Posted by Greg Maklae on February 14, 2010 at 10:50 AM
What type of metering is best for landscape photography, matrix or spot? Or should I just be concerned about a good histogram and avoiding the "flashies". My manual say maxtrix, just read an article that says spot. The more I study, the less I know!
Posted by Fred Stearns on February 14, 2010 at 12:09 PM
Greg,
It may sound like I am evading the question, but I would suggest that you try it both ways and see which one works best for you. I have known photographers who shoot either way. Personally, I use matrix (evaluative) metering. I started out with a manual camera where the only choice was center-weighted averaging. Matrix is better. If you have a large variation in the brightness, I suggest that you take 3 to 5 identical picture with different exposures (-1, 0, +1; -2, 0 +2; -2, -1, 0, +2, +2) and merge then using one of the HDR programs around. Another truth is that landscapes often have shadows too dark to be rendered by JPEG while having blown out highlights. Try re-framing in the viewfinder to eliminate whichever one you really don't really need.
If you are shooting RAW, you can overcome some of the "flashies" during processing. What you are looking at on the back of the camera is the embedded JPEG file in the RAW file. If you are shooting JPEG, that is what you get. If you are shooting JPEG, I suggest that you try a day of shooting RAW + JPEG and see what you can do with the RAW images.
I wish you luck!
Posted by Andy Richards on February 14, 2010 at 12:55 PM
Hi Greg. In my view, your question is really more about understanding exposure than about what kind of meter to use. A meter is a measuring tool, and the different types of metering are really just different methods of measurement. Reflective meters (spot, center-weighted, matrix, etc.) theoretically measure the light being reflected from what they are reading. Incident meters measure the light falling on the subject (handheld). In-camera meters are reflective.
The key to the reflected meter is my above comment "from what they are reading." A "spot" meter is a direct, reflected reading on a fairly small area. It just measures the light being reflected from that small area. Matrix reads a larger area inside your viewfinder, and then does some "computer" calculations to determine the "best" setting. I am only familiar with Nikon's matrix metering system. What they say is that they have input into the "computer" 1000's of different previously measured scenes and compares the current reading. Its actually pretty accurate about 90 - 95% of the time.
But, its not really a "smart" tool. It is very literal. And, as such, it can be fooled. In a way, it doesn't really matter which type you use, as long as the result is good (I know, kind of a ridiculous statement ). But the important thing is knowing which parts of the image you want to be "properly" exposed and knowing what these meters will "tell" you.
What they do is give you a reading that will produce a theoretically "neutral" result. If you understand what you are trying to accomplish, it is hard to go wrong using the spot meter. My experience with matrix is pretty darn good, but there are times when it gets "fooled" by the conditions.
As Fred notes, RAW and DSLR's have made this a much "easier" deal these days. I attended a John Shaw seminar last winter and he made this point. He used to write books on metering and exposure. Now, he notes that he sets the matrix meter, and shoots a test shot. Then he checks the histogram and adjusts. It works!
Posted by David Bostedo on February 14, 2010 at 1:17 PM
I'd say just be more concerned with a good histogram.
-- Choose the ISO you want (probably as low as possible if it's a static scene)
-- Choose the aperture you want for the DOF you want,
-- Adjust the shutter speed (or let the camera adjust it in aperture priority) to get a good histogram (the camera meter can be used here to judge your speed or exposure compensation)
Note that if you really want to capture the most data (most detail, least noise) the histogram will be pushed to the right just barely avoiding blowing out highlights. That could make the scene overexposed, and you'll have to bring it back down in post.
Otherwise, if you don't want to do the work in post, you can shoot for a more ideal histogram for the scene, or whatever look you're going for.
Matrix. Only if you have a subject taking up a small portion of the frame would you use spot metering. Matrix will give you a metering for the entire scene. Landscape photography is often about the entire scene, so I would use that. Using HDR photography or ND filters can improve the histogram.
Remember, a good histogram is just a guideline. Some of the best pictures are overexposed or underexposed. Go for whatever looks good.
Posted by Dan Borzynski on February 14, 2010 at 6:42 PM
Whatever Canon's least respected metering pattern is (evaluative??). Why? Because the first real camera I owned (Canon Elan 7) had evaluative and center-weighted. To confuse things a bit, I use spot metering or incident metering when I shoot 4x5. Why? Because that is what my handheld meter allows me to do.
To a large extent, I don't think the metering pattern really makes all that much difference, *assuming* that you understand what it does under the conditions you shoot it. I very rarely get exposure surprises in any of the metering modes I use, assuming that I was paying at least some attention to the scene I am shooting. When I am not, the metering mode becomes irrelevant - I am going to screw up the exposure anyway.
The caution I give though, is that you have to shoot a lot, critically look at exposures, understand why the meter went wrong AND learn from it. Just passively looking at a histogram or blown shots doesn't accomplish that. In short, you need to be an active participant in the creation of and evaluation of the image.
Here is a drill - compose and meter something white or black and/or bright and/or contrasty and/or reflective. What do you think the exposure compensations should be? Does this agree with the histogram? How about the actual image shot? Repeat every time you pick up the camera. Don't forget to keep the metering mode the same, whatever that means for your camera.
Back in the old days when Velviasausus Rex ruled the Earth, you learned exposure pretty quickly or spent a lot of money on bad shots. There wasn't really all that much choice in the matter. Of course, we also walked to school 10 miles, uphill each way, in the snow...
Posted by David Taylor on February 15, 2010 at 12:28 AM
I think it really depends on your scene. The idea of shooting "only in Matrix" metering can get you in trouble - even with landscapes. The first thought about exposure (any exposure, of any subject - be it wildlife, landscape, whatever) should ALWAYS be "what is most important to me" in this scene? For instance, if you are shooting a dark wolf walking across a bright snowy scene (or conversely, a gull set against a dark black body of water), your Matrix meter reading will likely mis-interpret the scene. With each of these scenarios, your camera will either under-expose the wolf, or over-expose the gull (typically) if shot in Matrix/Evaluative metering modes. For this kind of scene a spot meter would likely work best, or if the subject filled more of the frame, consider using center-weighted metering.
Exposure is all about "playing scales". Yes, RAW format is more flexible about recovering highlights or shadows than film was - but every pro will tell you - "Get it right in the camera". Shoot as if you don't have any post processing options. The more you are able to get the exposure correct at capture, the less "massaging" you will need to do in post, and the higher the quality your final image will be.
Over the past year or so, I have shot 99.9% of my landscape shots in Live View. As I'm composing the shot, I'll have only the Live View screen on, with no distractions except for the exposure data below the live image. When I work on the actual exposure, I cycle the Live View screens to the full data option, which shows the live image and the exposure data + the RGB/Luminosity histogram overlaying the image. When I have time, I usually do what Andy mentioned regarding John Shaw. Take a test exposure and reevaluate for the real thing.
One thing to remember: the histogram on your camera is not infallible. They are generated off a jpeg preview image taken from the actual RAW file, and as such, may not exactly match your RAW file histogram once imported into Lightroom/Aperture/Photoshop/etc. It's important to understand how your camera actually captures data, versus how it displays an approximation of that same data through a histogram.
Hope this helps.
Posted by Mel Ladewig on February 16, 2010 at 6:58 PM
I personally use a spot meter (Pentax digital 1 degree) for large format and evaluative for my 35mm. As Dan said, you just need to understand what the meter is doing and what compensation you need to make.
Posted by Dan Baumbach on February 20, 2010 at 10:17 AM
I've been shooting large format with spot metering before I started shooting digital, so I'm comfortable with using a hand held spot meter for exposures. When I shoot digital, I use matrix and then check the histogram. I then vary the exposure to make sure I don't have any blown highlights.
- Dan
Posted by Buck Ward on February 20, 2010 at 11:59 AM
I use evaluative metering with exposure compensation based on my judgment from my experience. Then I check the histogram and the 'blinkies' and adjust as needed. It works for me. Each photographer develops his own way of doing things. It really is "whatever works for you."
Digital cameras with instant feedback pretty much take the guesswork out of exposure. In this day of cameras with automatic everything, however, it may be not so easy to develop an instinct for exposure. Most modern cameras are going to get it right most of the time. With experience, you learn to apply the compensation to make the exposure be the way you want it to be. I think most photographers probably don't choose spot metering for landscapes, but in some circumstances it may be a good choice.
Matrix (evaluative) or spot - pick one and learn it. It's like learning to drive a car - with a bit of practice it becomes second nature.
Discussion
9 repliesIt may sound like I am evading the question, but I would suggest that you try it both ways and see which one works best for you. I have known photographers who shoot either way. Personally, I use matrix (evaluative) metering. I started out with a manual camera where the only choice was center-weighted averaging. Matrix is better. If you have a large variation in the brightness, I suggest that you take 3 to 5 identical picture with different exposures (-1, 0, +1; -2, 0 +2; -2, -1, 0, +2, +2) and merge then using one of the HDR programs around. Another truth is that landscapes often have shadows too dark to be rendered by JPEG while having blown out highlights. Try re-framing in the viewfinder to eliminate whichever one you really don't really need.
If you are shooting RAW, you can overcome some of the "flashies" during processing. What you are looking at on the back of the camera is the embedded JPEG file in the RAW file. If you are shooting JPEG, that is what you get. If you are shooting JPEG, I suggest that you try a day of shooting RAW + JPEG and see what you can do with the RAW images.
I wish you luck!
The key to the reflected meter is my above comment "from what they are reading." A "spot" meter is a direct, reflected reading on a fairly small area. It just measures the light being reflected from that small area. Matrix reads a larger area inside your viewfinder, and then does some "computer" calculations to determine the "best" setting. I am only familiar with Nikon's matrix metering system. What they say is that they have input into the "computer" 1000's of different previously measured scenes and compares the current reading. Its actually pretty accurate about 90 - 95% of the time.
But, its not really a "smart" tool. It is very literal. And, as such, it can be fooled. In a way, it doesn't really matter which type you use, as long as the result is good (I know, kind of a ridiculous statement
What they do is give you a reading that will produce a theoretically "neutral" result. If you understand what you are trying to accomplish, it is hard to go wrong using the spot meter. My experience with matrix is pretty darn good, but there are times when it gets "fooled" by the conditions.
As Fred notes, RAW and DSLR's have made this a much "easier" deal these days. I attended a John Shaw seminar last winter and he made this point. He used to write books on metering and exposure. Now, he notes that he sets the matrix meter, and shoots a test shot. Then he checks the histogram and adjusts. It works!
-- Choose the ISO you want (probably as low as possible if it's a static scene)
-- Choose the aperture you want for the DOF you want,
-- Adjust the shutter speed (or let the camera adjust it in aperture priority) to get a good histogram (the camera meter can be used here to judge your speed or exposure compensation)
Note that if you really want to capture the most data (most detail, least noise) the histogram will be pushed to the right just barely avoiding blowing out highlights. That could make the scene overexposed, and you'll have to bring it back down in post.
Otherwise, if you don't want to do the work in post, you can shoot for a more ideal histogram for the scene, or whatever look you're going for.
Remember, a good histogram is just a guideline. Some of the best pictures are overexposed or underexposed. Go for whatever looks good.
To a large extent, I don't think the metering pattern really makes all that much difference, *assuming* that you understand what it does under the conditions you shoot it. I very rarely get exposure surprises in any of the metering modes I use, assuming that I was paying at least some attention to the scene I am shooting.
The caution I give though, is that you have to shoot a lot, critically look at exposures, understand why the meter went wrong AND learn from it. Just passively looking at a histogram or blown shots doesn't accomplish that. In short, you need to be an active participant in the creation of and evaluation of the image.
Here is a drill - compose and meter something white or black and/or bright and/or contrasty and/or reflective. What do you think the exposure compensations should be? Does this agree with the histogram? How about the actual image shot? Repeat every time you pick up the camera. Don't forget to keep the metering mode the same, whatever that means for your camera.
Back in the old days when Velviasausus Rex ruled the Earth, you learned exposure pretty quickly or spent a lot of money on bad shots. There wasn't really all that much choice in the matter. Of course, we also walked to school 10 miles, uphill each way, in the snow...
Exposure is all about "playing scales". Yes, RAW format is more flexible about recovering highlights or shadows than film was - but every pro will tell you - "Get it right in the camera". Shoot as if you don't have any post processing options. The more you are able to get the exposure correct at capture, the less "massaging" you will need to do in post, and the higher the quality your final image will be.
Over the past year or so, I have shot 99.9% of my landscape shots in Live View. As I'm composing the shot, I'll have only the Live View screen on, with no distractions except for the exposure data below the live image. When I work on the actual exposure, I cycle the Live View screens to the full data option, which shows the live image and the exposure data + the RGB/Luminosity histogram overlaying the image. When I have time, I usually do what Andy mentioned regarding John Shaw. Take a test exposure and reevaluate for the real thing.
One thing to remember: the histogram on your camera is not infallible. They are generated off a jpeg preview image taken from the actual RAW file, and as such, may not exactly match your RAW file histogram once imported into Lightroom/Aperture/Photoshop/etc. It's important to understand how your camera actually captures data, versus how it displays an approximation of that same data through a histogram.
Hope this helps.
- Dan
Digital cameras with instant feedback pretty much take the guesswork out of exposure. In this day of cameras with automatic everything, however, it may be not so easy to develop an instinct for exposure. Most modern cameras are going to get it right most of the time. With experience, you learn to apply the compensation to make the exposure be the way you want it to be. I think most photographers probably don't choose spot metering for landscapes, but in some circumstances it may be a good choice.
Matrix (evaluative) or spot - pick one and learn it. It's like learning to drive a car - with a bit of practice it becomes second nature.