PDA

View Full Version : Using a DSLR as a meter



olwick
27-Feb-2010, 07:35
Does anyone use a DSLR as a meter for their LF photography, rather than a hand held meter? I've heard of people doing this, so I'm curious.

If you do, how do you use it? (matrix, spot, etc)

Which metering system is best?

Thanks,

Mark

JasonT
27-Feb-2010, 07:45
I did. It became too much of a PITA! A point and shoot, however, that I can tolerate.

Jack Dahlgren
27-Feb-2010, 09:27
I do. Metering mode doesn't matter. Take a shot and look at the histogram. Then figure out what you are going to do (long tonal scale vs. short, highlights etc.)

Chris Strobel
27-Feb-2010, 09:35
I do. Metering mode doesn't matter. Take a shot and look at the histogram. Then figure out what you are going to do (long tonal scale vs. short, highlights etc.)

And don't forget to add bellows extension factor and reciprocity.Myself, even though I have a Canon G10, I still find it easier to meter with my spotmeter via the zone system.

neil poulsen
27-Feb-2010, 19:46
I haven't tried it yet, but this intrigues me for transparencies. Couldn't the histogram be used to make sure the highlights don't blow out on the transparency? One would select an f-stop that would keep the histogram from clipping on the right.

Through experimentation, one could determine the ASA on the digital camera (for a given transparency film) that would enable the photographer to use the same f-stop on the view camera lens. Or, it could be a combination of using the lowest ASA on the digital camera, plus an off-set in stops in determining the view camera f-stop.

Marko
27-Feb-2010, 20:28
Does anyone use a DSLR as a meter for their LF photography, rather than a hand held meter? I've heard of people doing this, so I'm curious.

If you do, how do you use it? (matrix, spot, etc)

Which metering system is best?

Thanks,

Mark

I don't think anybody would use it rather than a hand held meter, but yes, it can be very successfully used for that purpose if necessary.

Like Jack said, metering system doesn't matter (much) as long as you know how to read the histogram and know how to apply it to the film you use.

Marko

Frank Petronio
27-Feb-2010, 21:04
I use a dslr, I find it reliable. But you have to realize that the highlights will burn out and the shadows will plug at different exposure levels with your digital camera and various film types.

In theory you could map it with film tests... but in practice you can assume the digital will have a bit less range than your negative films. So if it is "safe" on the digital then it will be well within the range of your film.

But yes, compared to a tested Zone System set-up and a good spot meter then you're working like a slob ;-)

And I just set mine on Aperture Priority at f/5.6 to make the math easier to do, just scaling up and down a few stops.

Of course bellows and filter factors need to be included when you use a traditional meter too.

Atul Mohidekar
27-Feb-2010, 22:26
I have been using first an SLR and then a DSLR for metering for years. I equally appreciate DSLR's usage such as a composition previewing tool and to select appropriate LF lens (based on the zoom lens focal length setting) as well as it's metering and histogram capabilities. Many times I pre-visualize the scene by trying out the specific three or four focal lengths, depending on how many LF lenses I'm carrying, with the DLSR zoom lens to decide which LF lens would work best for the scene. Also, with the DSLR I record the capture data digitally for each LF composition.


// Atul

gnuyork
1-Mar-2010, 09:02
I don't, but then my DSLR is as heavy as my 4x5 setup if not heavier. Usually I like to hike with my 4x5 setup so bringing the DSLR is not practical.

I do think it would be useful if someone could make a handheld meter that could also give a histogram.

Jack Dahlgren
1-Mar-2010, 09:38
I think a few point and shoots have histograms, but it would be nice if one of them was tailored for metering and data logging.

Kirk Gittings
1-Mar-2010, 10:34
I use a dslr, I find it reliable. But you have to realize that the highlights will burn out and the shadows will plug at different exposure levels with your digital camera and various film types.

In theory you could map it with film tests... but in practice you can assume the digital will have a bit less range than your negative films. So if it is "safe" on the digital then it will be well within the range of your film.

But yes, compared to a tested Zone System set-up and a good spot meter then you're working like a slob ;-)

And I just set mine on Aperture Priority at f/5.6 to make the math easier to do, just scaling up and down a few stops.

Of course bellows and filter factors need to be included when you use a traditional meter too.

There is allot of discussion on other forums like LuLa about how inaccurate histograms are. They are based on the Jpeg rather than the raw file. As a result many people (myself) tweak the Jpeg "picture style" (even if they are shooting raw) to get the histogram as close as possible to the raw image. It might be a good exercise to do this with the Jpeg rendering to get a histogram that is closer to a particular film and processing. Give me a minute and i will pull up an article about this from LuLa.

Kirk Gittings
1-Mar-2010, 10:42
from http://www.luminous-landscape.com/tutorials/right-hista.shtml


Try the following experiment with your camera. Place your camera on a tripod. Evenly light a test pattern like a step wedge. This will give you a subject with a known long tonality range. Now set your contrast to minimum. Shoot an image and note the histogram on the camera. Note how close the maximum step and the minimum step of the step wedge are apart on the histogram. Now set the contrast to maximum. Repeat the experiment. You will now find that the minimum and maximum steps in the wedge are off scale to the right and left. Now open the two images in your RAW image processor on your computer. You will now see that the two histograms displayed by your processing software are the same. The contrast setting had no affect on the RAW file.

What does all this mean? In order to display the full dynamic range of your camera in the histogram displayed on the back, you must set the contrast to minimum. You now can see the same dynamic range in the histogram on the camera as you will see it in your RAW processor on your computer.

When shooting RAW always set the contrast to minimum so that the histogram displays the full dynamic range the camera can capture. This makes for rather dull images on the back of your camera, but gives you better information about exposure and clipping.

Taking that into account then read this.

http://www.luminous-landscape.com/tutorials/understanding-series/understanding-histograms.shtml

There are other discussions that push this "how to get an accurate histogram" further and I will try and find them.

Jack Dahlgren
1-Mar-2010, 11:29
When you are dealing with +/-10% variance on shutter speed, changing light conditions, bellows extension, tilt and swing etc. it would not seem that a huge amount of time be spent on fine-tuning the DSLR histogram as it is just going to be a guide.

It is important to know how it will translate into your film results at the upper and lower edges, but beyond that it is probably "close" enough even if you haven't uploaded a custom profile into your dslr.

neil poulsen
1-Mar-2010, 11:38
There is allot of discussion on other forums like LuLa about how inaccurate histograms are. They are based on the Jpeg rather than the raw file. As a result many people (myself) tweak the Jpeg "picture style" (even if they are shooting raw) to get the histogram as close as possible to the raw image. It might be a good exercise to do this with the Jpeg rendering to get a histogram that is closer to a particular film and processing. Give me a minute and i will pull up an article about this from LuLa.

Well, bowl me over. I didn't realize this. Hmmm. Thanks for this information.

neil poulsen
1-Mar-2010, 11:47
. . . In theory you could map it with film tests... but in practice you can assume the digital will have a bit less range than your negative films. So if it is "safe" on the digital then it will be well within the range of your film. . .

Being more interested in transparency film for this discussion, I'm wondering how the dynamic range compares to that of transparencies? And anyway, I'm thinking that one primarily controls for the highlights with transparencies and lets the shadows fall where they may. If a histogram can control for highlights well, then that's it right there. Thinking about it in this fashion, a histogram is like an efficient summary of millions of little spot readings.

Jack Dahlgren
1-Mar-2010, 12:39
If a histogram can control for highlights well, then that's it right there. Thinking about it in this fashion, a histogram is like an efficient summary of millions of little spot readings.

The histogram doesn't control for the highlights, you do. But you are correct, the histogram is a graph of millions of spot readings. If you can translate it to film response, then it is a powerful tool.

Ivan J. Eberle
1-Mar-2010, 14:34
Digital sensors have a much more linear response to light than the characteristic curve of transparency film, so there's still the need to test and/or consult the data sheets on your particular emulsion. (Sensors are also largely immune to reciprocity failure).

The histogram is nevertheless useful for roughly gauging dynamic range in a scene and deciding whether to use transparency or neg film (or a GND).

But I'm finding other, more compelling reasons to carrying a DSLR rather than a handheld meter, besides the histogram.

I already owned several Nikons before I got into LF and these all have a separate 1003 pixel RGB metering sensor that's arguably more accurate under more difficult lighting and color temperature than any single Silicon CdS or Gallium Arsenide cell. Other than portability, a handheld wouldn't buy me much. Flash meter measuring, perhaps, but I don't do enough non-digital studio work to bother.

(And when I want to lighten up, I've got a really spiffy General Electric DW58 Selenium job that's certainly close enough for negative film -- and it doesn't even require batteries!)

mrladewig
1-Mar-2010, 17:34
I came to LF from a DSLR, though I'd shot 35mm film for many years before going to the DSLR.

In my experience, using the DSLR histogram to judge exposure was costing me alot in ruined slide film. Since the histogram does nothing to display a range of exposure values (at least that isn't an option on my DSLR), I couldn't accurately say how many stops the spread of an image would cover. With a pentax digital spot, I can quickly scan high, low and find my intended mid-tone. I don't get that feedback from a histogram, I just see a distribution graph. My E-6 keeper rate went way up when I switched to a hand-held spotmeter. Your mileage may vary.

ethics_gradient
2-Mar-2010, 07:56
I do, although I seldom shoot slides. I'm probably going to buy a dedicated spot meter (or maybe one of those Voigtlander VC shoe-mounted meters) though, as the DSLR can be a bit cumbersome.

Acheron Photography
3-Mar-2010, 07:06
I may be an utter philistine here, but I have been using my Canon S90 as a meter for three or four months, and have been really impressed by it. The results are reliable, and unlike and DSLR, the S90 really is pocketable. If you want to keep weight down and need something that is both an excellent small digital camera and a proxy for a light meter, the S90 is a good candidate.

Brian Ellis
3-Mar-2010, 09:27
"Try the following experiment with your camera. Place your camera on a tripod. Evenly light a test pattern like a step wedge. This will give you a subject with a known long tonality range. Now set your contrast to minimum. Shoot an image and note the histogram on the camera. Note how close the maximum step and the minimum step of the step wedge are apart on the histogram. Now set the contrast to maximum. Repeat the experiment. You will now find that the minimum and maximum steps in the wedge are off scale to the right and left. Now open the two images in your RAW image processor on your computer. You will now see that the two histograms displayed by your processing software are the same. The contrast setting had no affect (sic) on the RAW file.

What does all this mean? In order to display the full dynamic range of your camera in the histogram displayed on the back, you must set the contrast to minimum. You now can see the same dynamic range in the histogram on the camera as you will see it in your RAW processor on your computer.

When shooting RAW always set the contrast to minimum so that the histogram displays the full dynamic range the camera can capture. This makes for rather dull images on the back of your camera, but gives you better information about exposure and clipping."

O.K., I feel like a dunce but I'll ask the question anyhow. What's he mean when he says "set the contrast to minimum" and "set the contrast to maximum?" I don't offhand remember ever "setting the contrast" on any camera. The contrast of the scene is what it is. Does he have some special camera that I've never heard of or is he using different terminology for something we routinely do or what?

Kirk Gittings
3-Mar-2010, 09:37
Brian, he is talking about the Jpeg rendering on the camera screen and accompanying histogram by adjusting the "picture style" settings in the camera. I think you have 5DII? It is in the picture style menu.

Brian Ellis
3-Mar-2010, 10:01
Brian, he is talking about the Jpeg rendering on the camera screen and accompanying histogram by adjusting the "picture style" settings in the camera. I think you have 5DII? It is in the picture style menu.

Ah, thanks Kirk. I never use the picture styles so I don't know anything about them, I've always just used manual or aperture preferred. I'll have to check them out.

Dr Bellows
4-Mar-2010, 00:18
I've used my Canon S45 as a meter in the past, with good results. When lazy, I just set it to automatic, take a test photo and if it looks good, use the exposure indicated by the camera. When less lazy, I set the camera to spot metering and evaluate shadow and high values to come up with an exposure for my 4x5. I write the digital camera's photo number on my 4x5 exposure record for reference.

When I shoot with a filter, my digital test shot is done trough the same filter to see how it looks and to compensate for filter factor. This is a real help for B&W through colored filters.

I have yet to use a DSLR for this purpose due to weight and space. I plan a National Park trip soon and that will pose a dilemma. There's no way I'm leaving the 4x5 home, but I would also like to bring the DSLR since it is more flexible and makes better photos than the compact camera. I'm trying to coax a friend that shoots the same brand of DSLR to join me so we can share lenses. Since I can't talk any of my friends into winter backpacking, it turns into day hikes anyways, so the weight is never too bad, but space may require the full-frame pack.

Jerry Kaiser
4-Mar-2010, 13:03
I shoot 8x10 so the film and developing adds up fast. I use my Nikon DSLR on manual modes to take my spot meter readings and set up a digital shot. I then bracket with the digital and use the camera screen and other options to pick which of the bracketed digital pics I like best. Then I cross that back to the spot meter dials and then take the readings off the spot meter for my LF lens.

Example: I lock the Nikon to ISO 100 since that is the film I use; I take a spot meter reading (say 14 for this example);then you have to know about how much F-stop you want for your LF shot (so say I want my LF shot at near max of f-64 which gives a shutter of 1/4 second); then I manually set the Nikon to near max of f-22 which gives a shutter of 1/30 of a second (all using the dials on the spot meter setting of 14). Then I bracket the shot by adjusting shutter or f-stop depending on the end result I am chasing for the film; I pick the best picture from the back of the camera, look at histograms, check for any overexposures, etc; I then look at the settings on the Nikon for that shot (say I chose f-22 at 1/45 of a second); then by setting the spot meter dials with these settings it it shows an equivalent spot meter reading of about 14.5; then you can see the settings for the LF the usual way and shoot 1 picture instead of 2 or 3. This works really well for me and the film/developing savings have paid for the Nikon a long time ago.

Things like the amount of F-stop, desired shutter speed range, focus, etc that you want your film to be exposed to, are still from experience. All the DSLR does is help you stop bracketing with film or even avoiding taking a shot sometimes. I also compose the picture sometimes using the DSLR if I am in a hurry or not sure if I really want to shoot or not. You just need to know where to set the DSLR zoom to make the field of view similar to the lens on your LF camera and then you can look at the picture composition on the back screen of the DSLR.

Hope this is clear, if not email me. Jer

Alan Davenport
4-Mar-2010, 13:15
I'm amazed anyone ever got a decent exposure before there were histograms. A histogram is nothing but an artifact of the digital processing of the image; it tells you nothing you can't get from a meter.

For transparencies, pick the brightest part of the scene where you need to retain detail on the film and meter that area. Set your exposure to 1-2/3 or 2 stops more than the metered value (you'll have determined in advance how much difference to allow) and make the shot. Voila, no blown highlights.

Use an analog meter. They're more precise.

Robert Hughes
4-Mar-2010, 13:25
I'm amazed anyone ever got a decent exposure before there were histograms...
Use an analog meter. They're more precise. Imagine that people got decent photographs before meters, ISO ratings, or any of that fancy 20th Century stuff. Tell it to Mr. Brady...

Sunny 16 works as well as it ever did.

Marko
4-Mar-2010, 14:15
I'm amazed anyone ever got a decent exposure before there were histograms. A histogram is nothing but an artifact of the digital processing of the image; it tells you nothing you can't get from a meter.

For transparencies, pick the brightest part of the scene where you need to retain detail on the film and meter that area. Set your exposure to 1-2/3 or 2 stops more than the metered value (you'll have determined in advance how much difference to allow) and make the shot. Voila, no blown highlights.

Use an analog meter. They're more precise.


No they aren't. A DSLR you own will trump any light meter you don't own. Which is a number one reason why people keep asking this question.

A histogram can tell you much more than a simple needle on a light meter if you bother to learn how to interpret it. Calling it an artifact is simply ignorant.

I own both and use a spotmeter for LF most of the time. If I didn't have one or couldn't afford one, I could (did) get along just fine without it.

No need for bloviations about good old times being ruined by evil new technology. It's getting really old, to the point of becoming demented. And even the technology in question is not exactly new any more either.

Alan Davenport
4-Mar-2010, 16:38
A histogram can tell you much more than a simple needle on a light meter if you bother to learn how to interpret it. Calling it an artifact is simply ignorant.

Your histogram is an artifact of the processed image that was created by your digital camera, using responses programmed into its processor. When you consult the histogram, you aren't measuring the original scene; you're just seeing what resulted from the processing done in-camera. That's an artifact.

I'm not saying that DSLRs don't make serviceable light meters, but it's delusional to believe digital technology is somehow required to make 100 year old film technology work to its best effect.

Jack Dahlgren
4-Mar-2010, 17:10
Alan,

The movement of the needle or display is just an artifact of the processed photons in your lightmeter using responses programmed into its processor (or by current passing through a coil made by someone somewhere). When you consult the light meter you aren't measuring the original scene, you're just seeing what resulted from the processing done in-lightmeter. That's an artifact.

No one is saying is a DSLR necessary to make film work to best effect. It is delusional to think that someone would insist that digital technology is somehow required to make 100 year old film technology work to its best effect.

Marko
4-Mar-2010, 18:21
What Jack said.

A quick look at Webster produces two definitions of the term "artifact":


1.
a: Something created by humans usually for a practical purpose; especially: an object remaining from a particular period <caves containing prehistoric artifacts>

b: Something characteristic of or resulting from a particular human institution, period, trend, or individual

2. A product of artificial character (as in a scientific test) due usually to extraneous (as human) agency

A histogram is not created by humans, it is immaterial - a graphic display of numeric measurement results - therefore it cannot be an artifact.

Film, on the other hand, seems to satisfy all the above definitions... :D

Now, shall we leave the dead horse alone or should we keep looking up word definitions? Say "delusional"...

Robert Hughes
4-Mar-2010, 18:29
A quick look at Webster produces two definitions of the term "artifact":
2.A product of artificial character (as in a scientific test) due usually to extraneous (as human) agency
A histogram is not created by humans, it is immaterial - a graphic display of numeric measurement results - therefore it cannot be an artifact.
Nonsense - a histogram is an output of a digital light sensor recording a scene, just as much as a jpeg file is, and fits squarely into your definition #2, as much as any other experimental measurement would.

Marko
4-Mar-2010, 18:42
Nonsense - a histogram is an output of a digital light sensor recording a scene, just as much as a jpeg file is, and fits squarely into your definition #2, as much as any other experimental measurement would.

Then so is what you call an "analog" light meter - essentially a digital sensor reading the light intensity of the scene (or its part) and producing a discrete electrical signal as a result. Depending on the particular instrument, the strength of that signal is represented either by the movement of a solenoid with a needle attached or via a digital readout, as in more modern light meters.

A histogram is simply a distribution curve for a certain number of such readings. 10-20 million of them, depending on the vintage of a particular DSLR used.

If you are right, then both are artifacts, only one is much more precise than the other.

Alan Davenport
4-Mar-2010, 19:37
Then so is what you call an "analog" light meter - essentially a digital sensor reading the light intensity of the scene (or its part) and producing a discrete electrical signal as a result.

There ain't nuthin' digital in my Luna Pro F; from the sensor to the needle it's stepless, as in "analog."


A histogram is simply a distribution curve for a certain number of such readings. 10-20 million of them, depending on the vintage of a particular DSLR used.

Ten to twenty million readings, when we're talking about something a human brain has to digest, is beyond even science fiction. Few of us really comprehend even the magnitude of such numbers, much less do we have the computing power in our heads to decipher what they mean. No advantage there.


If you are right, then both are artifacts, only one is much more precise than the other.

So you agree the needle is better! http://home.comcast.net/~w7apd/public/rofl.gif

Jack Dahlgren
4-Mar-2010, 20:12
So you agree the needle is better! http://home.comcast.net/~w7apd/public/rofl.gif

The needle and the damage done...

Robert Hughes
5-Mar-2010, 09:57
Uh, what were we talking about? :confused:

Marko
5-Mar-2010, 13:30
At first, we were having a discussion about using a DSLR as a light meter for LF. Then we defined the meaning of "artifact" but we never got to delusion (http://www.merriam-webster.com/dictionary/delusion):



1: the act of deluding: the state of being deluded

2.
a: something that is falsely or delusively believed or propagated

b : a persistent false psychotic belief regarding the self or persons or objects outside the self that is maintained despite indisputable evidence to the contrary;

also: the abnormal state marked by such beliefs


;)