Merciless shadows and blown out skies

[tudor_apmadoc]

<p>I would agree with some of the comments above....</p>

 

<ul>

<li>Shoot raw. It allows you to do a lot more in terms of shadows</li>

<li>You can't recover white areas that are 'blown out'. Shoot at -1/3 or even -2/3 whenever you have bright white areas in the scened - that includes clouds, snow covered mountains, bright white clothes, etc.</li>

<li>You can recover dark areas fairly easily. It can be pretty automatic using a program like DxO, or fairly manual using PS - I've been using DxO for about 3.5 years now and have been extremely pleased with the results.</li>

</ul>

[hjoseph7]

Thanks you all I think I'll start shooting RAW from now on unless I'm sure I can nail the shot.

[indraneel]

<p><a href="../photodb/user?user_id=2398817">D. L.</a> , Nov 10, 2009; 05:30 a.m.<br /> Sorry, I will buy that you perhaps had to bring it up two stops, but not four. Also, no D200 I have seen would handle the lost colour fidelity and noise that well, no matter what the post-processing. Sorry if this post is a bit aggressive, but I just don't buy it.</p>

 

<p>Beautiful portfolio Dawid, but you still miss some things. Here's what happened: I have a bunch of manual focus lenses, so for easy exposure control I use the aperture ring on all (manual and AF, and manual exposure too). It was around 9 in the morning, Dallas downtown with tall buildings all around, sun shining through between them. We were shooting on a covered porch with graffiti wall and a solid wooden fence around it. I had metered the scene, set up a reflector and took a few shots. Then moved in closer, changed my lens and took a dozen shots before looking at the LCD. Realised my mistake, changed the aperture, moved the reflector around, and continued.<br>

And here's the processing: Pulled up 2 stops in viewNX just to be able to rate the image. Put the picture to portrait mode with -3 contrast and +1 brightness in captureNX2. Added contrast/saturation/brightness selectively using control points to the image. And then retouched as normal.</p>

<p>Early morning light is directional and quite hard, will always cast a shadow that is very strong. The lack of contrast should have told you how much the exposure had to be changed. +1 for noticing it. -1 for wrong interpretation.</p>

<div></div>

[philosomatographer]

<p>Indraneel, your series of images (as well as your model... oh my...) are beautiful, and very skillfully processed. Thanks for having a look at my amateur work also :-)</p>

<p>I will take your explanation at face value - I am still mightily impressed, and perhaps cannot believe, that a four-stops underexposed digital image from almost any camera, never mind a D200, can clean up <strong><em>that</em></strong> nicely in post-processing. But regardless, in the context of this thread, you do bring the point across - with digital imaging, expose for the highlights and bring up the shadows.</p>

<p>It was interesting for me when I switched to B&W film to have to apply the opposite, i.e. exposing for the shadows, letting the practically-limitless "shoulder" of the film take care of capturing the highlights. That is exactly what I did in my first image (bridges) that I posted earlier in this thread - I over-exposed the image by at least two stops, and in printing brought back down the highlights, in fact I specifically went for an HDR-like look of surreal compressed tones. It's a lot easier with a digital image though, assuming the data is there to begin with.</p>

<p>Anyway, I am still learning this analogue stuff after a year or so (I am focusing on a 100% analogue workflow, no film scanning), its completely different to how I used to process RAW files :-)</p>

[KenPapai]

<p><em>"Large JPEG, P, ISO 320."</em></p>

<p>That combo makes no sense to me. ISO 320.... why? Why on sunny day? You have a thing against ISO 200 or 400?</p>

<p>P -- again, why?</p>

<p>JPG only, WHY? If you want CONTROL shoot Raw.</p>

 

[indraneel]

<p>Thanks Dawid. Since you're taking me at face value, and not all images have exif, here are two more:<br>

<a href="http://indraneel.info/slides/21-scan_0117-1_1.html">http://indraneel.info/slides/21-scan_0117-1_1.html</a><br>

<a href="http://indraneel.info/slides/21-scan_0096-1_1.html">http://indraneel.info/slides/21-scan_0096-1_1.html</a><br>

HP5+ pushed to 6400, intended as a test after I had good results at 1600. Gentle agitation every minute with 1:1 ID-11. Film scanned on Minolta scan dual IV and burning/dodging with capture NX2.</p>

 

[philosomatographer]

<p>Ah, Indraneel, now we're talking :-) See, I expect your D200 images to have at least some form of visible noise, similar to the extreme grain your rather extreme HP5 push has produced. I myself have never tried to push it beyond 1600, I must try it someday.<br>

I wonder if an HP5 push to such an extreme ISO is much better, or worse, than say Delta3200 or TMZ P3200 pushed that far, or perhaps beyond... thanks for showing.</p>

[dan_south]

<p>Harry, whether you're shooting film or using a digital camera there are going to be situations where the recording medium simply cannot capture all of the information between the brightest highlight and the darkest shadow. That's just one of the uncomfortable realities of photography.</p>

<p>In these cases, you have two choices: (1) wait for the lighting to change (or change it yourself), or (2) CHOOSE which part of the scene is most important to you an expose for that.</p>

<p>Evaluative metering and histograms and other automated bells and whistles will never be able to make these decisions for you. You have to take full control of the metering process and exposure decisions, or you have to settle for the camera's best guess. Sorry, but there's no magic shortcut to a perfect exposure in high contrast light. For example, Ansel Adams spent a huge amount of time researching how to handle this type of lighting with black and white film. He wouldn't have dedicated so much time to the task (i.e. the development of The Zone System) if achieving a good exposure was trivial.</p>

 

[indraneel]

<p>There is noise, but then I also always shoot at iso 100. And, I didn't use noise reduction, didn't use shadow protection, darkened all the blacks using higher contrast. That maintained the skin tones and softer colors, didn't reveal noise in shadows, and masked the blacks from showing noise. And multiple controlled edge sharpening steps don't let the eye wander to where the noise is. But one thing I've observed, if the subject and story are powerful enough, the viewer will forgive and forget anything -- even bad lighting.</p>

<p>The film was processed last year when I knew much less, so was less processed. But I kind of like the grain in that, and much better than what NIK filters can generate for digital. I've no idea about other films, only shot HP5, FP4, PanF, ektachrome, and a few stray agfa and fuji chromes.</p>

[indraneel]

<p>I finally know what to suggest to the OP. If the shot had been from the other side of the table (on the left or even from beside the seller), this would have been a very strong image. Actually showing what the lady is doing and the sellers expression better. Even better would be from the left but from higher up. We would possibly not even have noticed the bad lighting. Incidentally, because of the position, it would probably also have eliminated the blank wall and brought the lighting under control.<br>

Maybe it's just me that looks for a purpose, and aesthetics does not appeal to me a whole lot. Probably have to work on that, but I'm not sure.</p>

[garydem]

<p>harry j-you are trying to capture a scene that has too high of a dr for the recording medium. this is going to true no matter if you use film or digital. for your info- ngative has a dr of about 8-9 stops, film slides has a dr of 4+ stops, raw has a dr of 8-9 stops, and jpeg has a dr of about 6 stops. if you shoot sa scene thay has a dr of 10stops or more you are dead with your method no matter what you do. therefore the photographer has to decide what is the subject and shoot the exposure for THAT, or shoot for the highlights and let the rest of the scene fall where it may in exposure. if the latter some effort and success can be made in pping to get some of those shadows back. also, generally shooting for the highlights is what most shooters do. your dslr and its exposure was set for the general scene. this means that you lost it on both the highlights and the shadows. it is up to the shooter to know when the camera is yellking for help in the exposure and to provide that help. in your pics this could have been done with exposure compensation(EC) in a negative direction. this would have preserved the highlifghts. a dslr is not a p&s, the user must provide the decision making that the camera asks for under certain conditions. a p&s will genewrally provide a pretty good pic left to its own, but a dslr will not do this.</p>

<p>consider that you said go back to film. i shot slides for 32yrs and had only 4+stops of dr to use. now with digital, raw or jpeg, there is more.</p>

[eco_foto]

<p>IMHO it looks like your shooting in not great conditions one being mid day light,which even our eyes have a hard time handling, let alone our cameras,don't expect so much in poor conditions!</p>

[mendel_leisk]

<p>I totally agree with a couple of points already made: Harry you should be shooting RAW, and in iffy conditions setting negative 1/3 to 2/3 stop exposure compensation, as good insurance against blown highlights. I find a lot of times my camera's metering results in too bright a scene anyway.</p>

<p>Also, I don't understand why you (and a lot of others) are averse to reviewing your shots, and give the practice a derogatory nickname (chimping). It's an available tool, one of the powerful plusses of the digital capture, why not use it?</p>

<p>Something else to consider with tricky high contrast lighting: get your exposure dialed in <em>manually</em>, with copious "chimping" till you establish a level that minimizes blow-out (not necessarily eliminates, but just a few scattered extremes blowing out), and then leave set thus for all your similar shots.</p>

[randall_pukalo]

<p>Welcome to the Achilles Heel of Digital. Even slide film, notorious for narrow lattitude, has more dynamic range than my Panasonic FX-150 or KM7D.<br>

I regularly take dual shots, in which the slide highlights comes out fine looking (maybe a little light in the highlights, but still "normal" looking), but my digital highlights are blown beyond recovery. It can be frustrating.</p>

[szrimaging]

<p>Ahem, three words for you, external light meter. You will get a much much more accurate reading using an external light meter than any ttl system can provide. The problem with ttl is that it needs to be metering a middle gray, or it is taking the entire scene and averaging it to middle gray. This works, but not as well as figuring out exactly how much light is falling onto a little dome that is constant and calibrated.</p>

<p>That said, as others have said, you would be wise to bracket shots. There are a lot of extra things you can do with bracketed shots in post that are just not possible otherwise.</p>

[maciek_stankiewicz]

<p>Shoot film, seriously :) Who needs digital camera for daylight pictures?</p>

[keith reeder]

<p>That first image would be <em>so easy</em> to fix if shot in RAW and converted with software that has even half-decent highlight recovery functionality...</p>

[colin carron]

<p>I normally have my camwera set at -1/3 or -2/3 stop compensation and check the histogram for most shots. Reshoot where necessary. The main thing is to avoid blown out high lights as you can usually recover shadow detail.</p>

[colin carron]

<p>duplicate post</p>

[matthew_walker4]

<p>I think part of the problem is composition. There's a lot going on in the picture and you are trying to bring everything in - "cram" is a word I would use but it is your picture so it's not fair to say that. Maybe use the available light in a different way to help you rather than use it against you.<br>

I also agree with Mendel. Why is reviewing a photo on the LCD so bad?</p>

[sem_svizec]

<p>Many recent cameras have sensors with considerable DR, but both in-camera JPG rendering and also most raw converters are not intended to deal with harsh light nowadays. They are rather made to provide pleasant rendering with good light, and essentially do not deal with DR compression. DR needs to be compressed because the DR captured by the sensor is too much for viewing and printing. This is generally the second phase of HDR processing; certain raw converters (DxO, Adobe) enable this to a certain extent without having to go HDR (with less tuning capabilities than HDR programs). <br>

Therefore, not only should you shoot raw and expose carefully in such conditions (so that you make the best use of the highlights headroom) - for best lifting of shadows you should make multiple raw conversions at different exposure compensation, and render in a HDR program.<br>

Unfortunately this is a cumbersome procedure, especially because the results are not expected to be stellar, so usually not worth the effort. For documentary purposes it would be much more convenient if cameras could do this automatically (that is, provide good midtones and acceptable shadows while not blowing the higlights, assuming that raw data is not blown). This is not a simple task, but I don't believe it is a totally infeasible one. For example, Nikon's ADL (and similar stuff of Canon and others) could be considered a first step in this direction - unfortunately it amplifies noise more than needed (by wasting a bit of the highlights range, rather than making use of the highlights headroom) and does not really get to more DR.<br>

Whenever feasible, better results are achievable with multiple exposures and true HDR - because the shadows in raw data are recorded sparsely and prone to noise. And having better light in the first place is even better. Due to flare, also the multiple exposures approach is not unlimited regarding DR.</p>

 

[danielleetaylor]

<p><em>Ha! Posts like these make me enjoy shooting B&W film in my Olympus OM-1 even more. For example, this image was shot in the same conditions, 5 stops difference between the shadows and the highlights. No problem, full detail retained in both:</em></p>

<p>Even Velvia 50 can hold a 5 stop range, which is about half of what modern DSLRs can hold. Your B&W film sample does not show a scene with contrast that would challenge either DSLRs or B&W film. Indeed, the image is muddy and lacks any true whites which indicates a scene that did not cover the full range of the film, and a print which used too low of a contrast filter on multigrade paper.</p>

<p>The digital sample you posted was not comparable to the B&W scene at all. In the B&W scene the sun is to the left and low in the sky. It's producing long shadows but there are also plenty of surfaces bouncing light around filling in those shadows. It would not be difficult to capture on just about any film or sensor. As I pointed out above, it actually needs a contrast boost in print. The digital sample, by comparison, was shot from open shade straight into the sunlit sky. I would guess that either the sun was very close to being in the scene, or the sky was just hazy enough to appear white even to human eyes. Portra 160NC would not hold those highlights. You would need a neutral grad filter (won't work in this situation because there's no straight line split between foreground/background), or multiple exposures to be merged later.</p>

<p>As to Harry's photos: considering the detail others pulled out with nothing more than a screen size JPEG to work from, I think the problem is primarily one of post processing, in camera or otherwise. Harry is shooting JPEG, so what are the settings on his camera? And what is he doing, if anything, to the images after they come off the card? Shooting RAW and putting a little work into the images in post would enable him to easily and consistently capture these scenes despite minor metering errors.</p>

<p>Something you must keep in mind when shooting any SLR, digital or film, is that matrix metering is not designed to try and fit the entire brightness range of a scene into the exposure. Rather it is designed to give the subject, as chosen by the camera's AF system, a "normal" exposure. Center weighted meters from the 1970's / early 1980's would fail in the common snapshot situation of a person shot against a much brighter background. Camera manufacturers wanted to overcome this because most people, most of the time, are taking snapshots and want their family or friends to be properly exposed even if that means much of the background goes white. Hence the matrix meter which gives bias to the AF point. Despite all the claims of sophisticated processing and scene memories and other such technobabble, matrix meters still ignore blown highlights in favor of giving the perceived subject a normal exposure.</p>

<p>This consistently throws people who want as much brightness range recorded as possible. It's why I tell people to master their spot meters and use them in high contrast situations. Just about every DSLR has a spot or partial metering mode and 2-3 meter readings will tell you all you need to know about a scene to nail it in a single shot with zero chimping. (Not that I'm against chimping, but you can shoot with that level of confidence once you understand your spot meter.)</p>

<p>Contrary to popular belief, DSLRs do not suffer from poor dynamic range. They blow slide film out of the water on this mark and are actually roughly the same as many popular B&W emulsions. (I don't know why B&W film has the generic reputation of being high dynamic range. It's the C-14 color portrait films which really knock the ball out of the park here.) What they suffer from is matrix metering which will let highlights clip any time the subject is perceived to be on the shadow side. The meters are probably making the right decisions for snapshooters. But advanced amateurs need a metering mode which will attempt to capture everything in the scene. The lighting optimizers in the latest DSLRs are a sign that Canon and Nikon know about this problem, but they need to take the next step and design the meter to work with the lighting optimizer firmware. This would eliminate 99% of the complaints like those in this thread.</p>

[philosomatographer]

<p><em>Daniel Lee Taylor wote: Even Velvia 50 can hold a 5 stop range, which is about half of what modern DSLRs can hold. Your B&W film sample does not show a scene with contrast that would challenge either DSLRs or B&W film. Indeed, the image is muddy and lacks any true whites which indicates a scene that did not cover the full range of the film, and a print which used too low of a contrast filter on multigrade paper.</em><br>

<em><br /></em><br>

Daniel, the way I printed <strong>my</strong> image is <strong>my</strong> aesthetic concern, thank you very much. You have no idea what kind end result I had in mind. </p>

<p>You also seemed to misunderstand my post. You appear to be claiming that, in my particular scene, which metered EV10 in the shadows, and EV15.5 in the highlights (5.5 stops difference) the entire brightness range of the scene is 5.5 stops. That's patently incorrect. In a world where everything was coloured 18% grey, maybe, but down here in South Africa (and I suspect where you live also) things have various colours ranging from black to white.</p>

<p>In my image, the brightness range from the dark, dirty pavement in the shadows, to the white clothing in the direct sunlight, both of which hold full detail, are well in excess of 10 stops. It was however my choice to compress them down into a very narrow display range, and in fact this exaggerates the original DR available to me in order to do so. It will still also take some more time than the year I have spent to learn to fully realise my vision in the darkroom, I have to control one variable at a time.</p>

<p>Please have a look at my portfolio, will you. I have for many years been shooting a range of digital cameras, ranging from point and shoots, to $7000 DSLR bodies. Why on earth do you think that, at this point in my life, I would switch to black and white film?</p>

<p>Because: apart from the very different tonal rendering which I find pleasing (personal aesthetics), there is absolutely no contest whatsoever in terms of the dynamic range between a typical black and white film like Ilford FP4, and <strong>any</strong> digital camera. Live with it, it's a matter inherent to the physical process of producing the captured image. This has been debated to death already, on this site and others.</p>

<p>Saying, at this point in time, that a digital sensor, with its linear sampling characteristics, has the same or greater dynamic range than a traditional monochrome film (because of the self-attenuating, non-linear characteristic) makes you look like a fool.</p>

<p>Why is it that, over the past year, I have not in fact *seen* a single blown highlight in any single image I have taken on black and white film? And this with "inaccurate" metering and guesstimation by virtue of me using battery-less/meter-less cameras. With a DSLR, using the same shooting/metering techniques, I fought blown highlights all the time. Just as I fight them when I shoot colour slide film. There is a <strong>huge</strong> difference, man! Negative film is simply incomparable to either of those two.</p>

<p>I do, however, wholeheartedly agree that <em>multi-layer</em> negative films (be it colour film like Portra NC 160, or Ilford XP2) have even greater highlight range, because when one layer of silver grains reaches maximum saturation, the others will often not have.</p>

<p>I agree fully that my two shots are in no way comparable, I simply posted my digital shot as an example of what happens when I boost the shadows on a Canon Pro body to maximise dynamic range, including the issues (such as read noise).</p>

<p>And yes, Harry's shot is quite fixable if it were taken in RAW format, and even his JPEG image is somewhat recoverable with careful processing.</p>

<p>In the end, I am still amazed at the directness of your attack against my post, as well as my image(s) - I was merely making conversation. We are on a Forum, are we not? Of course, you are merely sharing your "opinion" - so all's fair, I guess.</p>

<p>Cheerio.</p>

 

[philosomatographer]

<p>Oh, and Daniel, in response to your comment of my image holding no true whites, please, if you'll be so kind, open it up in photoshop and measure the RGB values of the woman on the bridge's shirt, or the hairdresser's homemade white poster in the bottom-right corner, and report back on them here telling us why it's <strong>not white?</strong></p>

<p>I believe the image I posted (which is not the most artistic image ever, I admit), with full detail visible in these whites, as well as actually <em>inside</em> the hairdresser's hut (which surely must have measured EV7 or less) is simply not possible with a single digital capture with current state-of-the-art DSLR technology. So much information is captured over such a wide range that I could "play around" to produce the image I posted here. It may look muddy to you, but it realises my vision for the shot (kinda) and represents detail across the full spectrum going from full black to full white.</p>

<p>Your statement of it not containing whites is thus simply incorrect. And I know for a fact that no DSLR can represent that range of tones in a single shot; I have done a lot of HDR in my life, using various techniques and tone-mapping strategies, and you'd need a minimum of two raw files for this particular range, in my experience.</p>

<p>Of course, the DSLR images will be nice and clean, not grainy like my image. There are many drawbacks to film as well. But DR is not one of them.</p>

[matt_needham]

<p><em>"I think film could have handled these situations allot better."</em></p>

<p>Print film would be more forgiving and allows for less exposure accuracy. Your results are typical of someone who is used to the forgiving nature of film when they first try digital. Getting the exposure right within a stop or so is no longer close enough. Keep practicing. It's not up to the process and materials to handle the situation. It's up to the photographer. :)</p>

<p><em>"I hate to Chimp..."</em></p>

<p>Why? Chefs taste the food as it is being prepared. Writers read what they have written. All other visual artists get to use their eyeballs when working. Only film photographers try to create visual art blind and in the dark. It's not a strength. It's a weakness. Use your eyes and mind; they are their for you to take advantage of.</p>

<p> </p>