alundeb wrote:
Suddenly I can see the light. This is about semantics.
The term "Expose To The Right" is totally misleading. What I mean, when I say "Expose To The Right", is "Expose From The Right". And I am sure I am not alone. We tie up the brightest part of the image (The right hand end of the histogram) at the clipping level. The direction of where you move things may either be To The Left or To The Right, but we start at the right hand side, hence "EFTR".
I am sure you masters of the English language can refine my idea about a renewal of the ETTR to EFTR. ...Show more →
Huzzah! Me too! I guess we're all more or less on the same page and its more of a language issue. I do agree with you though in that Peter's image is a good example of ETTR.
There are exceptions to every rule - probably even to the rule that states "there are exceptions to every rule..." ;-)
However, I really do think that the "expose to the right" concept is more often than not a good place to start, exceptions aside. (Edit: "expose from the right" works, too... )
Dan
PetKal wrote:
I think Anders, Dan and others are saying much the same thing yet looked upon from different angles.
When doing real photography, "expose to the right (or left, or center)" should not be a governing principle of the process. There is a host of other issues which take precedence in the creation of a photograph. However, one should perhaps be cognizant of the different noise and tonal impact which over- or under-exposure has on the final product.
In applying such awareness, we do it differently because we all have different objectives and aesthetic criteria.
Let me remind you of the exquisite "Egrets" image made and shown on this board by Hrow (Henry). That one is just about the most extremely "exposed to the left" image you have probably seen, yet it works beautifully because it was driven by Henry's vision, and not by shadow noise considerations and some such pixel peeping minutiae. ...Show more →
cgardner wrote:
Noise is a given with any scene that exceeds the sensor range. Lower ISO is simply one strategy to minimize it. The better strategy is to eliminate it by finding ways to fit as much of the scene range onto the sensor as possible. That is possible indoors or outdoors in the foreground if using two flashes in an overlapping key-over-fill configuration.
In situations in which you have to shoot in poor light and are shooting subjects where the flash can produce a result that isn't as awful as the poor light might allow, using a flash in the daytime is certainly a viable option. It works well in several of your shots.
Of course, this approach has its (very serious) limits when you look at photography more broadly, considering other subjects and other approaches such as, when possible, choosing to do the photography in better light...
Photon wrote:
Using an "optimal" ISO, such as 100 on a 5D2, is a bit like shooting at, say, f/6.3 with a particular lens because careful testing has determined that at that aperture you get the highest average resolution across the entire frame, and aberrations are minimized. It doesn't hurt to be aware of such things, but in the real world with good lenses and good (i.e., recent) DSLRs, decisions about settings would usually be based primarily on conditions. Is the subject moving? How fast? What is the light like? How much of the scene do we want to be in focus? Are the shadow areas important to the composition? Is highlight detail an important part of the texture that we want to show? I will certainly admit that when I'm shooting a landscape with a tripod, ISO 100 is my starting point, but going nuts deciding whether to set ISO 160 or 200, 200 or 50, etc. is not something I want to bother with. In fact, I have my bodies set to only allow "full stop" ISOs, just for simplicity and speed. (Apertures and shutter speeds I do by third stops. Works for me.) Sometimes for action, I use Manual exposure and auto ISO on the 1D4, and it then sets intermediate ISOs when needed. I never find noise differences jumping out at me. YMMV...Show more →
Getting back on topic for the OP, did we come to agreement that there is a "native" or "base" ISO for every Canon camera? If so, is it typically ISO 100?
RE: "So, under favourable conditions, i.e. f-stop and aperture can be widely adjusted, movement not a factor, etc. etc. is there a "BEST" ISO to use for a particular camera? Or just use the lowest ISO setting for optimal results?"
Given "favorable conditions," did we come to agreement that the lowest ISO (e.g. 100) is the optimum sensitivity setting to provide the best IQ? (e.g. greatest dynamic range or lowest noise?)
Note: Aesthetics or artistic considerations are presumed to be non-sequiters to the OP's question; he seems to be mostly concerned with technical considerations only. This is not to say we can't include these considerations in the discussion, but they seem to deviate somewhat from the OP's original questions.
OP--given what you've read in this thread, do you have any related or follow-up questions?
Very amusing. He says ETTR is a bunch of bull. Then the method he recommends is exactly ETTR done the right way. Place the histogram as far to the right as you can without clipping the highlights. He doesn't either understand that this is perfectly in line with underexposing sometimes.
alundeb wrote:
Very amusing. He says ETTR is a bunch of bull. Then the method he recommends is exactly ETTR done the right way. Place the histogram as far to the right as you can without clipping the highlights. He doesn't either understand that this is perfectly in line with underexposing sometimes.
Yep, I don't really understand the point of the article. It's already known that ETTR isn't possible in all shooting situations, like high DR scenes having no room for shifting the exposure. And his comment about newer sensors not needing/benefiting as much from ETTR is only true for shadows and even then only for the read-noise aspect. There's still plenty of shot noise across most of the tonal range and even more so in the lesser-exposed color channels and this holds true for any sensor available today, including the D3s. Whether the noise is intrusive or would show up at a given print size is certainly open to debate but considering how easy it is to ETTR once you have a workflow in place I think even that point is moot. Why not get the best technical exposure possible and leave all your post-processing and print-size options open whenever you need them?
This thread is one of the best that I have read in a long, long time.
Thank you.
I have found it very informative too...
I believe a tenet and key takeaway is to always shoot RAW and expose for your subject matter in a manner which provides the greatest flexibility for later manipulation in post processing.
Whether that means ETTR or ETTL (or neither) is dependent on the shooting conditions, the equipment being used, the needs of the photographer, etc.
Getting back on topic for the OP, did we come to agreement that there is a "native" or "base" ISO for every Canon camera? If so, is it typically ISO 100?
RE: "So, under favourable conditions, i.e. f-stop and aperture can be widely adjusted, movement not a factor, etc. etc. is there a "BEST" ISO to use for a particular camera? Or just use the lowest ISO setting for optimal results?"
Given "favorable conditions," did we come to agreement that the lowest ISO (e.g. 100) is the optimum sensitivity setting to provide the best IQ? (e.g. greatest dynamic range or lowest noise?)
Note: Aesthetics or artistic considerations are presumed to be non-sequiters to the OP's question; he seems to be mostly concerned with technical considerations only. This is not to say we can't include these considerations in the discussion, but they seem to deviate somewhat from the OP's original questions.
OP--given what you've read in this thread, do you have any related or follow-up questions?...Show more →
100 ISO is native on all Canon SLRs that I know of (give or take a little) but not the 5D2.....it was pushed closer to 200
cameron12x wrote:
I believe a tenet and key takeaway is to always shoot RAW and expose for your subject matter in a manner which provides the greatest flexibility for later manipulation in post processing.
Whether that means ETTR or ETTL (or neither) is dependent on the shooting conditions, the equipment being used, the needs of the photographer, etc.
I agree. Many will say that they prefer to shoot jpg and accomplish everything in the camera. This can work very well for some, and certainly is preferable when your images will be immediately sent out for publication (breaking news, sports, etc.), but sometimes an excellent image can be truly improved in the computer, just as a great negative could be massaged in the darkroom. There is a craft to shooting a great color slide that wows viewers when it's projected on a screen, but some scenes can barely be captured by the limited dynamic range of color positive film, and some perfectly exposed slides only made great prints when masking, dodging, and other techniques were used. Easier now when they're scanned...but I digress...
Anyway, if we agree that sometimes manipulation is called for, then we might as well prepare for it as well as possible.
There is "technically correct" exposure, which occurs when a scene fits the sensor exactly and it is reproduced with detail from lightest to darkest content with texture (i.e., Zone 1 to Zone 9) and there is "perceptually correct" exposure, which can occur when exposure isn't technically correct but the viewer accepts it as real.
Back in the days of shooting B&W we didn't have ETTR vs. ETTL debates because it was a given you could fit any scene range to a print and reproduce the entire tonal range, either by changing paper grades to fit scene contrast on the negative, or with the Zone System change development of the negative based on the scene range to fit one grade of print paper. The workflow with the zone system was: 1) read the Zone 1 shadow with a spot meter to determine exposure; 2) read the the Zone 9 solid white to determine the scene range and development time needed. All negatives, regardless of scene range wound up with exactly the same shadow>highlight density range. Every scene fit the #2 print paper with a full range. The up-front anal retentive metering and systematic film development made the print making step a no-brainer.
Color negative film changed the paradigm of exposure. It had half the range of B&W and the range couldn't be altered with development because the color balance would go wonky. That's when photographers first faced the dilemma of whether to clip the highlights, shadows or both on a sunny day. Whether you exposed the faces correctly and blew out the sky, or rendered the sky correctly and rendered the faces underexposed depended on which was more important in the photo — exposure based on what is perceptually correct based on viewer expections. 60 years later we are in the same place with digital because the DR is about the same as a color print — about 6 stops of detail.
Camera manufacturers could make exposure easier to understand if they put three different colored indicators in the playback:
1) Red highlight clipping warning - what we already have
2) Green shadow clipping warning
3) Yellow indicator all areas reflecting 12% — the perceptual middle gray in the middle of the sensor's range
That of course will never happen because it would reveal how poorly the sensor records the range of an outdoor scene.
Lacking the other two warnings the way to see if scene exceeds sensor is:
1) Raise exposure until highlights start to clip in the playback then lower it 1/3 stop.
2) Look at the left side if the histogram. If it piled up on the left scene exceeds sensor.
To find out how much the scene exceeds sensor range increase exposure until the histogram is no longer running off the right, counting the dial clicks (one click = 1/3 stop).
To find out the range of your sensor in stops set Custom WB on a gray card, will the viewfinder with it and bracket exposure from the point it is1/3 stop below clipping until you render it as dark as possible... Here the test from my 20D..
So I know my sensor range is about 6 stops with detail. If I expose the highlights accurately and the left side of the histogram is showing loss of shadow detail and it takes + 3 stops exposure to render the shadows accurately, know the scene range is 9 stops.
Unless there is are large areas of important white highlight detail the exposure which looks "normal" perceptually will usually be somewhere in the middle of that range: exposing highlights 1-1/2 stops above the point they start to clip.
That's about where Canon metering pegs exposure at EC=0. By shooting in Av mode in backlight and adjusting until highlights are below clipping I find that I predictably need to dial in EC = -2 stops. That means the metering default would choose to blow the backlit highlights by 2 stops to render the shaded content in the center of the viewfinder accurately.
The "Cannon" approach to Av exposure is like sighting in a new gun. Take a shot of new scene at the camera baseline of EC=0 (what it thinks is the correct aim point), see where that lands (what if anything is clipping), then adjust aim with EC from that baseline to put exposure where you need it for that scene. If nothing is clipping and there is also a gap on the right of the histogram is tells you the camera's "best guess" is underexposed. If there is clipping at EC= 0 and the left side of the histogram is piled up it tells you the scene exceeds the sensor. Then you decide to clip more or less of the scene based on seeing what scene content is clipping.
The same thing can be done in M mode from a Sunny 16 starting baseline.
I started using that approach back in 2004 shortly after buying my first Canon, a 20D, to try to figure out how Canon metering modes interpreted scenes, both ambient and flash. By always starting from the baseline of a EC=0 and feC=0 test shot and keeping track of how much adjustment EC and FEC was needed for front-lit, back-lit, side-lit, etc. scenes I got a better understanding of when the metering defaults guess well and when they didn't.
After testing all 19 metering combinations on my 20D I concluded that the quickest route to the best "perceptual" exposure was the Cannon approach: let the camera take it's best guess on its own (i.e. EC =0 and FEC =0) first, evaluate, and adjust from that baseline until it looks right. I found that all things considered evaluative metering made the best guess. The test shots are here: http://photo.nova.org/Canon/TTL/
Basically, my approach starts with ETTR. Most often it is the "right" way, or close to it, on a number of levels. When might I not quite exactly use ETTR?
- If the scene contains a lot of subtle and important near white variations, it is my belief that I can get a better starting point if I expose a bit below ETTR. (There are other ways to enhance or recover very bright details in post, if you prefer.)
- With scenes that are very hot in one of the color channels (usually red) and cool in others - e.g. certain kinds of flowers, sunsets/sunrises, etc. Here, getting too close to the right - even in the affected color channel - can sometimes over saturate that channel in ways that are not apparent from what the histogram shows. Here I just try to be a bit conservative.
- With scenes in which the highlights are few and small and in which I can afford to let them blow a bit in order to keep better exposure in the rest of the frame - the familiar specular highlights situation.
- When confronted with extremely wide dynamic range scenes, if I want to capture them in single exposures I may need to consider carefully what can be done in post and what I'm willing to do. I may protect the highlights and plan to recover shadow detail in post, or...
- ... I may use exposure blending in order to capture a wider dynamic range, from which I can create the best image in post.
"There is "technically correct" exposure, which occurs when a scene fits the sensor exactly and it is reproduced with detail from lightest to darkest content with texture (i.e., Zone 1 to Zone 9) and there is "perceptually correct" exposure, which can occur when exposure isn't technically correct but the viewer accepts it as real."
I love the way anyone can cut through the BS. The largest file size shot wins. The same scene with the same camera, the smaller file size shot is never going to be the winner. ETTR wins every time with a larger file size as proof.
OP here - thank you Cameron12x, cgardner and all others.
I posed a technical question thinking there is an absolute, empirical answer re ISO generated digital noise. Oh, gimme back the good ol' days of film !!!
Yes, shooting digital means more than just the ISO, but one must deal with each such determinant separately. You have to establish all these things in your head so that in the heat of the moment you can choose the settings you think will give you what you perceive as the best possible image. We're not talking deliberate studio setups or snapping at a mountain, but being able to produce viable, sell-able images taken at events, parades and in other quickly-changing circumstances. Thinking EFTR, etc. is not an option.
Thank Canon for the 3 user custom settings, for example on the 7D and 5DII.
Also, I have the highlight over-blow protection active on all my cameras, hence must use ISO 200.
So then, I'll just keep using use the lowest ISO possible, factoring in desired aperture (i.e. DOF), shutter speed, subject reflectance, etc.
Andrew J wrote:
...The largest file size shot wins. The same scene with the same camera, the smaller file size shot is never going to be the winner. ETTR wins every time with a larger file size as proof.
Often that's true, but keep in mind that increasing the ISO enough to introduce a substantial amount of additional noise will make the file size significantly larger, and in this case, it's generally not what you want. File size can help pick a winner from a series of hand held shots where you want to quickly find the sharpest one. I know both of these situations are different from what you're describing, where I'm sure you were referring to tripod mounted shots, identical in all ways except for exposure modified through shutter speed. In that case, I agree that the larger file size will be the winner.
winman3 wrote:
OP here - thank you Cameron12x, cgardner and all others.
I posed a technical question thinking there is an absolute, empirical answer re ISO generated digital noise. Oh, gimme back the good ol' days of film !!!
Yes, shooting digital means more than just the ISO, but one must deal with each such determinant separately. You have to establish all these things in your head so that in the heat of the moment you can choose the settings you think will give you what you perceive as the best possible image. We're not talking deliberate studio setups or snapping at a mountain, but being able to produce viable, sell-able images taken at events, parades and in other quickly-changing circumstances. Thinking EFTR, etc. is not an option.
Thank Canon for the 3 user custom settings, for example on the 7D and 5DII.
Also, I have the highlight over-blow protection active on all my cameras, hence must use ISO 200.
So then, I'll just keep using use the lowest ISO possible, factoring in desired aperture (i.e. DOF), shutter speed, subject reflectance, etc.
...Show more →
You've got it. I don't happen to like the compromise of that custom function, but there's no reason you shouldn't shoot at ISO 200, and go to higher speeds when the conditions call for it. Use the settings that work for you. You've got experience and a handle on the important things in photography, so don't sweat the technical explanations peculiar to digital sensors and amps. Just make your photos look good!
winman3 wrote:
I posed a technical question thinking there is an absolute, empirical answer re ISO generated digital noise. Oh, gimme back the good ol' days of film !!!
The good old days weren't as halcyon as you remember either.
Back then then the "noise" problem was film grain which increased in proportion with the film speed. You had the choice of buttery smooth 35mm negs with Panatomic-X (ASA 32), the gritty textured look of Tri-X (ASA 400), or you could split the difference with Plus-X (ASA 100). Same with color negative and reversal (transparency) film.
As was the case with B&W noise can be used as part of the aesthetic in the overall look of a photo to give it a rough edged, less polished look.
Indoors it's a balance of the need for speed for ambient exposure vs. how much noise will be acceptable for the content and context. A high ISO noisy shot from a live rock concert will work OK because it will be in context with the action. The same amount of noise in a posed shot for the album cover wouldn't be.
When using flash for candids I'll sometimes crank up my ISO to 800 or 1600 when I want more background ambient-lit detail in the shots, often gelling the flash to match the tungsten background so the two merge seamlessly color-wise.
Outdoors the choice of ISO speed is dictated by the sync limit of the shutter if conventional flash is used as fill. At ISO 100 and 1/250th a "Sunny 16" exposure is f/11. HSS flash eliminates the sync limit allowing "Sunny 16" exposures at ISO 100 @ f/2.8 @ 1/4000th. With HHS there isn't a compelling reason to use an ISO higher than 100. At ISO 200 you'd need to change the shutter to 1/8000th to expose the background the same and not affect flash range. Beyond that you run out of shutter speeds. Changing aperture as ISO was raised would cut the effective range the flash.
