Calibrating Your System for Accurate Results
All right, well, welcome welcome to creative life and where this is the the complete course in skin uh but actually it's a course in digital photography so, um we're going toe really see a lot about workflow and lighting and then post production which are all aspects of photography not just skin, so let me get my little clicker here so we're goingto start with sort of overview of of image capture workflow but before you do anything else as a professional photographer, you really need to, um, calibrate your system, get really get intimate knowledge of how your camera and your lands in your lighting perform. So uh, that means we have to shoot a test and the test is about the camera lens and lighting um and what I'm I put a bit of this discussion in my book I have another book called mastering exposure. I went into this in a bit more detail and where we we can't go into too much detail here because we have limited amount of time, but I want to talk about how to shoot a test which is essen...
tially shooting in a range of exposures to kind of determine what the dynamic range of your camera is, um and calibrating color I like the little x right color passport and we'll talk about that is very simple and how to save your settings in light room or adobe camera I I'm a big proponent of light room on that that's my sort of of the the hub of my work flow everything goes in and out of light room um and once you you've shot this test and you've determined your settings you can save a new camera default um so that you know whenever you open up the files from your camera a sort of standard setting can be applied all right? So shooting the test um always shoot a test with a color checker target and this is this is my testing set up um the twenty four patch it's now ex right color checker known as a man's l chart has been standard in the industry for many, many years um and a human subject thes these two here I'm gonna point to them over here these two patches on the color checker are supposed to represent dark and light skin but I've never met a person yet that actually had skin that was that color. So um it's it is I think important tio to get a real true picture of how your camera captures color to have a real person in the test shot and if you're stuck for a model you can use the self timer like I did here I set the camera on self timer and run into the picture myself um I have this little black trap uh which is a box with a hole cut in it inside the box is a black velvet so it's sort of an absolute reference for absolute black um not absolutely necessary but it's a nice thing to be able to take a reading of what absolute black actually is in iraq capture and I have something similar here this is a diffused white reference eyes a styrofoam on lenz packing uh, sort of material which you can't really get any more uh they now use a paper to pack the lenses but it happens to be a very reflective surface so it will always clipped a white and it's nice toe in your in your exposure range to be able to identify when the exposure clips this lightest possible thing starts to clipped a light um and then this part here I write down the number that would represent the I s o for that exposure. So when we shoot a range of exposures uh sometimes now very often actually theis so as recommended by the manufacturer the camera is not the actual best ice so for that camera at any under any given lighting conditions so you have to shoot a test to find out and, uh, the tests um you wantto evaluate the test not on the lcd of the camera so you don't want to rely on that to determine your exposure um everyone I know everybody looks at the back of the camera but the lcd really is not accurate it's it's it sort of represents the in camera j peg and we should be shooting raw because we want better than j peg quality and this the lcd can't show you exactly what what you're really getting in the in the raw file and often the hist a gram on most cameras you get a cut it's kind of a composite hissed a gram you don't actually see red green and blue separatist a grams you can on some of the cannons offer that feature but it's often inconvenient to switch to it. Ah, and many cameras just simply show you the composite history ram so it's not really accurate because you can't tell if you're clipping in the red channel for instance because a composite will shift the representation you may think you're ok and you either have clipping or you can actually over expose more than that history is indicating, uh I am a big proponent of using a light meter I have I like the sick connick meters and I'm going to be talking about my meter here is the is the l seven fifty eight p r it's the digital meter uh going toe come back to this in a minute but any kind of handheld meter is is ah preferred for shooting the test and I actually ideally it's it's good to use a reflective meter in this iconic can do both so you have the little white dome which is your incident meter and you can switch it so that you can actually take a spot meter reading um and I use that to take a meter reading off of the the test target to get the middle gray value ok so uh we're not going to actually do this live here but the idea is to shoot your test subject and brackett exposure so by one third in my case shooting can and shooting one third f stops and I shoot multiple exposures getting brighter and getting darker um I recommend that on the lighter side you uh exposed at least four f stops overexposed and then on the darker side at least five if you have a nikon camera you might need to go a little more um the interesting thing is that from middle gray the the camera actually isn't really linear in the sense that middle gray is actually a middle value for the ast faras the cameras concern you have more latitude on the shadow side before you end up losing that you know by under exposing at some point you're going to drive that middle gray patch two black and it will you know, sort of fall off the cliff as it were um that tends to be between four and five s stops on the on the shadow side on the highlight side it's more like maybe three f stops you know, so you have much less room to overexpose before you clip on the highlight side and more room in the shadow side um so you want to run your brackets to make sure that you cover that range um and then what you're going to do is evaluate the exposure is based on the mid gray patch, so the mid grade patch is this patch here on the color checker target that's supposed to be fifty percent luminosity um and this is you're sort of white patch and you're black patch these aren't actually maximum white or black um but uh for the target they are but this is what you basically exposure on then as we under expose that this patrick get darker and darker and darker when it reaches the point where it's indistinguishable from this patch which will already be black uh, that's your your your, uh, sort of your clipping point you don't want to be underexposed past that when you're over exposing it's going to get brighter and brighter and brighter and it right before it turns white so that it will match the white of this patch that's your end point on the highlight side, so again don't look at the lcd this's not going get tell you, which is the correct or the best exposure you're going to read, you're going to measure the mid gray patch in light room and, um base exposes off the off the meter reading and then evaluate which one is the actual best exposure and you may I what I my methodologies always start on what I expect the best exposure to be and then open up and we'll come back to it this would be my expected best exposure and then I'm stopping down um if I if at all possible trying news shutter speed to change the exposure um the f stop eyes not it's not quite as accurate actually most of the time for most with most lenses um and shutter speeds air timed electronically nowadays, they're not actually timed by the curtain that, uh, night can still have a shutter curtain, but the actual timing of the exposure is elektronik, so the shutter curtain just their toe make them to make the chip visible to the light, but the the duration is has controlled more tightly by the electronic timing off of the signal off the chip so you can actually get more accurate exposure differences by using the shutter um and then you're you're going toe look at that midgrade patch in light room and find the value that's closest to fifty percent light room gives you percentage numbers if you have noticed in the the rgb number read underneath the hist a gram because your percentage numbers so you want you'll find exposure the best exposure will be the one that's closest to fifty percent and you may find that your camera deviates from what the manufacturer says is it's true? I so, um so again here to review uh you're determining the best exposure is the one that's closest to fifty percent after that, you're going to look for those end points as I was talking about before you, right before the mid grade patch turns white that's your end point for your dynamic range and you'll count from your ideal mid tone exposure you'll count f stops to the exposure that gets that patch just toe just before it goes toe white and that's your dynamic range on the same thing with on the on the shadow side, right before that midgrade patch turns black that's your the low point of your dynamic range so you're plus and minus so you you count stops over and then you added together and that gives you the number total number of stop dynamic range, really dynamic range um and not just what the manufacturer says eleven stops or whatever they mean I've never found a camera that actually lived up to that um in real world uh actual total value, so this is I made this little table here represents my cannon five d mark to every camera is slightly different so you have to actually shoot this test tio determine it so what I what I've done is I shot the same exposure bracket test in different lighting conditions shade son tungsten and flash so those are all the lighting if you on ly use one condition you know like if you only shoot with tungsten lights indoors and we only need to test for that but if you shoot in all these different conditions that's a good idea to test all of them and what's interesting um you see here the exposure compensation column here so this it means minus one third means that the cameras actually mohr sensitive than the s o indicates so it was more like one twenty five instead of ice. So one hundred I set the eye so here in a shade and son both one hundred also in flash in tungsten I said it to six forty because I almost never have enough light in tungsten lighting situation. So I'm always shooting with aya so at least it's six forty I'd like to do these tests that the lowest I so that I'm likely to use because that's generally the highest quality so so here shade you see my exposure compensation was minus one third is actually more sensitive to light so the also would be more like one twenty five so I have to dial in a minus one third exposure compensation in the camera when I'm shooting and shade uh with sun it's actually, uh I have to actually add one third of exposure to overexpose slightly it's kind of strange not every camera behaves this way, but very often cameras will not behave linear lee in different colored light you can kind of see a pattern here uh tungsten the warmest light I'm having to add two thirds of an f stop compensation to get a correct exposure two thirds of a stop I mean that's like that's that's fairly big eso if you're not doing that with this camera I would be under exposed all the time by two thirds of a stop um flash works out to be no compensation it's exactly I so one hundred so it's interesting that the cooler light because the chip seems to be more sensitive to cooler light and shade I'm actually stopping down to get the right exposure as the light warms up I have to add exposure until I get to tungsten, which is the warmest light I'm adding the more the most exposure that just seems to be the way this particular chip has been responding so it's not linear across different colored lights very interesting um and the only way you're going to find out about your cameras to shoot that test yeah, chart. You're making a reference to the zone system for his own ones on nine? Yeah, remind me it's been a little while, but aren't zone one end zone to having a rich black with no detail? Yes, but I'm, uh, with digital it's kind of interesting because you have so much more latitude. Um, in terms of what you can do after the fact, right, I've found that if you have, uh, a tone in zone one, which is just just slightly, you know, lighter than black, uh, you can actually pull detail out of that in your post processing and same thing for oddly and digital that it behaves kind of more linearly on the highlight side, where as it gets brighter and brighter and brighter until finally, it'll clip toe white, but it's a straight line, right? So instead of having that like way would think of ah, the, um the tone curve, a cz being a curve, so as the on film as you increase exposure, kind of rounds off the top, and the demarcations from one's own to the other are much narrower, with digital it's more linear in the sense that each step each zone is approximately the same distance away from its neighbor, so we can go all the way to zone nine and there's actually still enough gray there that I can pull a sense of texture out of it if I need to and also with the dynamic range of these cameras um the high bit depth capture the raw file when we do our kind of recovery operation to recover detail that looks like it's missing in the in the preview suddenly we confined lots of tonal information there, so so yes, I'm using the zone system it kind of expanded for digital, so going from just for clarification so a part of what I'm seeing here is based on what you're getting by looking at the image and then understanding that you can recover those deals I'm actually measuring his own uh, you know, zone one in his own nine exposure on percentages in light okay, so I can see what the percentage numbers are and so in percentage numbers, if you're ninety percent, that would be a zone nine and if you're a ten percent that's his own one, okay, so you can see how many f stops you have to be to get to that zone. Nine so in this case on my camera was two from them from the mid tone from zones on five in two and a third stops I'm reaching so nine so it's it's not linear on either side of that's interesting of the mid tone yes so, lee just for some folks including mary costa rica and wearing costa rica to ask this but who are not as familiar with his own system could you just do a really brief not going to in depth? But what is his own system and maybe where people can learn more about okay, um the zone system was developed by ansel adams on dh sort of the made it famous as a way of controlling, uh, exposure and, um, uh rendering tonal rendering of the image and it was it's a very rigorous system of, uh, visualising tonal range is in terms of these nine zones, so he split from black to white. He split the image down into ten steps. So, um in digital I use eleven steps because the fifth, the midpoint, the fifty percent ends up right in the middle that way. But anyway, the zone system is just a way of kind of like dividing the tones up and being able to identify, say, ok, that's his own one, this is a two ah, and you'd have a kind of this reference of the zonal reference which you could apply to any image that you're working with and he built ah adams developed a whole strategy for exposing to the zones and and he was dealing with wet chemistry and you know the time that a film spending the developer and how you could push from his own five get it to go toe you know plus one zone and all these different things would affect the contrast in total rendering now in the world of digital that is so easy to do we could push zones around willy nilly so the digital zone system is actually kind of a simplification because all I'm really concerned with is making sure that I've captured information between zone one in zone nine then the zones between those can be pushed around and manipulated with curves and photoshopped like, you know, unbelievable amount yeah would it be fair to say just generally speaking that ansel adams was essentially developing his own version of a history graham to be able to play with by developing the zone system with zone one being the blacks and zone nine being the highlight detail in the sort of, you know, kind of I mean it's actually much more valuable than hissed a gram hissed a gram despite its being, you know, promoted so heavily, it doesn't really tell you that much it's a statistical distribution of values in an image but, you know, that's gonna look radically different if you're shooting a snow scene versus, you know, a black cat in a coal mine kind of thing, you know, so a good history graham could look horrible depending on what image it's applying you know, uh it's applied to he certainly wouldn't want hissed a gram that looked like a normal spread if you're seen was the black clad black cat in a coma so knowing on the other hand that you know in where your zone nine is in that scene you know, so you're not under exposing it gives you a lot more information and so when we shoot this test and you determine like your dynamic range, you know, here uh I'm just adding the two numbers how many stops below zone five and how many stops above that's the total I have seven stops uh and you know then where your limits are for your camera so using ah spot meter is really ideal for this sort of zonal thinking because you can take your spot meter and put it on any tone in in the scene. So if you have a white t shirt, you could put the spot meter on and go and no ok, am I going to be overexposing that white t shirt or white wedding dress saying, uh, if you know how many have stops, can I open up to make sure that that stays within that that limit and doesn't fall past zone nine and consequently also for the shadow like the black tuxedo the you know, the spot meter you can actually measure that it's a little less convenient to use the meter in the camera for that sort of thing you'd have to actually walk way up make make sure you were putting a shadow on the subject and try to fill the frame with with white or black to get you know those those limits and it you know you could apply to anything any scene that you're you're photographing um in general though, you don't have to get all that crazy u here mostly going to be basing your exposure off of ah mid tone placement and so averaging the values in the camera if you're using the in camera meter um you you can kind of get a feel for it, but this is the first step is really valuable information to kind of know what your cameras doing, you know, so that you know, if you're inside under tungsten light is kind of orange warm colored light, you're probably going to be having to add exposure beyond what your meter seems to be telling it so you're going to cheat you're your exposure to the plus side? Yeah have you seen a lot of variation from one mark to another mark too? I mean, is that a thought behind? Well, I I don't have to mark two, so I haven't seen it in my testing, but between the five day mark two and a one d s there there was a difference and from one camera to the next you know there's a difference it's not going to be huge maybe a third of the stop you know different between the two cameras of the same make but it you know you you won't know unless you test your own camera exactly where you happen to fall you might be lucky and you fall right in the middle of the you know the manufacturing tolerance and then your golden right then it's going to behave just exactly the way they expected to be a but very often doesn't work out that way uh if if all of that testing is like too much for you because it's shooting a lot of exposures and getting in there and analyzing it um sick connick has kind of an interesting system that they use for their they've developed this for their digital meters so this is the l seven fifty eight dvr the digital master meter and it has a it has ah usb port on it so you can actually hook the meter up to a computer and they have software that what you do is you shoot this target if the by this target from them and it's it's not cheap I forget what it is but you buy this target it's sort of like getting a color checker a big color sugar target but it's just it has these gray patches on it and you take three exposures, so you take a normal exposure uh a plus three and a minus three stops so you only have to take three exposures and then you feed these images, process him in light on the way to normal the way wood and feed it to the meter software and then the software analyzes all the patches and figures out your dynamic range tells you where you're in points are and puts that into the meter itself as a profile it also calibrates the meter so you're you take a media re reading with the spot meter you know you go up a measure the mid gray patch here measure that and then you also taken incident reading in the same light very often they don't agree they may be off by usually like a third or two thirds or something like that um and the process of running these images through the software builds a profile that calibrates the meters so that both the incident and the spot will agree in terms of how they record the exposure. So that's a good thing is that something they have to do on a regular basis and I don't have to do it once really for the meter, you probably, uh you khun store three profiles in the meter and you can switch between which profile you're reading uh it's not like a color management profile it's just a profile for the meter exposure ring system s o what I recommend is to shoot the test in three different kinds of light so that you could be switching because apparently the digital cameras don't responded different colored light the same way so the lights cooler they're a little more sensitive you wantto put that sort of calibration into your meter if you're going to be using the meter, you know all the time, then you can calibrate it for those different colored lights and then I've used take a meter reading and set your camera whatever the meter tells you and you don't have to think about oh, I've got to do plus three, you know? Or you know, a bus two thirds or a plus one third here minus one third unit just you just said your camera or whatever the meter says, um so kind of cool question really from a number of people on the internet does this change by lens that you're using, uh and so that's a good question, if you are, say, a cannon shooter and using all cannon glass, the difference between one lens and the others is so minimal that it's almost not worth, you know, bothering with, uh, I would say unless it's a kind of an exotic lens, you know, like you were using a fish eye or something like that but we use those types of lenses so infrequently that uh it's probably not worth bothering with now if you are using cannon glass and say takina or a sigma or some other manufacturers glass there will be a difference between how those two uh lenses perform in terms of both exposure and kind of contrast and s o if it's important to you then you may need to test um the other lands yeah good question. All right, move on. Okay, so uh the other thing that's that is interesting about a meter and really like here using flash you know, trying to determine flash exposure you have to use you have to use a meter um you can use the meter to measure of contrast ratios and that's I'm showing here the the dome retracted so in this configuration I would pointed it at the light so I would say if I have a back light I would point the meter at the light take a reading then point the meter at the front light take a reading and I can then measure like the contrast ratio between those two lights and uh cinema of dps director photography's use this approach all the time so they're always like measuring the lights and figure out contrast ratios we don't see it used it that way is often with stills but it is something that you know it's a good thing to know how to dio um the other thing that the meter does in in this system so here I show you the camera this little that's how you would change from one profile that other they call it camera one, two, three um the other thing you can do when you they have this little mid tone button and when you push the midtown button, the next exposure you're reading you take will be considered the midtown and then the next to exposure meter readings you take will be considered ah highlighting a shadow so you better be reading something that's appropriate there and they'll all three be displayed on the meter in a graphic display which you're going to play with this little jog wheel so let me let me show you how that works. So for instance, here this is the display at the bottom the zero is considered the mid tone and then the sort of spread of points on either side of that. So I took three readings. So my first reading after pushing that midtown button uh gets put here by default and I measured the highlight it end up there and I measured the shadow and it ended up there and I'm looking at this going these areas down here these air might clipping point see the two triangles, two triangles over here those that represents my limits, my personal limits based on my testing and so I don't wantto I don't want to put thie low exposure that close to the end if I can help it and I've got more room in the highlight side so I can use the jog wheel to just move that whole spread over here just push the jog wheel and then I can move these points too put them on the higher end place that my hi my highlight reading closer toe the white clipping point so now I've got better uh, better data to work with, and I can see that it still fits within my limit so my highlight, melatonin and shadow fall well within my range and I whenever possible, I'm trying to keep it on the upper end of that range, and this is measured from their scene it's not like it, just a guess looking at the history graham of the camera yeah, so you're keeping it pushed more towards the highlights because you know that you're going to be able to recover more from the shadows is that I can always make it darker and you know, the more you have, um you haven't well lit or well exposed, the less noise you're gonna have, so if my shadow and I can put the you know I can manipulate the zonal placement my shadow should be black right so if my shadow is not even his own nine but his own eight there would be less noise in that area and I can force it to go to black very easily so that's why I wanted tend to want to be overexposing just a little bit but at least this way I know that I'm not pushing it off the cliff that highlight values falls within the range and ike look on the meter and see it really quickly so if you get good at this with this meter it you know you can be confident in kind of difficult situations like in a wedding you know you can sometimes get crazy contrast in light of course the bride is standing in the white dress right as the ray of light is hitting it and you know the groom in the black tuxedos in the shadow and your leg you know you could take a meter reading and no, just how far you can overexpose that brightly with lit dress before you run out of room okay uh now we've all been just talking about exposure calibration right? Has nothing do with color color actually turns out to be a much easier problem as faras calibration goes because we just take a picture of this color passport this x right system has this little kind of booklet thing and you just get a good exposure of it which you've now already know howto make a good exposure right and that's all you need to do and then you run it through their software so I take it I shoot it in both tungsten this is tungsten light and daylight great balance so I'll do the white balance to this patch and um and then you export it in light room the color passport software uh runs as an export plug in so you go to export it and you get this x right pre set and it sort of walks you through it you select those two um shots the one done in tungsten light one done and take daylight and you can build a dual illuminate profile which is designed to work with light room or adobe camera ross system of color of their whole the way that renders color uh and it makes a profile that you you know for your camera so you just hit export but it makes the profile puts it in the right place, you know, in the preferences and you're good to go so then when you actually set up your your camera defaults you can choose this profile for that camera save it is your camera default and then every time you develop something from that particular camera you you've already got that profiled so you get the best possible color you can um when we're actually shooting after you've done all this calibration, and the only thing left, really, when you shoot is to try and get a good white balance. So the ideal thing is to use some sort of great card. I have a couple of great cards with me, and we'll show that when we're actually doing the shoot, um, this I didn't bring this one with me. Just because it's, you didn't want to carry that much stuff on the airplane. But this was kind of nice because it pops open and the other side's a white, uh, reflector so you can use it as a reflector or, uh, like a fifty percent gray card. So this also works for when you're far away from the subject and have them hold that you can take a meter reading off this larger gray area of the easier to fill the frame of your dslr with a big thing like this than it would with a tiny little, you know, great card.