Slashdot Mirror
Photographers, You're Being Replaced By Software
Mrs. Grundy writes "CGI software, even open-source software like Blender, continues to improve in quality, speed and ease-of-use. Photographer Mark Meyer wonders how long it will be before large segments of the photography industry are replaced by software and become the latest casualty to fall to outsourcing. Some imagery once the domain of photographers has already moved to CGI. Is any segment of the photography market safe? Will we soon accept digital renderings in places where we used to expect photographs?"
CGI has a LONG way to go before it can replace a good photograph. A well-composed, well-lit photograph can say more than most 3D animations ever could. And a photo is a lot easier and cheaper to produce. Who is going to pay a team of digital artists $100 an hour to create a 3D model of something when you can just tell Jimmy Olsen to go take a picture of it for a pittance?
The software to do 3D may be getting easier and cheaper. But good 3D artists aren't. And a single picture of a wounded, crying girl in Syria will always have a helluva lot more power than any 3D rendering of the deployment of Syrian forces. Photography isn't going all-CGI any more than movies are.
What political party do you join when you don't like Bible-thumpers *or* hippies?
I know that when I get married, I won't be hiring a photographer. Instead I'll hire Pixar to make a 15 minute short commemorating the occasion.
Not to be semantic, but this is not outsourcing. Outsourcing would suggest that they'd hire a photographer overseas to do the job at a lower rate. This is elimination of the job by technological advance (not sure if there is a buzzword synonym or not).
I've done all sorts of photography professionally - from fine art, documentary, photojournalism, weddings, to commercial (not at the same time). And by professional, I mean, actually getting paid for it and making a living and renting cameras, grip, lights, assistants - the whole gamut. I have since switched to 3D and I tell you it's slower because I have to do everything myself. It's not like animators or modelers are clamoring for still image jobs. I have to model, texture, build the shader, and light the scene everything myself (which isn't hard with my background - but radiosity is another matter). That's at least a 2-3 week additional work time for a project.
Photography won't be replaced by CGI any time soon because the former is faster. I can hire a crew and equipment and finish a shoot in 10-12 hours tops. CG supplements it with set extension or environment/ object replacement, but to create something CG from scratch takes a very long time. I give CG this: it's easier to setup lights whereas in real life you need an electrician or a generator for larger projects, especially if it's on-location outdoors. You also need a lift and an experienced assistant to operate them, and an impeccable sense of where the wires are of course, have safety in mind at all times. With CG, I just click a light node and bam, I can duplicate 2k lights down a tunnel for a car shoot. Obviously, the downside is the render time, particularly when you have to bounce and diffuse it but if you can segment the 6k image to different quadrants per render node, and rent a render farm, it's efficient.
Overall workflow, photography is faster in my experience only because there's people available to hire. Where I'm at, there's not too many freelance CG artists, or artists who knows lighting (because it affects the shader and vice-versa), and almost no photographer/ assistant know how to do CG. I seriously doubt CG will overtake commercially produced still photography (as opposed to wedding, event, documentary, etc).
A photograph requires a subject - similarly, a cgi render requires scene data. If you have the scene data, such as a product model, or a mountain, then you can take a virtual photograph by setting the lighting, framing the scene, etc.
So let's say I want an image looking up a tall skycraper from the ground. I could go out, find a location, wait for the right weather and lighting conditions and take my traditional photograph. Or, if I happen to be able to find a skyscraper model, I could easily compose the exact scene I want in my computer. Faster, probably. And maybe with Google's or someone else's increasingly accurate data, it could be an actual skyscraper and not just some stock model. So yeah, this will replace a lot of traditional photography, without a doubt.
But art is always up to the artist.
must... stay... awake...
Chalk another one for the lawyers.
One advantage of CG is the elimination of concern over buildings and landmarks that must be licensed to appear in film or advertising. Movies and TV shows filming in NYC (especially Times Square) commonly replace buildings and advertising with computer images to avoid licensing and copyright concerns. It can be expesive when something falls between the cracks. An example - The movie poster for Spiderman featured a reflection of the Chrystler Building in Spidey's lens. The owners of the Chrystler Building sued and the poster was recalled, reprinted with a photoshoped image, and redistributed.
When starting from scratch with a full CG image all parties involved can be sure there will be no such FUBARs. Advertisers want eye popping images. Most are pretty flexible about the source.
Already being done, I know from friends of mine who work in IKEA PR-department that they have been constructing scenes in 3D modelling software instead of photographing for at least 3 years already.
All joking of robots replacing artists aside, there is some truth to Mr. Meyer's concern, and I have felt the consequences. Allow me to share a brief story as my first-ever Slashdot comment.
My father migrated to the US as a young man and high school dropout and learned to make his living as a commercial photographer. He got good at it, and though our family has never seen an excess of money, we've been able to survive from his steady income from the variety of low- and high-profile clients. (Though the payoff wouldn't suggest it, some are extremely well known -- anyone with a credit card looks at one of his old photos every day. #brag)
The big, then-ugly switch to digital came and he closed his dark room and turned on his Mac. Since then, his most steady client was one of the top major US mobile phone carriers. For well over a decade (maybe two, I have a terrible sense of time), he shot all of their new phones for packaging, billboards, etc. The market slowly changed and other customers stopped asking for photos, but this single company kept our family afloat.
Last year the company hired a new marketing firm who convinced them to switch from hi-res photographs of their products to 3D computer-rendered graphics. My dad was no longer needed and after many years of nonstop work they dropped him. My dad, who is now 72 years old, with no other steady photo customers, has had to make changes. He packed up his photo studio and set up shop at home for occasional shoots, but is applying to grocery stores so he can pay the bills. (Being adaptable and resilient, he has also been teaching himself digital video editing in an effort to get into that market.)
What about the phone company's switch to rendered models? Apparently it has gone poorly for them. Word has it that the switch was a disaster -- turnaround for new images was neither cheaper nor faster -- and they have fired the marketing firm, who in turn, consequentially, has had to make massive layoffs.
So back on topic, yes, there is CG replacing some photography in the real world. But such an "upgrade" seems yet early. Maybe even premature or naively futuristic. Make what you will of its outcome.