital, and get more from the scene rather than just try to average out your exposure.

None of the metering modes will give me what I want because they're not
smart, they're just tools on the camera.

I don't disagree.

One doesn't always have time to go meter the scene properly and then
go back to take the shot.

Then that photographer will get results proportionate to that level of work.

Who does? Photography is an art that takes time.

You told people to take time they don't have by stating: "Expose the
subject close up, use exposure lock, or, if your camera doesn't have
exposure lock, meter the scene up close and then use manual mode to
regain the proper exposure setting after you recompose. "

Then they aren't photographers, they are snapshooters. Photographers take time to learn and practice exposure. There are no shortcuts in photography. Film latitude and sensor DR don't make exceptions and behave magically because people don't have time to set up their shots.

This makes sense in some situations, but none for your average
snapshot.

Like I said above...

Not everyone is trying to make art with every shot.

Most people who take the time to study these forums on a Friday night/Saturday am do!

There's no need for people to do this 99% of the time.

And again, their results will match their skill and time investment.

Exposure problems reported with certain models of camera does not
mean that Evaluative metering is the problem, even if the other
metering modes seems to help the problem.

I didn't make that connection. You are creating a false argument.
Read my original post.

Yes, you did.

Sorry, i don't see the connection even though you would like to insist that I did.

You specifically mentioned the Nikon and Canon(you
don't have to use 's they aren't cursewords.)

I use asterisks not because I believe those brands are bad words, but as a courtesty to those who are doing searches on the forum to learn about those camera brands. It is a matter of etiquette. Read other posts for other examples.

You stated that using another metering mode improved exposure problems for these people. Your message insinuates Evaluative metering as the cause, even if you didn't mean to.

You have a creative imagination - perhaps that is why you enjoy photography.

What about the other 200 cameras? Their EM works, but not as well as
taking the time to learn how to expose properly - unless of course
you have the perfect scene with near perfectly distributed light
reflections.

Proper exposure is subjective.

Oh not it is not! Read a histogram or look at slides. No offense, but that statement clearly shows that you do not understand the very basics of exposure.

It's always a choice.

Again, no.

If Evaluative gives you what you want, then it is the proper exposure. Trying to
make it seem like you can't get a proper exposure without a "perfect
scene" makes no sense.

You said that, not me. I said that evaluative is useful for a scene that has similar reflected light because the exposure will be similar on all subject matter. But your use of evaluative metering for your bright sky with low/average light on buildings is a perfect example of where evaluative is almost completely useless - unless for some artistic reason you enjoy underexposed main subjects - but don't tell a real photographer that it is correctly exposed!

--
'Be right, fearless, faithfull, and true to others...'--T.S. Elliott

 

[Brian]

Boy, someone must have woken up on the wrong side of the bed to go off on a rant about how bad evaluative metering is.

The fact is that it is quite useful a lot of the time. It is like any other tool. If you use it a lot, you'll learn how best to handle it.

If you don't like evaluative metering, don't use it.

It certainly works for me most of the time and when it doesn't I try a different method. It certainly isn't rocket science.

--
Brian

 

[boxf1]

Thank you for the brilliant summary.

Why do people on this forum call anything they don't like or don't understand a "rant?"

Do you know what a forum is? See my corrections to your false assertions below...

Boy, someone must have woken up on the wrong side of the bed to go
off on a rant about how bad evaluative metering is.

I got a nice pay increase at work today and am feeling very good actually.

The fact is that it is quite useful a lot of the time. It is like
any other tool. If you use it a lot, you'll learn how best to handle
it.

That is nonsense babble - made with a poor attempt - to sound like something credible.

If you don't like evaluative metering, don't use it.

Wow - what genius! Tell us more, please.

It certainly works for me most of the time and when it doesn't I try
a different method.

What mastery of the subject you possess!

It certainly isn't rocket science.

No, it isn't, but it seems you need to read a book or two on the subject though.

And now it is time for me to leave. Have a great weekend everyone.

--
Brian

 

[Booga_roo]

I'd respond, but There are so many things to respond to. They're a very fast typist, that's for sure.

I think there must be a language barrier or tone of voice issue. He misunderstands and says my statements are untrue when he talks about my opinion. I aparently misunderstand what he says since he keeps saying "I never said that" then still denies it when I show him his own text. I think this is one of those times I'd actually prefer to talk face to face to cut down on the amount of time it takes to hash all this out.

It's not like it's going to change how most of us work anyway. DPReview is already mostly people who are taking the time to learn about their equipment. I

I just hope that people get the idea that the camera's meter is capable of more than one setting, instead of the idea that Evaluative is a useless marketing ploy.
--
'Be right, fearless, faithfull, and true to others...'--T.S. Elliott

 

[Brian]

You welcome for the brilliant summary. It was a gift to one who has not quite grasped the benefits of evaluative metering. However, it appears that my gift was as pearls before swine. Don't worry though, it cost me nothing.

But seriously, won't you really tell us your feelings about evaluative metering. I sense that you may have something against it. perhaps it wronged you in some way. Do your cameras' evaluative metering not work properly and so you want all the rest of us not to have the fortune of using it?

and oh, btw it was a rant.

Why do people on this forum call anything they don't like or don't
understand a "rant?"

Do you know what a forum is? See my corrections to your false
assertions below...

Boy, someone must have woken up on the wrong side of the bed to go
off on a rant about how bad evaluative metering is.

I got a nice pay increase at work today and am feeling very good
actually.

congrads. that is always a good thing.

The fact is that it is quite useful a lot of the time. It is like
any other tool. If you use it a lot, you'll learn how best to handle
it.

That is nonsense babble - made with a poor attempt - to sound like
something credible.

you must be right then. after rereading it, i could not make out a thing I said. I'm going to right a letter to my camera manufacturer right now and demand that they remove this feature.

If you don't like evaluative metering, don't use it.

Wow - what genius! Tell us more, please.

I'm terribly sorry, but I already said all there was to say on the subject, after all, I am the genius.

It certainly works for me most of the time and when it doesn't I try
a different method.

What mastery of the subject you possess!

yep.

It certainly isn't rocket science.

No, it isn't, but it seems you need to read a book or two on the
subject though.

Could you please explain to me what aspect of metering I don't understand so I will be better able to read the proper books. Perhaps you can give me the name of the book that tells one to never use evaluative metering.

And now it is time for me to leave. Have a great weekend everyone.

Certainly will.

--
Brian

 

[Staxis]

The key point is to learn what metering can and cannot do. Matrix metering is good a lot of the time but when it does not get it right it goes wrong in an unpredictable manor (in my experience). matrix metering rarely gets it right in difficult lighting conditions when the most stunning images are taken - this is where the photographer's skill comes in.

Centre weighted gets it right some of the time but when it fails to deliver it does so in a predictable way so it is easy to compensate for its shortcomings.

Spot metering is very useful but you have to know that your target is equivalent to 18% grey - or how far off 18% grey it is.

The point is you need to understand which metering pattern to use and when, or do you.....

.....with DSLRs it is very easy to check the exposure using the playback and histogram. If you got it wrong first time then reshoot - exposure nailed every time no matter which in built metering mode you use. If you are shooting sport, for example, where you cannot reshoot, you can go through this procedure using a target with the same lighting as you real one.

For the record, my DSLR is set to use centre weighted metering and I use exposure compensation and manuel exposure mode a lot. The AEL buton is set to use spot metering so this is available to me whenever I want it, which is quite frequently.
--
Greg

When you've got a moment, have a look at my newly updated site:
http://www.wrightphotos.co.uk
also http://www.wrightphotos.co.uk/FromeInFocus

And now the blog: http://www.wrightphotos.co.uk/blog

Winner of the South West Rural section of the BBC's Picture of Britain Competition.

 

[Rumpis]

It depends. My wife is using Nikon D40X in auto modes including matrix metering. I have checked the result by histograms and find D40X matrix metering very good. It lets her to spend more attention to composition and not too much to technical details.

Evaluative metering is nothing more than a marketing scheme. It
averages the entire scene. Translation: Jack of all trades, master of
none. When you use matrix, multi-segment, or whatever your camera
calls evaluative metering, it tries to figure out the best overall
exposure. The problem is, it will not hit the exact exposure
correctly on most of the subjects since the algorithm is designed to
find the best compromise. For that reason, spot, partial, or
center-weighted metering are better options, depending on your
situation.

The relatively new "Face-detection technology" is nothing more than
spot metering on specific points (faces) and that is why faces are
accurately exposed and everything else in the scene is ignored (and
thus inaccurately exposed). This is acceptable because in a scene
with people, the things we want to expose accurately are the people,
and especially their faces.

Several cameras on the market today, like the Nik*n D80 and Ca*on
Rebel XTI (400D) have problematic "matrix/evaluative" metering. Users
are finding center-weighted metering (among the others I mentioned
above - again, dependent on the situation) addresses the varying
exposure issues. So my message is simple: turn off the
multi-everything metering and use the more traditional, tried and
true method that was perfectly fine until marketers told you it
wasn't. Expose the subject close up, use exposure lock, or, if your
camera doesn't have exposure lock, meter the scene up close and then
use manual mode to regain the proper exposure setting after you
recompose.

--
Rumpis )

 

[John Bean (UK)]

There is no such thing as "exact" exposure...only the effect you want.

Yes, there is. Read "Exposure." Exposure is a science when you
understand latitude (film) or dynamic range (digital sensor).

Only if it was negative film, or in the case of digital the resulting capture is going to be post-processed.

If you are using slide film or want to use a digital image "straight from the camera" then Barry is absolutely correct, exposure is correct if - and only if - it delivers the image content that you visualised.

--
John Bean [BST/GMT+1] ('British Stupid Time')

PAW 2007 Week 30:
http://waterfoot.smugmug.com/gallery/2321711/2/177298069/Large



Index page: http://waterfoot.smugmug.com
Latest walkabout (21 March 2007):
http://waterfoot.smugmug.com/gallery/2641073

 

[aukhawk]

.....with DSLRs it is very easy to check the exposure using the
playback and histogram. If you got it wrong first time then reshoot

• exposure nailed every time

and not just DSLRs - almost every digital camera

Shoot - check - adjust - reshoot is a whole lot quicker than angsting about which metering mode to use and then deciding on suitable mid-tones in the subject area etc etc
OK that is de-skilling the process but that's progress for you.

(For the record I use centre-weighted metering and usually keep the camera in manual mode - having gone through the above procedure once I wouldn't bother to check exposure again unless the light changes or my viewpoint does.)

Francis