« Marty Forscher (A Keeper of the Flame) | Main | Citizen's Arrest »

Monday, 05 October 2009

Comments

Thanks - makes me realize how Neanderthal my process still is. I just look at the thumbnails, select consecutive image groups (3. 5, 10 etc) which are clearly of the same item/event/viewpoint and cycle through them in Preview until I have decided on one (or more) worth carrying forward. BUT I write the winning image numbers down on a notepad. (OK pause for laughter). Crude but it works.

I have a similar workflow in PhotoMechanic. The main difference is, the very first thing I do is apply a batch caption to every frame. I include the date, location, and general subject (not specific to each image). PhotoMechanic has an excellent global captioning tool that does this to all the images at once, very, very fast.

Later, when I've made my final selects, I write a full caption for those images.

Adding this small but important part of the workflow insures that I can find my images later, and know what was going on in them.

Thank you Carl for this helpfull contribution. I used to work like this but have since changed it somewhat. Instead of 600 pics I shoot 60. Focussing on what's important helps me get better pictures and it saves me a lot of time behind the computer.
I download the new pics, maybe select 1 or 2 for direct use on blog or flickr and leave the rest sitting there for a while. I then go back about 6 months and look through a day of old pics. I usually delete 70% of them in 5 minutes and tag what's left with keywords.
Mind you, this only works for my private work, which is what I do mostly.

My first edit isn't for the good photos, it's for the bad. Blinking, bad focus, wrong moment, bad exposure, missed light, or just plain awful....all get tossed. I'm ruthless in this edit, and will trash up to 1/4 of the photos. Only then will I attempt to look for the winners.

I know there are many who shudder at the thought of throwing out digital files, but for me, it's necessary to have that clutter gone.

Thanks for the article Mr. Weese, but I gotta say, this is exactly the problem with "free" shooting in digital. Digital is definitely free if by that you mean "shooting on a cheap card with $2000 worth of equipment and ending up with 2000 snap shots that will require 20 hours of your life just to 'process.'"

That is not cheap unless your life is cheap. Time is not free. As I'm sure you know (or else you would not be trying to waste time as efficiently as possible with your "streamlined workflow") time is the most precious currency we have and digital, frankly, wastes it. In my opinion, film because it has intrinsic value, makes you more likely to invest each shot with more value, ie, time.

But even with digital, it will PAY you (in time) to take your time when you're shooting. Open your eyes, wait for something interesting, wait for good light, and only shoot when you know you have something. And only shoot it once or maybe twice if it's really awesome. Without that approach, you'll produce nothing memorable, but you'll produce a lot of it.

(Unless you're a pro shooting sports).

As you pointed out, you are going to miss shots whether you shoot 10,000 frames or 100 and coming home with 100 frames will give you time to mix your chemistry or read the paper or make a soup or watch the grass grow or write an article about photography (as opposed to software) or be with your family.

Love your wife/dog/husband, not your computer. That is, unless you're a techno guy and in that case why even take pictures in the first place? Just download some pics of goats off the internet.

I was driven to buying PhotoMechanic some time ago by the performance issues of sorting hundreds of RAW captures. I've never run Bridge intentionally, but the times it's come up accidentally it appears to be glacially slow (nor does it produce interesting lakes :-)). I use Thumbs Plus from Cerious Software for "digital asset management", but it was even slower at viewing NEFs than Bridge was, so doing my sort there wasn't an issue.

Like Chuck, my first pass is mostly eliminating technical failures. Those are quick and easy to kill -- and I want them gone BEFORE I rename the files to my standard naming scheme, to reduce the gaps in the sequence numbering. (My files are named as "ddb yyyymmdd ses-seq-orig.ext" where "ses" is a session number, to distinguish separate locations shot the same day (or potentially different sessions could even be at the same location; two client shoots would be two sessions even if they were in the same location), and "seq" is a sequence number within the session. The rest is obvious, right?)

My big problem is selecting from among very similar photos -- if I'm pushing the limits of shutter speed (relative to subject motion or camera stability) for example I'll deliberately overshoot fairly heavily. Or if the subject is fast-moving. I'll even use continuous shooting mode for some of those cases. Those can be hard to sort, but they don't upset me since I know what I was doing and why. Now and then though I end up taking 5 nearly identical pictures carefully one at a time. What WAS I thinking? Bah.

I suspect I'll be happier if I adapt your unabashedly multi-pass process. I've been trying to do this mostly in one pass, and I keep going back and resorting things a bit; I'm clearly sneaking over that direction when I think I'm not watching.

That's why I prefer a medium format film camera with 8 pictures per roll. It
makes me look closely before making an exposure, and saves a lot of editing time.

"As an aside, if you can find the discipline, put off this whole process for a while after making the pictures.""
and
"Last note—the longer you let this process take, the more time you let pass between shooting and editing, or between the editing steps, the better your edit is likely to be."

This is the best guidance of Carl's essay. After a few days...or weeks...or months...your non-visual allegiances ("That took me hours to snap!", "I nearly got killed getting this!") to an image will fade leaving mostly the residue of the image's core value.

But there is also great value in making some very fast judgements, too. The best images often have a universal grabbing power irrespective of capture backstories. Personally I always make a fast, hard cut with Bridge as soon as possible, deleting the real stinkers and the obvious under-performers while assigning a star to those that immediately grab me. Then I walk away for a week or more before importing the remaining images into my Lightroom data base, assigning keywords, etc. At that time I nearly always make more cuts.

I think the bedrock message value in Carl's post is to be consistent with your processes and self-aware of what biases you. If you're an amateur photographing for your own enjoyment you can adopt any process time-frame you choose. Successful and experienced vocational photographers adopt efficient screening processes early and tend to stick with them.

One additional note regarding the time value of judgements. Using Lightroom's powerful "smart collections" facility you can dynamically evaluate the status of your own judgements. You can, for example, create collection specifications such as "4-star-rated" + "pumpkin" keyword, from which Lightroom will continuously gather images meeting these conditions. It's been fascinating for me to revisit what I considered "best" from even two or three years ago.

Finally, Carl's remark about "intention and audience" is worth reinforcing. Lensers are often poor judges of image value. Their eyes get clouded by the camera/lens they used, "dynamic range", and assorted other useless drivelites. If you plan to show your images to others consider their point of view and what, if anything, you're trying to covey.

The complexity of it all makes me long for a Leica M6 and a few rolls of B&W film.

Shoot the pictures.....take the rolls of film to the developer......wait a bit......pick them up.

Game over.

There's no "right" answer to shooting a lot or a little. Some people may have a natural tendency to be more comfortable one way or the other. I like to go back and forth--for many projects/subjects I work with 8x10 and larger view cameras and six or eight exposures is a good productive day's work. Other subjects/projects are better suited to exploration with a small hand-held camera and lots of film/memory.

One way to look at it is that in one case you edit before shooting, in the other case, after. Neither is inherently superior, each is a valid way to work. One advantage to the latter is that editing a take is an excellent opportunity for self-education, which is not at all a waste of time.

I keep it simple: in Lightroom I quickly scan and images, flag some as rejects, flag some as picks, and leave the rest alone. Then go back through once or twice more. I'm relentless and rejects are obvious rejects and I almost immediately delete them. The 1-5 stars rating system is too complicated for me. I either don't like an image, like it, or not sure.

Carl’s suggested process is a good one. It intrigues me, however, to find his repeated point that it is better to wait—the longer the better—is completely opposite to my preference. On a three- or four-hour shoot I may bring back a hundred or so images. I always try to review them as soon as possible after the shoot since my memory of the feeling I had when taking that picture is still lively and will absolutely condition whatever post-processing I do to the selected images. I regularly go back to old sessions after a few months to see whether I missed anything but truthfully in the past two years, say, I’ve only found three or four missed surprises out of a couple thousand.

A couple of people, some directly, have raised the issue of who you are making your selections for. It's a key question, of course.

If you're selecting for a client, you need to leave them some room to have their own needs and opinions. You may not even know if they need a horizontal or a vertical, depending on the terms of the assignment.

If you're shooting journalism, you *probably* don't know what size or orientation will be most useful to the people doing page layout.

I know Life photographers and others in that era frequently shipped exposed rolls back, and the processing, printing, and editing were handled back at headquarters, mostly without any participation from the photographer. I don't know how that works today for photojournalists; do they ship back entire flash cards full of images, or do they select first? If you ARE doing selections, you may well need to leave in mediocre photos, if they're the only ones you got of something you can tell was important.

I think (NOT MY FIELD!!) high-end advertising photography generally has a rough page layout showing exactly where the photo goes, doesn't it? So one can work towards something fairly precisely defined?

And art, of course, is more of a solo process, wherein you work however you like (or can afford), and are answerable to nobody (unless you're trying to make a living at it).

Herman: Wait a minute, *8* exposures per roll? I know 6x7 gets you 10 normally, so you must be shooting 6x8 or 6x9? Not I think 6x17, the math just won't work out. Even 6x12 seems wrong (or am I imagining that aspect ratio?).

Glad it works for you! A lot of my photos couldn't be shot that way; the subjects won't sit still long enough, or close enough, or far enough away, or something.

I'm with the "slow down" and "simplify" crowd. Forty+ plus years ago when I started using a camera I went through the "machine gun" phase. Then along came motor drives and machine gun became "gatling gun". That came to a crashing halt the first time I picked up a Rolleiflex. My exposure rate went way down but the "keeper" rate went way up. Saturday I spent with a couple of shooting buddies driving over 400 miles shooting fall color. Mattie shot over 300, Keith about half that and I got 91 of which 21 are worth keeping. If past experience is any guide that's about twice what they will get. Stop, look, consider and then (maybe) shoot.

Generally I'm winnowing down for post processing, so I use a similar multilevel sort in Lightroom. First past to flag hits (anything not flagged as a hit is assumed to be a reject), second pass to rate, third pass to do close examination on the ones above the threshhold and either correct the rating or start processing. Since I do most of what I'm going to do right in LR, if I've done my work at the shoot, I can plow through a pile of images quickly.

I know a lot of people are nostalgic for the 'good ole days' when photographers magically took better pictures because they were constrained by film, but honestly I think it's a question of what you bring to it. Motor winders were invented in the film era, and I'm old enough to remember bulk film backs, the silver halide equivalent of a high capacity memory card... people took plenty of images back in the day. The biggest difference was hobbyists didn't take tons of pictures. Digital has leveled that playing field.

There's something to be said for not ripping off 6 frames when one will do, but there's also something to be said for powering through a burst to make sure you've got at least one where no one is mid blink/sneeze/goofy facial expression. And digital's "free" images give you far more leeway to experiment; when I'm shooting with my film camera's I'm keenly aware of every $2/frame (final cost for 120, obviously 35mm is cheaper) shutter press, which means my focus is on getting images I know will work. Frame big. Stop down a bit more for extra DOF. Shutter high enough to guarantee no shake. Digital (in most cases, some things only happen once) gives the latitude to capture tons of images you know going in will likely be garbage, but if not they'll be really interesting. What happens if I drag the shutter/open up the lens/hold the camera over my head and shoot? Sure, you can do that with film but a) you don't know the outcome until later and b) have to be willing to spend a couple of bucks to find out.

And I never went from film to contact sheet to finished image as fast or easy as I can go from memory card to final print.

Despite the selection and tagging system being personal preference, I think there are ways that the presented system can be improved. For instance, I prefer not to use the star ratings as Carl has because it means that those ratings will have to change based on target audience at any given time. He states that he uses 2 stars to indicate photos appropriate for the target audience he's selecting for, while 3 stars is for good photos that don't fit the current set/purpose. What about a year from now when you want to extract a different set for a different audience from the same session? The star ratings have to all be changed, and if you ever want to revisit a previous set you're out of luck because you've changed your star ratings.

I don't know if Bridge has this, but in Lightroom for instance you can create a "collection" and add photos to it. Say I go to cover an event but also bring my family because it's a family-friendly event. From the resulting photos I can create "work" and "personal" collections and "copy" appropriate photos to each (This doesn't take up significant storage space since you're not making a physical copy of the file but rather creating an entry for it in the Lightroom database, which takes up minimal space). In a year if I have a different target audience I can create another collection and copy appropriate photos to it as well without disrupting the 2 existing collections. You can do this same thing with keywords also. I feel star ratings are best used for rating the quality of a photo instead of trying to categorize them for a particular target audience.

To Jeff Glass, your suggestion to spend more time on each photo and shoot less of them is ok for some situations but not all. Sports, as you noted, is one example. Event photography (weddings, parties, concerts, etc.) is another. Any situation where the technical challenges of the shoot are not insubstantial (moving targets, rapidly changing light, etc.) is a situation where it behooves you to take multiple shots to make sure you have at least one that is technically acceptable. The freedom of shooting digital is that you can shoot more profusely when the situation warrants without worrying about film and processing costs. That's not to say that you should shoot wastefully, but high volume does not necessarily equal waste.

We used to "shoot" now we "capture". Do we plan on letting them go later? I'd rather "take pictures"-less violent!

If this is really how you are using bridge then you really really really should be using Lightroom instead.

One metacomment...I'm all for slowing down and shooting less too, but let's face one fact here--the most useful answer to the question "how do you edit a large set of image captures" is not "don't shoot a large set of image captures."

I'm not criticizin', just sayin'.

Mike

There have been a number of comments about what a waste of time it is to edit a large set of pictures. I'd like to emphasize a subtext here by paraphrasing some self-help I've seen in other places, "If you regard editing your photographs as drudgery, you ain't doin' it right!"

And I don't mean the technicalities, but that editing is an important learning tool (quick nod to folks who've pointed out that Lightroom, and other programs, do a better job of catalog tasks than Bridge, nolo contendere, but I don't like its user interface and do like the way Bridge and ACR interact: my catalog needs are pretty simple). My point was not to give Bridge tips, but to recommend a general approach to editing that can make it a valuable experience where we actually learn about our work.

Way back in, I think, his first book, David Vestal wrote (paraphrasing here) that a photographer's most important piece of "equipment" was the collection of binders or boxes of contact sheets that constitute one's photographic autobiography. The contact sheet has been replaced for digital capture by the browser folder, but the opportunity to use the editing/study process for self-analysis and improvement of our photographs remains.

Just shot 10,000 images over a 135 day outdoor photography contest and won $1,600 - almost a 1/3 of the way toward the new M9. I commented about this on the rangefinder forum which prompted Al Kaplan to reply he would put a bullet to his head if he ever had to edit 10,000 images. I am sure others look at 10,000 images and say "no way." Wrong. Having been a pjist for almost 30 years, I never "rationed" film (and never ran out of film on a job either - almost did one time in the Middle East however I found some rolls in "Aramco City") and I will NEVER ration pixels. The problem is with wildlife and birds especially, even with a 1dmk3 at 10fps, you will miss some captures. Some good captures too. Your editing technique is almost identical to how I cull 99% of my images - I use CS4/Bridge - and it leaves me peace of mind when I have that shutter pressed on continous, the buffering is heating up and the card is about at maximum. My mantra with DSLRs is keep shooting, I will cull on the computer. Good editing/culling balances very nicely with keeping the trigger finger pressed.

PhotoMechanic has a two-dimensional rating system -- 8 colors and 5 star ratings, I think it is. I use only the colors (which I use as three sets of quality levels roughly meaning A, B, and C, plus one I use to mean "delete this"). (Conveniently, PM lets you select which set has the really convenient set of single-keystroke hot-keys.)

Does anybody use BOTH? Seems like the point has to be to use them together, but I haven't yet found that useful. Nobody else has mentioned it yet; but if somebody does find it useful, I'd love to see an explanation of how it's useful for you.

I have to disagree with Jeff Glass. To me, this doesn't illustrate a problem with digital, but rather a strength of digital. Narrowing 631 shots down to a few doesn't suggest poor shooting at all. Nick Kelsh' "How to Photography Your Baby/Kids/Life, etc" books demonstrate the difference between how a pro photographer and an amateur shoot by showing contact prints from 35mm negatives by a pro which show someone "working the scene". And in "On Being a Photographer" the authors discuss how most pros who shoot events/photojournalism can show contact sheets that demonstrate how they use their camera to explore their subject, working toward their best shots in each "situation" as Carl calls is. And that was pre-digital, no chimping.

I shot nature before I started shooting people & events, but even shooting nature (scenics, macro) I rarely ever analyzed a scene and stumbled upon the best shot without some experimentation. Certainly now that I'm shooting people, I shoot a lot more. I try things, some don't work. Some do, I get better. And I don't care if I've got the worst "hit rate" on the planet (which I don't, I'm sure) ... the more I shoot, the more good pictures I get.

We shouldn't look to the large format photographer as the ultimate authority on anything here; there's nothing inherently more artistic or successful about the works of large format photographers than the works of small format photographers, so even if their reasons for limiting their exposures were more about working style than logistics, there's no reason to adopt their practices. (I bet a photographer like Garry Winogrand could outshoot Carl in an afternoon, even with film :)

I shoot a lot. I delete a lot, but shooting a lot presents me with the 'problem' that I simply end up with many shots I like, far more than I ever shot with film. Partly due to simply shooting in more situations, partly due to exploring each situation more, and partly because the more I shoot, the more I learn.

- Dennis

Years ago, I lived in the same neighborhood as David Hume Kennerly, who was Gerald Ford's White House photographer and has numerous Time magazine covers to his credit. I *think* it was David...I'm really not 100% sure, so don't quote me...who told me that there's really only one difference between professionals and amateurs in photography. "We shoot more. Not a little bit more. LOTS more."

Or words to that effect. Again, sorry about my porous memory and the possible wrongful attribution.

Mike

Interesting post ...... & a reminder that there's more than one way to skin a cat. (Don't flame me, I just hate cats. But dogs are good).
I cover sports, motorcycle road racing predominantly & can (& often do) take 3000 images a day. These are then edited, keyworded, captioned etc & shot off to agencies, all one with Aperture 2 within a few hours. Live & learn, there never is a "right" way. Just as well or we'd all be clones. Scary.

I really suck at editing my own work but -

For a while I worked as an in house paparazi/PR/ celebrity party photographer, back in the Studio 54 days. One of my co-workers was tasked with going though a pile of about 1400 chromes I had shot at a party. First thing he did was cull the really bad ones where there is something obviously wrong like weird composition, blown out foreground general strangeness. Then he made a few more passes to include get all the celebrities, rich clients, random nakedness and whatever until there were about 120 slides left. I had tossed the first cull onto another light table minus the completely blank, and out of focus and was poking around at them when the the art director and the PR woman came to make their pick. They of course thought the culled photos were more interesting than the "good" photos and ended up using a 50/50 mix, and I ended up being the guy that could photograph really boring events and make them look like an out of control riot. Going through what I can find of my work from 25 years ago for a project, what I find to be good and interesting is completely different from what I thought was good and interesting back then. Somewhere I have probably hundreds of negatives with Madonna standing in the background from 1981, and if I were in the habit of throwing out boring photos, they would be long gone.

Oh, and back on topic, some stuff looks good big , some stuff looks small , hardly anything looks good both ways


I'm always intrigued and maybe a little baffled by the whole idea of leaving time before processing or editing pictures.
For me, my thoughts, feelings, intentions at the time are imprtant to the end prodcut and so i want as little time as possible between shutter release and first processing.
A final edit for presentation may benefit from a wait but picking those that meet my intent needs to be done in short order.

Here's another anecdote. Ken Heyman was a strong influence on me when I was a teenager going nuts over photography (the 1960s). I read an interview, probably in POP, with Heyman where the main point was that he got his incredible intimate pictures of people by being infinitely patient and "as boring as possible, so I just disappear." But it also mentioned that he thought 18 rolls of film was just about right for a working day. If he shot less than that, it meant things weren't happening, he wasn't up on plane. But if he shot a lot more than that, it might mean he'd been struggling and pushing too hard instead of finding the groove. Those would have been 36 exposure rolls of Tri-X or Plus-X shot in non-motor-drive Leica and Pentax cameras, with little or no bracketing of exposures. (He probably averaged more like 30 exposures per roll, because the practice was to reload as you approached the end of roll, during lulls in the shoot, to avoid having a great situation develop when the camera body with the right lens on it was on frame 35...)

Thom Hogan had a recent (24 September) post similar to this (but a different variation).

http://bythom.com/shootless.htm

I shoot a lot and I delete *gasp* nothing.

Well, almost nothing. There are a few gaps in my photo file names that resulted from "no visual/emotional value whatsoever" tests. I keep everything else, even the utterly rotten failures.

Why? Because I shoot primarily because I become able to augment my memory of the event, outing, situation, etc. Of course, I like nailing great shots that can tell stories and grab attention. I don't want to bore my friends or relatives with the sub-par.

Still, going back and browsing through my monthly folders (many of which contain 600+ shots) is a large part of the fun I find in photography. I can go back and recreate those moments or hours or even weeks simply by sequentially browsing through each shot, as shot. Most of these events are parties with friends, so a large portion of this behavior is related to us crazies having fun.

I'm not a machine-gunner and rarely use my cameras' continuous shooting function. I just press the shutter button when I see something that hits me as worth saving. I've never tried to measure my efficiency, but I'd guess I end up really liking about 30% of my shots. But nearly all of that 70% still have value.

I act differently when shooting film, of course. I enjoy the limits of the medium, but I take this gear out knowing one sad fact: I will almost certainly miss a great moment because I'm holding back to save film for later. And those great moments often don't reveal themselves until I'm browsing these on a computer.

Such are our differences.

I don't think that slowing down and being careful necessarily means that you are shooting a small number of pictures. You should be shooting a large number of pictures all of which have a good chance of being a winner. I find it can take me a few hundred pictures to get "back in practice" again to be able to see the interesting things that are in front of me. I didn't shoot enough before a recent vacation and as a result only really got in a groove taking pictures in the last half of the trip. That's tragic!

After all, who knows when you can get back to that spot with that light with an opportunity to capture *that* moment. It's likely you never will be, so get it together and get everything you can. I say.

That said, editing needs to be fast and efficient. I find that with LIghtroom, a modern laptop, and 4GB of memory the main bottleneck in editing is getting the pictures off the card in the first place. After that it's resisting the urge to tweak the settings instead of editing down the shoot.

Otherwise, I do something similar to what is here. But I really only make two passes. I delete everything that isn't worth keeping, and then later when the time comes I pick the "selects". I am not a pro or even a semi-pro so my criteria for selects is entirely personal and just means I'm tossing on the web site or to Costco for a print.

I should probably be more careful about categorization and organization. But instead I keyword the stuff and keep the best ones. Later in my life I'll probably go through the files again year by year and find new favorites that I missed.

Sports Illustrated's digital workflow at the 2004 Super Bowl (industrial-grade stuff):

http://www.robgalbraith.com/bins/multi_page.asp?cid=7-6453-6821

Great stuff Carl. Love the Sudek quote (and agree!)

I am reminded of William Eggleston who allegedly only takes one shot - if he misses it he says there's always another one around the corner. In his case, there always does seem to be another one.

1. I'm a "techno guy", and I find Jeff Glass's comment

"That is, unless you're a techno guy and in that case why even take pictures in the first place? Just download some pics of goats off the internet."

both illogical and insulting. It's a good thing us techno guys have no emotions, or I would feel hurt.

2. Deleting individual images from the card (which someone here suggested) while it's in the camera and therefore represents your only copy is a really bad idea. Random glitches can destroy the whole card, and deletion is one way to encourage those glitches.

--Marc

Seriously, don't delete in camera. It's just going to get you into trouble. I'd say don't delete at all (why would you...you maybe are going to save 10% of your disk space if that) but in camera you are opening yourself up to all kinds of screwups.

I did this once, when on a long trip with somewhat limited memory cards, and I think it worked out ok. But who knows...I deleted a lot of duplicates, and the duplicates might have been better than the ones I ended up selecting in camera. I really wish I had all those pictures.

This is probably a dead topic by now, but let me just note re: David Kennerley's comment about pros shooting lots more: the difference between a pro and an amateur is not necessarily the difference between a good photographer and a mediocre one. Often, exactly the opposite.

I shot a two day event recently and ended with about 900 shots. Which, in turn, ended as about 200 keepers and that only because the client asked for approximately that number. (I think it's a wrong approach on the part of the client, but...)

For my personal stuff, the final ratio is usually about one to ten. And then I narrow the number even more before letting the photos go out in the public.

I almost never use burst mode, sometimes even making the mistake of not using it when it would be more convenient or would result in a keeper. But I do have 3-4-5 photos of the same thing comparatively often: I take the first one and don't like it or miss something. Take another one, not so sure about it anymore. Take another one, the subject has closed eyes when they should have been opened. Etc.

My approach to winnowing is quite similar to Carl's. I download the photos into a folder, open them in Bridge, check for those that I like, develop them and save as 16-bit PSD and then post process them in Photoshop. I leave the undeveloped ones in the folder and afterwards (long afterwards) pass through the photos again to see whether I missed something. And then delete the unsuccessful/undeveloped ones.

The comments to this entry are closed.