Jump to content

ligh metering options on powershot?


storie

Recommended Posts

<p>I've been taking advice and just playing with my camera (powershot 1400sd is) for the last few days. I just discovered that I can change my light meter settings between spot, center weighted and evaluated. I can't find the manual and am wondering in what situations would I use the different settings? and is one a better option for most situations?</p>
Link to comment
Share on other sites

<p>The User's manual is included in your package when you buy the camera, it is in the CD in the box. At page 88, it says:<br>

- <em>Evaluative: Suitable for standard shooting conditions, including back lit shots.</em><br>

- Center weighted: Averages the light metered from the entire frame, but gives greater weight to the center<br>

- Spot: Only meters within the [ ] that appears at the center ofthe screen</p>

<p>The emphasis is added by me</p>

Link to comment
Share on other sites

<p>To elaborate slightly on that:</p>

<dl>

<dt>Evaluative</dt>

<dd>The camera analyses the scene, compares it to an internal database, and tries to work out what exposure might be right. The benefit is that this is an intelligent system, and very often it gets it right - it's a sensible default. The disadvantage is that it can get confused, and it's hard to tell what it's going to do.</dd>

<dt>Centre-weighted</dt>

<dd>This was a common approach before people started putting computers into camera meters: the camera will look at the scene, paying more attention to the area around the middle than the edges, and try to make the average brightness of the scene appear as a mid-grey (in brightness) in the result. This is surprisingly effective a lot of the time. You can change whether the camera's aiming at mid-grey (to make it lighter or darker, if - for example - you're taking an image of snow or a coal pit) with <i>exposure compensation</i>.</dd>

<dt>Spot</dt>

<dd>This takes what you're pointing the "[ ]" at, and tries to make <i>that</i> look mid-grey (in brightness). It's useful if the rest of the scene is confusing the camera. For example, if you're trying to take an image of a person with the sun behind them, or hit by a stage spotlight (with a black background), it can be useful to spot meter by pointing the "[ ]" at their face - then the camera won't get confused by the rest of the scene. You may want to use exposure compensation so that you can make the thing you're pointing at lighter or darker than mid-grey.</dd>

</dl>

<p>For what it's worth, my camera usually lives in (the equivalent of) evaluative mode, and I switch to spot metering when it gets confused; I rarely bother with centre-weighted. I hope that helps.</p>

Link to comment
Share on other sites

<p>Try this simple test (see attached photo). Set your camera to full program, aperture or shutter priority, whichever you prefer - as long as it's some form of auto-exposure without manual override or exposure compensation. If your camera in evaluative (or matrix, multi, etc.) metering mode delivers the same background exposure with and without the light directly in the frame, use it.</p>

<p>Otherwise, if you get significant underexposure with the light turned on and aimed into the camera lens, you'll want to use center-weighted or spot metering for tricky situations.</p>

<p>Incidentally I tried this test last night on three different cameras. My Nikon D2H failed miserably in Matrix metering mode, a four-stop underexposure. My decade-old Olympus C-3040Z delivered reasonable results, a two-stop underexposure in ESP metering mode (Olympus' equivalent to Evaluative, Matrix or Multi mode). The Ricoh GRD IV wasn't fooled in the least - its Multi metering mode consistently delivers accurate metering in most situations (although it struggled with some stage lighting situations in live opera and ballet, where spotmetering would have been better).</p>

<hr>

<i>Note: The attached image illustrates successful "smart" metering with the Ricoh GRD IV's Multi metering pattern, which evaluates the entire scene and wasn't fooled by the presence of a small patch of bright light.</i><div>00aXtn-477179584.jpg.5ca9f1a61b956b91ca0eaf73b63054b7.jpg</div>

Link to comment
Share on other sites

<p>Lex: To be fair, the matrix meter on the D2H didn't "fail miserably" in matrix mode - it just decided that you wanted a different exposure from the one that you wanted (presumably trying to expose for the light bulb). This is the problem with matrix/evaluative metering: it's based on the camera guessing what you want and, sometimes, it won't guess right. Worse, it's hard to guess <i>when</i> it's going to guess wrong. The solution, really, is to check any images that have a chance of being confused, and try again with another metering mode or exposure compensation if you disagree with the camera about what result is "right".</p>
Link to comment
Share on other sites

<p>At the risk of digressing (since Jennifer has a Canon Powershot, not a Nikon dSLR)...<br>

...after extensively using various Nikons and their matrix metering in a wide variety of situations for many years, I'm willing to call this a failure. A D90 I tried in a local shop when it was a new model didn't fare much better with backlighting.</p>

<p>Since the early 1990s Nikon has touted matrix metering as being capable of adjusting to accommodate many lighting situations. In actual practice it's not much more sophisticated than the full screen averaging metering of previous decades. The one exception I'd make is that Nikon handles backlighting and TTL fill flash situations very well. But in available light situations with backlighting I'm continually riding the exposure compensation dial to avoid underexposure.</p>

<p>This isn't even a very challenging backlighting scenario - certainly nowhere nearly as difficult as some stage lighting. This setup is comparable to any indoor candid situation with, for example, backlighting from a window in daylight, or any bright room lighting without a lampshade. It's exactly the type of scenario that most beginners expect an auto-everything camera to handle gracefully.</p>

<p>If the lamp had been centered in the frame, or against a very dark background, I'd be willing to consider the test beyond reasonable capabilities for most cameras. However even a much older Olympus P&S digicam handled this scenario significantly better. And the Ricoh P&S digicam handled it without any problems.</p>

<p>It's a reasonable test for beginners to get an approximate idea of how their auto-everything cameras perform in typical real world situations, so they can make adjustments if necessary.</p>

Link to comment
Share on other sites

<p>Once you move away from averaging meters to "evaluative" you're inherently saying you'll go with the camera's idea of "right" rather than any definitive measure. I actually don't know that I'd have disagreed with the camera here (did you want the lamp in the foreground compared with the dull, featureless background?) but I agree that some cameras behave less helpfully in common scenarios than others do. Nikon claim to have improved this in their latest updates, although why they had to worry about skin tones in comparison with concentrating on what's in focus (which would seem to me to be the obvious approach) I don't know. It's possible that compact cameras use even more complex heuristics. To bring this back to topic, it would help if manufacturers told the users the algorithm that the camera was trying to use, so that the results were predictable - but they don't. As it stands, we have a "feature", not a "bug".<br />

<br />

The moral of this story is that, no matter how clever the camera's metering is, there's no substitute for checking the image and using exposure adjustment. (And in my case, trying to recover the detail from a RAW file.)</p>

Link to comment
Share on other sites

<p>It may not be clear from the sample photos I attached here...</p>

<p>Those photos were taken with a Ricoh GRD IV, using Multi metering mode (Ricoh's equivalent to Matrix, Evaluative, etc.), in aperture priority. It's what I'd call a successful example of this type of sophisticated metering. In both examples: ISO 400, f/1.9. In the top, the shutter speed was 1/48th second; in the bottom, with the lamp on, 1/68th. The minor differences could be accounted for by the reflection of the lamp off a nearby shelf with a beige PC. Note that the background is virtually identical.</p>

<p>When I'm referring to my D2H Matrix meter failing, I mean it underexposed four stops in this same test. The background was completely black. Considering how Nikon touts Matrix metering, that's a total failure. It's not a feature or operator error when a manufacturer claims its "smart" metering can cope with this type of scenario but consistently fails.</p>

<p>In less extreme situations, such as my street photography downtown where there are deep shadows on the streets and bright sky between the buildings, I typically have to ride the exposure compensation dial to offset the Nikon's tendency toward underexposure. Nikon's Matrix metering is simply nowhere near as capable as Ricoh's Multi metering. However after 40 years in photography, I'm comfortable switching to center weighted average or spot metering as needed to compensate for the shortcomings in Matrix metering. But this is the beginner's forum and expectations are different.</p>

<p>I have no idea about Canon's Evaluative metering for the Powershot series, but it's perfectly reasonable for typical P&S digicam owners to expect a camera to handle this type of scenario.</p>

Link to comment
Share on other sites

<p>Lex - I hear that you're not happy with this result, I'm just suggesting that Nikon's meter didn't "underexpose", it (probably) "exposed for the light" bulb. I very much doubt that the meter went wrong, but rather that it decided to concentrate on a bit of the image that you weren't expecting it to - you're expecting the background to stay the same exposure when the light is turned on, but I don't think there's any hard rule that says this is what the meter should choose to do. Especially on a digital camera, you quite often want to meter for the highlights anyway (and recover the shadow detail later if you need to). Your other cameras may have behaved as you expected, and they may even have made a choice which would be more sensible in the majority of scenarios, but the Nikon was less "wrong" than "different".<br />

<br />

Which, of course, comes back to the perils of trying to guess what a smart meter is going to try to do. Spot metering would let you definitively fix the exposure of the background; evaluative metering is for when you don't really want to spend your time picking a suitable spot and then tweaking the exposure compensation - trusting to the evaluative meter is dangerous, but since it's <i>usually</i> right, it's usually worth a try before resorting to manual controls.</p>

Link to comment
Share on other sites

<p>Thanks everybody. I get the idea I want to double check my photos and possibly spot meter in situations with bright contrasting backlight like sun through a window near the subject or bright sunlight behind the subject and adjust based on whether I do or don't want details or shadows in the background. This was very helpful I will definately try some test shots similar to the examples and maybe some bright sunlight shots to get an idea of what my camera is doing. Can't wait to get a DSLR. Someday when I am not desperately poor. In the meantime thanks for all the advice and support. Keep having fun with everyday life.</p>
Link to comment
Share on other sites

<p>James - why? Put a spotlight on someone on a stage and it's usually them you want to meter for, even if the background is 90% black. Stand someone in front of a sunset and you usually want to get their exposure right. I could argue that you usually want to prioritize metering at the focal point (which is the behaviour of most spot metering by default), but I'm not sure that assuming you should meter for the majority of the scene is necessarily valid. The distinction between "spotlit performer on a dark stage" and "table lamp in a dark room" is subtle.<br />

<br />

I'm not suggesting that I agree with Nikon's metering decision here, I'm just suggesting that their matrix meter may have been making a decision based on unknown but sometimes-justifiable heuristics, and that the problem is likely more to do with the meter deciding that you were trying to achieve something other than what you were actually trying to do, rather than being definitively "wrong". It's quite possible that, under these scenarios, what the meter decides you're trying to do and what you're actually trying to do is different more than it's the same, and that you could complain about the heuristics that Nikon use - and since we know that Nikon played with their metering algorithm in the D4 and D800 when it comes to backlit subjects we could guess that they might have acknowledged this. Still, it's less of an error than "a decision to behave in a certain way". We probably wouldn't complain about it if Nikon actually documented their algorithm and the photographer could predict the results better.</p>

Link to comment
Share on other sites

Jennifer, you would better off reading about these metering modes just in a general way on any instructional

photography website. The camera manuals don't tell you much about them. It doesn't really matter what make of

camera you have, nor whether it's called evaluative, matrix or multi-segment meter. For all intents and purposes, they

all pretty much do the same thing.

 

My own experience with several powershots is that it's usually better to just let the evaluative meter do its thing, apply +

or - exposure if you find your picture looks too dark or too bright and take the shot again.

 

You have to do that anyway with the spotmeter, and with digital cameras, using the old, traditional centre-weighted

meter is for old, traditional people like me. You need a bit of experience with it in order to get what the evaluative meter

would probably have given you anyway. It's usually not worth the bother.

 

The evaluative meter will almost never underexpose shadows, and when it overexposes highlights, it's because that

scene had too wide a difference between the bright parts and the dark parts. If so, it wasn't a well-chosen scene to

photograph in the first place. It also is better integrated with the flash system and the custom picture styles, in my

opinion. I usually just stick with it on my A1200 and S95, and use exposure compensation if I think it's necessary.

 

My S95 has full manual controls, but unless I'm doing fine art type still pictures using a tripod and I don't want the camera to change any settings on me while I work, I don't often bother with that. I'm of the opinion that small cameras are best used as Polaroids on steroids. Concentrate on choosing and framing the scene, and let the camera take care of the rest. Unlike a Polaroid, it costs nothing to take a second picture with appropriate exposure compensation if necessary. It's almost never more than one stop off anyway.

Link to comment
Share on other sites

<p>As an old timer who learnt his habits before all these options became available ... spot metering was the preserve of Asahi and expensive ... I always use centre weighted and if I have a tricky situation I can usually use centre weighted after the style of spot metering by taking the camera up close to the subject. The key is to know how each works and use it in the appropriate situation to get the correct exposure for the subject and how you want to render it. Unless you are quite certain about how to interpret the resulting information that spot metering gives you it can lead you horribly astray.</p>
Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...