State of the VR World

State of the VR world in 2003/2004:

A couple weeks ago I read a NY Times article in which a woman described waiting while her boyfriend went beyond shooting single shots to capture a 360 panorama with his new digital camera. I gathered that these two were not in the photography or panorama business. Digital cameras are totally mainstream, and if the recent NY Times article is any indication, shooting real (360 degree) panoramas with digital cameras is something many people either do or want to do with their new digicam toys. More and more digital camera companies are providing panoramic software tools with their cameras.

With nearly every citizen of the planet a user of Photoshop or equivalent s/w, and many billions of digital camera and cellphone pictures being shot every day, the Internet cup runneth over with digital camera images and panoramas. Earlier this year I was looking for images of a rarely used trail into the Sierras and was surprised at how many people were posting hundreds of snapshots from their trips to this remote area. Weblogs are a factor as well. Panoramas online have been relatively sparse only because no one has made it totally easy to build and post panoramas. Perhaps this will be the year someone finally puts all the pieces together and make it as easy to make panoramas as to crop an image and optimize the color in iPhoto. Because while it has never been easier to shoot panoramas, the amount of effort to build  and post them means many people never finish what they started.

The arrival of multiple Digital SLRs under the $1000 price point signals the end of film cameras as useful devices for most people, and the ubiquity of digital cameras. The snobbery of the upscale car or computer marketplace will probably evidence itself whenever someone whips out a digital SLR in any corner of the planet. Most digital cameras will have a useful life of 9-18 months, just like computers. Where do a billion old digital cameras go to die? Probably in a landfill. This impact on the environment will grow to rival big, bad silver processing. Whereas before a good 35mm lens could be used for most of one's  life, increasingly specialized lenses will only have a useful life in tandem with the camera system -- do you think anyone will keep their D100- or Rebel- specific lenses once these manufacturers come out with full-frame DSLRs? The digital imaging units of the Canons and Nikons will probably expand like the PC companies did in the 1980's.

Photographers shooting with film are increasingly a small minority. It's a video-game/instant results world and only monks have the patience to wait for 1) film to be processed, and 2) film to be scanned. I wouldn't be surprised to see the film SLR camera market consolidate to a few models per manufacturer. Specialized films will disappear as well. However film scanners (like the new Nikon CoolScan 5000) continue to improve what can be extracted from film, so those 1995 film panoramas can look better and better every year.

The percentage of people shooting full spherical or cubic panoramas will probably not grow dramatically since it still takes about three times as long to shoot a multi-row scene as a cylinder (assuming a typical scenario of a person with a wide-angle adapter on a CoolPix 5400, or a D100 owner with a 14mm lens). When I go on trips with others, most people will tolerate  5-10 minute breaks to shoot panoramas, but few people will accept a bunch of half hour breaks to shoot cubics. And then there's the added time and frustration that building cubics entails. The arrival of full-frame DSLRs will make it easier to capture cubic scenes; maybe this year someone will come out with a QTVRAS for cubics which will make it as easy and productive to build cubics as cylinders.

The announcement of more complete 16 bit support in both Photoshop CS and Stitcher 4.0 is a turning point for the shift from 8 bit per channel color. We've endured nearly ten years of stagnation in the color depth of the panoramic workflow. Now you can capture, edit, and stitch without throwing away color information. I expect HDR to be a key discriminator of high-end work for a few years until all the other tools catch up.

Both Manfrotto and Peace River getting into the cubic panhead business signals an era of increased competition which by any economist's measure should result in lower panhead prices. I bet it will take panheads made in China to deliver truly cheap panheads though.

GPS integration and geolocation are a key nexus for a lot of development this year with mobile technology and photography. The Nikon D2H hints at the possibilities. This will make it much easier to produce panoramas that are automatically posted on the Internet with names and locations. Presently associating metadata with images is one of the most time-consuming aspects of putting panos on the Web.

Since it now takes only a few thousand bucks to equip yourself with everything needed to shoot very detailed panoramas, I expect to see a new flood of people jumping into the panoramic photographer business. With every DSLR producing  high-resolution 5-14 megapixel images, the making of panoramas will increasingly become a boring matter of how fast you can shoot and stitch-on-a-laptop production. People who can cheaply put a panorama on the Web minutes after shooting, or return to the office and process 500 scenes overnight will probably dominate the field. Expect many of these newcomers to the field to offer to shoot VRs for free, as long as they can put ads in the scene. Who in this world ever passes up something for free? Consequently many VRs will start looking like webpages with flashing, eye-hurting color, fake-Windows-error-message, blinking ads embedded in them.

At the same time, as happened with desktop publishing, people who formerly hired photographers to shoot panoramas will just grab a digital camera and shoot a 360 themselves.

Increasingly panoramas have become hip. Planetary explorer craft are expected to produce panoramas. Whole stores do nothing but sell panoramic prints. But the fall from hip to dross goes quickly, and panoramas will have become routine, commonplace, and dirt-cheap-to-make. You could call it the Dell phase. And like when print photography was new, it will take years before the average person (or the average VR buyer) becomes experienced at distinguishing a quality panorama from a something a robot captured.

Less clear is how all these zillions of panoramas will be presented. I suspect many will be posted as super-wide JPEGs. Java playback using 10,000 different whacky and unpredictable user interfaces will continue to be popular, to the continued detriment of VR usability. I would guess QuickTime playback would grow in absolute numbers but shrink relative to the overall pie, and tend to be the niche of integrated multimedia VR projects.

Are the trends good or bad? Well this outcome is so much better than what happened to the VRML world.

    Copyright 2004 by erik goetze

Last update: 1/6/06; 2:41:58 PM.