fukkin meters, how do they work

tell me about meters, metering, and how to do it

Incident light meter - handheld thing with a dome, reads the levels of light falling on the dome. Most accurate, if you're in a controlled lighting environment.

Reflected light meter - reads the light on a scene against a control (used to be an 18% gray spot, now it's mostly computer algorithms since Nikon started matrix metering in the late '80s).
Most common are 'averaging' meters (every brand seems to have their own name for this, but it's all the same basic concept) - they read the light in an entire scene and try to balance it out as well as possible.
Fancier cameras will have a spot meter - you read the light at a single point (1-degree, 5-degree, etc.) in the scene. Used if you need to expose highlights absolutely correctly, for instance.

how much of this stuff is obviated by shooting RAW?

I think if you shoot RAW, you just want to slightly underexpose, right? Or, take a shot, and see if you clipped any highlights. If so then just do EV compensation minus 1/3 - 2/3 stops or so.

Do people have real-world 18% gray equivalents? Asphalt? Grass? I'm usually meterless, but now that I'm using the Rollei with the built-in meter I like to point it at something from time to time for a reality check.
When I'm not meterless, this is my preferred guy:

http://farm4.static.flickr.com/3291/3136827877_22c3b61649_m.jpg
Gossen luna-pro by pelukas2000, on Flickr

think I read that for shooting outdoors in the sun that the deepest blue in the sky is 18%

for people shooting I like to meter off my hand, which is usually one stop above 18% gray

it took me along time to realize that 18% gray means 'middle gray'. and that manufacturers sometimes calibrate their meters for 14%, or 16% gray...

middle gray is kind of a cool concept. so if you have a light meter and you meter off of your hand outside. it means that in the picture, your hand will fall in the exact middle of the luminance range captured by your camera. in terms of stops, there will be let's say 5 stops of light below your hand's value and 5 stops above it, at least. of course if there happens to be something in the picture that exceeds that range (i.e. the sun) it will clip the highlights because it exceeds the number of stops you can capture with that shutter/fstop setting. likewise if there is something in deep shadow in your picture.

that's about as basic as I can break it down.

http://cdn.7static.com/static/img/sleeveart/00/000/403/0000040328_350.jpg

Incident light meter - handheld thing with a dome, reads the levels of light falling on the dome. Most accurate, if you're in a controlled lighting environment.

it's become received wisdom that this is the best way to meter, and in some sense it's true - if you use incident metering you will always get a usable negative. however it might not be the negative you want - you might have wanted to make a skin tone brighter than what incident metering will give you.

it also took me a long time to figure out how this works. you put the meter where your subject is and point the dome at the camera, making sure the dome lines up in a straight line w/ the camera lens. the reason you do this is that the dome catches all the light falling on the dome, like milo says - including the light that's getting reflected off the ground, which is why you want to make sure the dome is parallel w/ the lens. the reading that you get will be exactly in the middle of the range of light values that the lens sees, and your subject will fall in the exact middle of the dynamic range captured. this is, famously, how you capture a white cat in a snow storm or a black cat in a coal mine.

some scenes have really wide dynamic ranges (i.e. a city where there's harsh sunlight and deep shadows). this may be something crazy like 20 or 25 stops? (spitballin) even negative film can only capture something like 11 or 12 stops. this is always triage in a sense - do you stop down your aperture to make sure your highlights don't get blown, losing everything in the shadows? or do you open up to get everything in the shadows, but risk blowing some highlights?

some scenes have a really narrow dynamic range - a beach on an overcast day maybe. or a snowstorm, where everything is white. or at night, when everything is dark. (throw a streetlight in there and then you have a problem!)

I use one of these

http://ecx.images-amazon.com/images/I/41UvYsslpmL._SL500_AA300_.jpg

it's fun to walk around with it and just take readings, even when you're not shooting. it helps you 'learn the light'. took me about a year but I can't remember the last time I used a meter outdoors, indoors however is a different matter since it's hard to gauge different kinds of indoor lighting (incandescent, tungsten, halogen).

I should add, from experience, that reflected readings are vastly better than incident readings when you're shooting landscapes.

another pro tip, when metering a landscape using a reflected light meter that has a relatively large beam (40 degrees in my case), point the meter about 2/3 at the ground, 1/3 at the sky. if you point it straight on the sky is gonna be so bright that it's gonna put itself as the middle value for your meter and you're gonna underexpose the scene.

using a fully manual camera without a meter puts you in the same state of mind as chopping your own firewood, tanning your own leather. YOU control the magic etc.

ok I am trying to understand the dome at the camera thing. is the idea that an incidence meter is measuring the light falling directly on the subject?

and a reflective meter is measuring the light being, uh, reflected from that subject?

hence incidence meters being used by studio photogs since they can actually walk over to the thing in question, right?

also: spot meters are reflective, right, and the meters in a modern digital camera are some type of weighted average? (ie preferentially read the thing in the middle when computing the fstop)

pretty much, yeah. Not just studio, though. When I was shooting (outdoors) with a view camera, I was using color negative film and relied on an incident meter. I liked it better than a handheld reflective meter and could tell if the area where I was standing was darker/lighter than what I was shooting (if I couldn't get near enough the subject to just meter there).

ok I am trying to understand the dome at the camera thing. is the idea that an incidence meter is measuring the light falling directly on the subject?

and a reflective meter is measuring the light being, uh, reflected from that subject?

hence incidence meters being used by studio photogs since they can actually walk over to the thing in question, right?

― remembrance of schwings past (gbx), Friday, September 9, 2011 7:10 PM (2 hours ago) Bookmark

yeah, it took me a long time to figure out the difference between light falling on an object and light reflected. so like, think about a single flashlight beam in a dark room and two things: a lump of coal and a mirror. you shine the flashlight at the mirror and you take a reflected light measurement and an incident light measurement of the mirror. it's going to come out pretty similar - since the mirror is probably reflecting like 90% of the light or something (w/e physics) the amount of light falling on the mirror is gonna be similar to the amount that's being reflected.

replace the mirror w/ the coal though, and then take the same two readings. the incident reading is gonna be the same since it's still measuring the flashlight beam that's falling onto the coal. the reflected reading is gonna be off though, since the coal is absorbing what, 50% of the light or something? and you're probably gonna.. hmm.. overexpose.

This is an interesting article putting forward the idea that you should expose-to-the-right with digital wherever possible.

It makes sense from to me from an audio-recording PoV (tracking as hot as possible without clipping) and yet I've been consistently underexposing when metering on dark objects for years so as to get the "true" black tones. But that pushes shadow detail down into the noisefloor and pulling those areas back up in Lightroom just introduces noise. Better to have an overexposed image (without clipping) and deepen the blacks in post - no increase in noise. (I don't mean +1EV when metering on someone's black jacket, just 0EV and sort out the inevitable grey jacket and slight wash-out later).

(I'll still underexpose in generally dark conditions though, to get a faster shutter speed or to avoid ISO 3200 (not great on my 40D)).

With almost every digital camera I've had, I had to always dial in +1/2 or +1 EV for exposing to the right. But I think exposure is almost an overrated issue, given a usable negative/file - if you blow a couple of highlights (unless they're overly distracting) or don't have shadow detail in spots, that's probably not what kills the value of a photo.

Content over focus and exposure, every time.

well sure, but for n00bs it's a good to get an understanding of what "good exposure" even means. like, until you and dayo explained incident light meters, i had no idea they even existed! i've only used built-in reflected meters before. and on all the film cameras i've used (canonet, leica, rollei SL35) i've never really gotten great exposures consistently.

yeah it's also worth considering how the different media types work - you would think they fall into neat categories, but they don't!

negative film - pretty unfuckwithable and just by being a 'negative' of a positive has many cool feature that you wouldn't expect, sort of like how mirrors just aren't reflections of us but are pretty non-intuitive (everything is reflected, how you look in a mirror isn't how you look in real life)...

basically with negative film, overexposure is generally handled pretty well whereas underexposure is the death. the reason being, when you underexpose, vast swaths of the light sensitive particles just don't get activated, and get washed away during development. so you end up with large patches of thin or clear film where there is no silver, no detail at all recorded. on the other hand, when you overexpose, I think you almost never exhaust the particles available to be exposed. for example, I think Ilford advertises FP4 as being able to handle up to '6 stops' of overexposure? in a traditional darkroom printing room, you can compensate for massive overexposure, but not for underexposure. look to the black and white work of ralph gibson - he intentionally 'overexposes' and overdevelops his film to get that look. his negatives must look like strips of nori.

on a related note, a lot of people routinely 'overexpose' color negative film by 2/3 or one stop to get richer, more saturated colors.

slide film - still not quite sure how this works - when you overexpose or push the particles towards the higher end of the spectrum, you risk blowing them (turning them to white) or washing out the colors. when you underexpose, you get dull colors I think? I don't shoot much slide. fwiw I hear modern slide films are much more forgiving than they used to be. some slide films can be pushed two stops, like provia 400x which you can even shoot at 1600, albeit with a lot of lost shadow detail I imagine.

digital sensors - more similar to slides - like michael said, overexpose and you blow highlights, underexpose and you get lots of noise. however, massive improvements have been made towards the sensitivity of digital sensors. the fastest true ISO negative film can give is like 1200 ISO, whereas dark-eaters like nikon sensors can go up to something like 6400 or even 12500 with usable results? have heard of people underexposing by as much as three stops and getting something perfectly usable out of the RAW file. it's kind of crazy and the opposite of negative film, which allows room for massive overexposure.

boy this stuff is confusing

look to the black and white work of ralph gibson - he intentionally 'overexposes' and overdevelops his film to get that look. his negatives must look like strips of nori.

...what does "overdevelop" mean? (i know i could google this stuff, but w/e)

so the way negative film gets developed is that the developer converts exposed silver salts which are unstable to their stable form, which is what you see on the negative (the dark stuff is silver left on the clear gelatin base).

when you overdevelop, you leave the negatives in the developer for longer than the standard time for a certain exposure speed. like publishers will publish recommended times for an ISO setting*

it's important to remember that the developer doesn't act on all parts of the negative equally. I have a theory** as to why this is but let's put that aside for now. areas of the negative that received little light and hence have little action get developed by the developer at a different rate than areas of the negative that received a lot of light. overdeveloping tends to develop the highlights much more than the shadow details. hence, overdeveloping usually increases contrast and creates what are called 'blocked' highlights, areas of the negative so dense that it almost becomes impossible to see through. ralph gibson shoots tri-x at 100 (thus overexposing) and develops for I think, 800? (thus overdeveloping). he gets this incredible high contrast neg that must take him ages to print in a darkroom.

*as any old school photobook will tell you, you need to determine for yourself what ISO 'works best for you' since there really is no such thing as a 'right' ISO value for any given scene. there are all sorts of tests that you can do to determine this but they are tedious. books will also tell you that films may not live up to the advertised 'box speed' and that's why you should test. this is true but testing is so tedious, plus usually people on photo forums will have done the testing for you. for example, some people think that the true 'box speed' of ilford HP5 (tri-x equiv) is actually 640. some people think that neopan 400 (tri-x equiv) is actually maybe 320.

**as I understand it, 400 speed film is not just composed of 400 speed silver particles - it's a mix of particles of different speeds, from 50 to 400. my theory is that even in events of massive overexposure, you've exhausted the 400 speed particles but there are 200 and 100 speed particles that aren't as overexposed. the developer continues to develop these slower particles when you overdevelop. whereas with shadow detail, probably only the fastest particles (the 400 particles) were exposed, and when they get developed to 'exhaustion' (i.e. the developer can't develop them anymore) it's done. that's why another old photography maxim is "expose for the shadows, develop for the highlights." I could be talking completely out of my ass here though.

what's even more confusing is that a photographic print of a negative made in a darkroom, is also a negative - photographic paper is itself a negative. you are making a negative from a negative thus getting a positive. there are interesting implications to this as well.

also there are negative films out there that are not tolerant of overexposure at all - ilford pan F being one. I think it's cause the film is only made up of one type of silver particle, so when you overexpose them to the point of black, that's it, game over man.

http://www.camerasandfilm.com/archives/471

so this is p neat

ayo dayo

when yr using yr incident meter (I've just "assembled" that iPhone meter per the link), how...do you do that? like if I'm metering, say, the bus I'm on, do I take the incident light on my palm and just go with that?

well you'd have to decide upon a subject first. if your subject is just 'interior of the bus' then you'd walk over to where you want to take a pic of, meter by pointing your iphone to where your camera is eventually going to be, and take the reading.

it's going to require a value judgment - are certain areas of the bus darker or lighter than others? you may want to take separate readings for those areas and see what the difference is, if there is any at all.

be careful of pointing the meter towards the windows - likely the light outside is gonna be much brighter (in the daytime) and will affect the reading. but sometimes that may be what you want, depending on what you're shooting!

you generally use reflected metering when metering off your hand - it's commonly accepted wisdom that a caucasian toned hand is gonna be one stop above 'middle grey' in a picture (which generally looks pretty pleasing, people like it when they look 'luminous'). so if your subject is the people on the bus you may want to do that - you risk blowing the highlights if the windows are in the pic, tho, and it's sunny outside

wait I just confused myself. so if you meter off your hand, open up one stop from the reading you get to place yr hand one stop above middle grey. I think that's right. *ponders*

so per the instructions on that link, you do a spot meter off your hand with your camera, overexpose one stop. then you take an "incident" reading with the iPhone app and stop down the EV until it matches your camera. this "calibrates" the iPhone and it can now be used as an incident meter.

haven't read the post yet, will later

fwiw, i took some shots outside (cloudy), and the iphone generally underexposed by about 2 stops as compared to what the center-weighted reflective light meter in my GF1 wanted (when taking a picture of a dark-ish building). inside, in this airy architecture library, the iphone meter is overexposing by four stops. \(•_°)/

http://www.camerasandfilm.com/wp-content/uploads/2011/06/71110019-3.jpg

gonna point out that though, that this is a great photograph, but you can see her face falls right in the middle value of brightness in the pic - might have been improved by opening up one stop? this is what incident metering does, though - it puts the thing behind the dome at the middle of the brightness values measured by the incident meter

gotcha...so for portraiture, it might be worth opening up so ppl can sparkle like a vampire in the sun

oh okay his method is pretty simple. take yer GF1, change your metering setting to the one w/ the smallest zone (the one that looks like [ . ]). measure yer palm outside in the sunlight, that'll give you a reading that's +1 stop above middle grey (or IOW it'll underexpose by +1 stop if you set that as middle grey.) you have to add one stop to it to get the middle grey reading for the scene.

use yer iphone in reflected mode (I'm assuming using the back camera and not the front) to take the same measurement of yer palm in the same lighting conditions. adjust the "EV" setting in the app until it matches with the adjusted reading you got from your GF1 in the first step.

yeah or you can just exposure bracket since yer shooting w/ digital, or pump it up in LR afterwards xp

xp right.

and i've got the tape over the back camera, so it's getting an averaged reading

oh okay. well you're using the iphone camera to take the calibrating reading w/o tape first, right?

oh wait am i supposed to do it that way?

well yeah - the calibrating reading should be taken in reflected (i.e. default), since that was what mode your GF1 used. after you calibrate the reflected mode, then you can tape it and use it as an incident reader. I'm pretty sure...

actually the more I think about his method the more I think I may be misunderstanding it. will ponder, busy now

no wait - I think you are supposed to point the tape covered iphone @ the sun. my bad.

so yeah just do what he says in the article. *shrug emoticon*

so guys are there like noted bibles of film photography basics? i've discovered a roll of delta 400 in my drawer (prolly...3 years old?) so maybe it blows but i've got this rollei sl35 sitting here so i may as well shoot it. the battery is dead, so the in camera meter doesn't work, so i'm gonna use that iphone trick and see if it actually works with film.

old 400 delta should be fine; my not-especially-technical approach if i think i'm going to get something washed out is to just push it up anyway, because then you'll get *some* kind of thing going on, rather than the just-washed-out-400-style-pics you fear.

yeah maybe expose at 200 since film sometimes loses sensitivity once it's expired.

my suggestion would be to keep a logbook of your exposure info. maybe bracket to see how changes in exposure affect the negative. other than that, just shoot!

http://www.amazon.com/Photography-10th-Barbara-London/dp/0205711499

This was the standard 'textbook' in my school's photo program - buy an old edition cheap, something like the sixth or seventh edition should be all film (though this covers way, way more than film/darkroom work)

http://www.amazon.com/Photography-7th-Barbara-London/dp/0130282715/ref=sr_1_1?s=books&ie=UTF8&qid=1316896141&sr=1-1

dunno if this is useful to you at all, but I referred to it quite a bit when I was developing my own film:
http://www.amazon.com/Film-Developing-Cookbook-Darkroom-Vol/dp/0240802772/ref=sr_1_5?ie=UTF8&qid=1316896204&sr=8-5

thx for the recs man

so I've been dicking around in my yard and testing my gf1 sensor against the sunny 16 rule and it's kind of amazing how well the two match up? also went through a few old photos and sure enough, if I (laboriously) spitball the exposure i can get pretty close (one stop) of the recorded meter value. p cool