| Author | Thread |
|
|
10/10/2005 05:58:01 PM · #1 |
I'm interested in getting some opinions (and pros and cons) on the use of an external light meter as opposed to the camera's built-in meter. Most of us have read the "meter is stupid" type of materialâ€Â¦and I researched all types of different methods for zone metering, integrated metering, bracketing, using a gray card, etc. The 20d does not have spot metering, and it seems to me (given a lot of what I have read) that spot metering appears to be one of the easiest and more precise methods for getting the "perfect exposure" (to quote Jim Zuckerman), so I was thinking of purchasing an external meter.
However, before incurring the expense of yet ANOTHER piece of equipment. I wanted to hear from the masses - any recommendations/suggestions? (fyi, I don't intend to do too much studio work, mostly outdoor stuff). |
|
|
|
10/10/2005 06:34:07 PM · #2 |
Here's a decent read on the different metering possibilities: //www.shutterbug.net/features/0703sb_spot/
True spot meters cost the earth. They are very useful, but there's a learning curve involved. I find that in landscape type work I have sufficient knowledge to calculate my own variations of the camera's recommended exposure based on what my eyes show me of the scene as a whole and my own knowledge of how the in-camera metering works. I used to use spot meters all the time for zone system shooting, but I feel no need for one now with the 20D.
Here's B&H page on spot meters; check the prices... spot meters.
R.
|
|
|
|
10/10/2005 06:51:42 PM · #3 |
| Thanks Robert. The article refers the tried and true film latitude method (5 stops over, 2 stops under). Do you find the 20d offers the same latitude, or is closure to a 9-stop latitude? |
|
|
|
10/10/2005 06:58:01 PM · #4 |
For outdoor stuff (landscapes and the like) I think I'd just stick to my internal light meter. The best a light meter will give you is a "spot meter" but even then it is using reflected light to measure the light.
If, on the other hand, your subject is up close and you want to get the exposure right regardless of what color or how reflective the subject is ... then a light meter is INDISPENSABLE. And I am *not* talking about the spot meter aspect of it, because even the spot meter will be wrong.
For example, shooting a bride and groom ... bride is wearing white, groom is wearing black. What do you meter off of? Doesn't matter if you use either your camera's internal light meter or even a spot meter, the reading will be wrong if you meter either the white dress or the black suit.
On the other hand, if you can take an ambient light reading by walking up to and holding the light meter where the bride or groom are standing (with the sphere pointing back towards the camera), you will get a perfect exposure reading every time.
The other situation where light meters are indispensable? When using strobes. Your camera can't measure their output... you're left to guessing using histograms.
|
|
|
|
10/10/2005 07:06:55 PM · #5 |
| dwterry - thanks, good example for when to use an incident meter. Have you ever used a gray card in that situation? |
|
|
|
10/10/2005 07:10:10 PM · #6 |
I have both a minolta auto meter IV F and a minolta spot meter f (which cost me a fortune at the time).
I use the spot meter when shooting film. In the studio I never use a meter anymore. I know the ouptut of my lights at given distances and can dial them in in about 3-4 test shots for the look I'm going for.
Here's a link to my...spot meter F
|
|
|
|
10/10/2005 10:51:25 PM · #7 |
Originally posted by macpapas:
dwterry - thanks, good example for when to use an incident meter. Have you ever used a gray card in that situation? |
You make a good point. I used to use the gray card method before I bought a light meter. It works quite well and it's certainly a lot cheaper than a light meter (which I actually purchased to use with my strobes and only later started using outside as well).
|
|
|
|
10/10/2005 11:10:02 PM · #8 |
Originally posted by dwterry:
For outdoor stuff (landscapes and the like) I think I'd just stick to my internal light meter. The best a light meter will give you is a "spot meter" but even then it is using reflected light to measure the light.
If, on the other hand, your subject is up close and you want to get the exposure right regardless of what color or how reflective the subject is ... then a light meter is INDISPENSABLE. And I am *not* talking about the spot meter aspect of it, because even the spot meter will be wrong.
For example, shooting a bride and groom ... bride is wearing white, groom is wearing black. What do you meter off of? Doesn't matter if you use either your camera's internal light meter or even a spot meter, the reading will be wrong if you meter either the white dress or the black suit.
|
This is true as far as it goes, but that's not the whole story; the use of a spot meter requires more than just metering and setting exposure. The whole point of a spot meter is to be able to take very accurate readings on the brightest and drakest parts of the image you wish to retaind etail in, and then calculate the dynamic range required to see detail in both areas. Armed with this information, you calculate an exposure that avoids utterly blowing the highlights and losing the shadows to darkness.
When you know the dynamic range you're working with, you're armed with as much information as you need to make intelligent decisions. But you still need to "place" your exposure accurately. This, actually, is what zone system is all about.
If you're working from RAW, you can use a form of zone system pretty effectively. You can place your highlights as high as zone 9 and bring them down to zone 7 in RAW processing and have full texture; or place them on 10 and bring them down to 8 where they're not really textured but are not totally blown out. Somewhere in that range.
To "place" a highlight, meter it and consider the indicated exposure a zone 5 exposure. Two stops more than that plces the highlights to zone 7. 4 stops more places them to zone 9 and RAW can bring them back down to zone 7.
Suppose the dynamic range between deep shadows and highlights is 6 stops. You want the deep shadows on zone 3, where a trace of detail is retained. That would place the highlights on zone 9, too bright for detail. RAW processing would allow you to bring the highlights down 2 stops to zone 7, and you're good to go. The key thing is to force the exposure over on the right side of the histogram, where more detail can be packed in, then deal with it afterwards in processing. But you have to be really precise, 'cuz a little too far blows the highlights irredeemably.
That's how a spot meter is used, anyway :-)
Robt.
|
|
|
|
10/10/2005 11:21:51 PM · #9 |
Originally posted by bear_music:
...If you're working from RAW, you can use a form of zone system pretty effectively. You can place your highlights as high as zone 9 and bring them down to zone 7 in RAW processing and have full texture; or place them on 10 and bring them down to 8 where they're not really textured but are not totally blown out. Somewhere in that range.
To "place" a highlight, meter it and consider the indicated exposure a zone 5 exposure. Two stops more than that plces the highlights to zone 7. 4 stops more places them to zone 9 and RAW can bring them back down to zone 7.
Suppose the dynamic range between deep shadows and highlights is 6 stops. You want the deep shadows on zone 3, where a trace of detail is retained. That would place the highlights on zone 9, too bright for detail. RAW processing would allow you to bring the highlights down 2 stops to zone 7, and you're good to go. The key thing is to force the exposure over on the right side of the histogram, where more detail can be packed in, then deal with it afterwards in processing. But you have to be really precise, 'cuz a little too far blows the highlights irredeemably.
That's how a spot meter is used, anyway :-)
Robt. |
I've been reading a lot about this whole zone system stuff, and then you write this....where have you been this whole time!? LoL
It's a very simple, yet extremelly effective way to understand what seemed to me, impossible! Those damned zones!
thx Robt!!
|
|
|
|
10/10/2005 11:25:36 PM · #10 |
Originally posted by scuds:
I've been reading a lot about this whole zone system stuff, and then you write this....where have you been this whole time!? LoL
It's a very simple, yet extremelly effective way to understand what seemed to me, impossible! Those damned zones!
thx Robt!! |
Sheesh, man, I've been driveling off and on about zone system as long as I've been here; I used to teach it, see? You managed to miss all those threads?
Robt.
|
|
|
|
10/10/2005 11:39:22 PM · #11 |
| So Robert, do you use the Zone system with your 20d meter (pointing at highlights and shadows - or their near proximity), or do you eyes just see f/stops, after years of taking pictures. :-) |
|
|
|
10/11/2005 12:27:56 PM · #12 |
Originally posted by macpapas:
So Robert, do you use the Zone system with your 20d meter (pointing at highlights and shadows - or their near proximity), or do you eyes just see f/stops, after years of taking pictures. :-) |
As I commented in my first post, "I find that in landscape type work I have sufficient knowledge to calculate my own variations of the camera's recommended exposure based on what my eyes show me of the scene as a whole and my own knowledge of how the in-camera metering works. I used to use spot meters all the time for zone system shooting, but I feel no need for one now with the 20D."
Interestingly enough, for the last 10 years or so of working as an architectural photographer, I rarely even used a light meter when working with the view camera. Only in "difficult" lighting situations did I feel it was necessary. Experience pretty well takes care of that. I use the 20D on matrix mode and review the exposure mentally as I shoot, on the average landscape shot. I'll use the exposure compensation dial to fine-tune and I'm pretty much always in the ballpark.
Robt.
|
|
|
|
10/11/2005 12:35:30 PM · #13 |
Originally posted by bear_music:
If you're working from RAW, you can use a form of zone system pretty effectively. You can place your highlights as high as zone 9 and bring them down to zone 7 in RAW processing and have full texture; or place them on 10 and bring them down to 8 where they're not really textured but are not totally blown out. Somewhere in that range.
To "place" a highlight, meter it and consider the indicated exposure a zone 5 exposure. Two stops more than that plces the highlights to zone 7. 4 stops more places them to zone 9 and RAW can bring them back down to zone 7.
Suppose the dynamic range between deep shadows and highlights is 6 stops. You want the deep shadows on zone 3, where a trace of detail is retained. That would place the highlights on zone 9, too bright for detail. RAW processing would allow you to bring the highlights down 2 stops to zone 7, and you're good to go. The key thing is to force the exposure over on the right side of the histogram, where more detail can be packed in, then deal with it afterwards in processing. But you have to be really precise, 'cuz a little too far blows the highlights irredeemably.
That's how a spot meter is used, anyway :-)
Robt. |
This is a wonderful explanation. Thank you.
A couple of questions if I may.
1. I assume that you can do all this using TTL spot metering.
2. You mention processing in RAW etc. I am assuming that there is no way of selectively adjusting highlight and shadow exposure in RAW. So if you bring back the highlights by adjusting your exposure setting then you're also losing some of your shadows at the same time. Or is there some wonderful way of doing this selectively?
Thanks
|
|
|
|
10/11/2005 12:59:39 PM · #14 |
Originally posted by joezl:
This is a wonderful explanation. Thank you.
A couple of questions if I may.
1. I assume that you can do all this using TTL spot metering.
2. You mention processing in RAW etc. I am assuming that there is no way of selectively adjusting highlight and shadow exposure in RAW. So if you bring back the highlights by adjusting your exposure setting then you're also losing some of your shadows at the same time. Or is there some wonderful way of doing this selectively?
Thanks |
1. To a certain extent, yes, but not very precisely. There's no dSLR of which I am aware that has the really narrow (1 degree or less) metering angle of a true spot meter. If you are working with a tripod, say, and photographing a wide landscape, you can put long glass on and do your metering calculations by pointing at the areas youw ant to meter, then calculate your exposure and lock it in manually, then replace with the WA glass to shoot. In other words, you're using the camera + telephoto as a separate meter... But this is cumbersome, and experience eventually would make it unnecessary except in really extreme cases.
2. Youy can selectively adjust shadow and highlight exposure in RAW processing by watching the histogram and adjusting both exposure and contrast. When you combine this with shadow/highlight masking techniques in photoshop you can make some really dramatic adjustments. The key here is that it's in the nature of our sensors that more information is packed into the bright end of the scale than the dark end, so it's better to bias in that directiona nd then compensate downwards in processing, always assuming you don't go so far up in the bright range as to lose all detail in the highlights. The RAW processing allows you to recover highlights that are overexposed within certain limits, and that's the way to go if the dynamic range is greater than can be "naturally" expressed.
The very best way to work is with multiple exposures of the same scene combined in post-processing, but that's a whole other subject...
R.
|
|
|
|
10/11/2005 01:14:02 PM · #15 |
Sounds like money is better invested in a bigger card and shoot in RAW mode. I was thinking of a light meter also. But by reading this I think I will buy a bigger card and start shooting in RAW
|
|
|
|
10/11/2005 01:22:58 PM · #16 |
Originally posted by rex:
Sounds like money is better invested in a bigger card and shoot in RAW mode. I was thinking of a light meter also. But by reading this I think I will buy a bigger card and start shooting in RAW |
That would be my gut reaction, yes. A decent incident light meter is very nice for landscapes, though, as an alternative to camera metering, and much less expensive.
Robt.
|
|
|
|
10/11/2005 01:41:23 PM · #17 |
bear_music: "The key here is that it's in the nature of our sensors that more information is packed into the bright end of the scale than the dark end, so it's better to bias in that directiona nd then compensate downwards in processing, always assuming you don't go so far up in the bright range as to lose all detail in the highlights."
HA! THAT's the information I was looking for! (reference my earlier comments on "film latitude"). Thanks Robert! |
|
|
|
10/11/2005 06:48:46 PM · #18 |
Originally posted by bear_music:
A decent incident light meter is very nice for landscapes, though, as an alternative to camera metering, and much less expensive.
Robt. |
Huh!
OK - I'm now totally lost here. I thought that landscapes would be when you would need the spot meter so that you could take readings off shadows and highlights that are away in the distance rather than hiking a few miles to take your readings. But clearly I'm missing some fundamentals.
|
|
|
|
10/11/2005 07:21:41 PM · #19 |
Once upon a time I lusted after a light meter. But then I realized that with digital (instant feedback, histograms) there was no reason in the world for me to bother carrying one.
I see light meters as tools used by professionals who have no time to dilly-dally around with exposure settings. Not rank amatuers, like myself, with plenty of time to adjust an exposure.
In the time it would take me to futz with a lightmeter... to get it out, get the reading, and then take the shot... I can take a shot and view the histogram, make any needed corrections and take another, corrected exposure if I need to. EZPeezy.
So I decided that, for me at least, there was no need to complicate the matter any further. |
|
|
|
10/11/2005 08:03:36 PM · #20 |
Originally posted by mcmurma:
Once upon a time I lusted after a light meter. But then I realized that with digital (instant feedback, histograms) there was no reason in the world for me to bother carrying one.
I see light meters as tools used by professionals who have no time to dilly-dally around with exposure settings. Not rank amatuers, like myself, with plenty of time to adjust an exposure.
In the time it would take me to futz with a lightmeter... to get it out, get the reading, and then take the shot... I can take a shot and view the histogram, make any needed corrections and take another, corrected exposure if I need to. EZPeezy.
So I decided that, for me at least, there was no need to complicate the matter any further. |
For landscapes, snaps etc, I would say that you are probably correct, however, even for amateur photographers, a good handheld meter can be essential, especially in the studio.
|
|
|
|
10/11/2005 11:42:03 PM · #21 |
This is classic timing. Just got home a little while ago and Popular Photography magazine came in the mail. One of the top articles “10 Reasons You Still Need a Handheld Meter”. You readyâ€Â¦
1. Speed – easier the removing camera from tripod to meter with, change lenses, etc.
2. Real Spotmetering – cameras that have spotmeters tend to meter at 10% or more
3. EV levels – though some cameras offer this as well
4. Contrast Range at a Glance – handheld meters often give you range on viewfinder, so no more calculating f/stops in your head
5. Incident readings – built-in meters measure reflected light
6. Quick Averaging – handheld takes multiple readings in succession and gives you average exposure
7. Authoritative 2nd opinion – in case you don’t like what the camera is telling you
8. Save on histogram waste – you may not be able to take multiple pictures to get the right histogram – why not get the exposure right the first time
9. Flash measurement
10. Combo readings – take flash measurement and ambient light reading and compare at a glance.
Just passing on the info! Viewer discretion is advisedâ€Â¦
|
|
|
|
10/12/2005 02:28:59 AM · #22 |
Those are all nice reasons.
But my favorite is:
7. Authoritative 2nd opinion – in case you don’t like what the camera is telling you.
WhooHoo! Stupid Meter :) |
|
|
|
10/12/2005 02:59:57 AM · #23 |
Originally posted by joezl:
Originally posted by bear_music:
A decent incident light meter is very nice for landscapes, though, as an alternative to camera metering, and much less expensive.
Robt. |
Huh!
OK - I'm now totally lost here. I thought that landscapes would be when you would need the spot meter so that you could take readings off shadows and highlights that are away in the distance rather than hiking a few miles to take your readings. But clearly I'm missing some fundamentals. |
Look at it this way; an incident light meter measures the light that's falling on a scene. If you use the exposure the incident light meter gives you, in theory all areas of the scene will be faithfully displayed. A zone 3 will be a zone 3, a zone 5 will be a zone 5, a zone will be a zone 7. Even if the image is 90% composed of zone 4 and the remainder is zone 7.
In the same situation, the camera's matrix metering will average everything out and yield an exposure based on the assumption that the tones in the scene would yield a zone 5 if they were dumped in a bucket and mixed up smoothly. So you will end up with a shot where the zone 4 part is rendered as zone 5, and the zone 7 part is bumped up to zone 8, the no-detail threshold. Now, in post-processing, you'll have to bring it down to what you wanted.
This isn't a bad thing, in the sense that a stop of "overexposure" is workable and guarantees a lot of detail & texture to work with in the putative zone 4 that's been bumped to zone 5, but you can see it would be a problem if a lot of the image area were in deep shadow and you wanted it to stay that way. The camera metering would force the shadows way up and the highlights would start to blow.
In fact, this is why we have EV compensation; so you can look at a scene and be sort of a "mental incident light meter" in your mind, and decide "Hmmm, there's a lot of darker tones in this scene, I'm gonna wheel down my exposure to compensate for that" or "We're metering a lot of reflection and clouds here, don't wand 'em to go gray and find my shadows underexposed, bteer bump it up a stop-and-a-half!".
You get the idea. You can avoid all that with an incident light meter, see? You just get an EV based on the light that's falling on the scene and that's a great starting point.
The only time you really NEED spot metering in landscape work is when you have a really extreme tonal range and you want to know how much the spread is so you can compensate for it in some way. But there's an easier way to do that than spot meteri8ng, although spot metering is fine if you can do it. Still, it's not like we're paying for film here, so the easy way to cope is to bracket your exposures. Also, when you know you have a real contrast problem (and experience will teach you that soon enough) you can, and should, go into your camera menu and shoot at low contrast. If you're shooting RAW, that's exactly what you get, btw. Then you dial in the contrast while viewing the histogram in the raw processor.
To see for yourself the effect of the matrix-metering mode on exposure, try this; set up your camera on a tripod and focused in close so it covers an 8x10 inch area, say. Set it on auto exposure and do a shot of a white paper, a shot of a black paper, and a shot that's half white paper and half black paper. Do this in jpg with normal contrast. Get all 3 images up on screen side-by-side with no post processing and look at them. The white paper and the black paper will be rendered in similar, if not identical, shades of gray and the mixed-paper will have much closer to a true black and a true white on it.
If you have an incident light meter, obviously the indicated exposure for all three shots will be the same, and the tones will be the same for all three shots; white whites and black blacks, more or less.
Robt.
|
|
