art_major Posted September 10, 2017 Share Posted September 10, 2017 hello, I'm interested in making some stereotypes with a large format camera. I was thinking about connecting 2 crown graphic cameras together, side by side. Would this work? I'm just not sure about how much spacing I need between the cameras. Thank you! Link to comment Share on other sites More sharing options...
chauncey_walden Posted September 10, 2017 Share Posted September 10, 2017 The normal spacing would, of course, be the spacing of your eyes. Impossible with 2 4x5 cameras unless you are looking for the hyperstereo tabletop look. Another option would be to use one camera on a left/right oriented slide bar where you could vary the spacing from eye distance to whatever the length of the bar would allow to get the effect you want (assuming the subject is static.) For dynamic subjects you would have to mount 2 lenses on the same camera and rig a divider in the bellows to separate the images giving you a pair of 2.5x2 images. Link to comment Share on other sites More sharing options...
art_major Posted September 10, 2017 Author Share Posted September 10, 2017 I want to use 2 cameras, I don't know how far to keep them apart from one another. Link to comment Share on other sites More sharing options...
dan_fromm2 Posted September 10, 2017 Share Posted September 10, 2017 Art, as Chauncey said, as close as possible. This means removing the strap and associated hardware from one of the cameras and the RF and associated hardware from the other. I'm not sure that 4x5 Pacemaker Graphic boards are large enough for Chauncey's two lenses plus divider solution to work. Perhaps two small lenses in barrel hung in front of a 4x5 Speed Graphic. Yes, I read that you have Crowns. Have you thought about using a pair of 2x3 Graphics, if you must use Graphics? Have you thought about matching lenses on focal length? Link to comment Share on other sites More sharing options...
rodeo_joe1 Posted September 10, 2017 Share Posted September 10, 2017 For static subjects you could just use a sliding bar to move the camera by the requisite 2.5 to 3 inches between consecutive shots. Or look for an old passport camera that takes 2 or 4 shots on a single sheet. Personally I've always found the popup book / cardboard cutout / diorama look of stereo viewers slightly comical, and most unconvincing of viewing a real scene. Link to comment Share on other sites More sharing options...
chauncey_walden Posted September 11, 2017 Share Posted September 11, 2017 OK, you can try this. A general guideline for an ortho view would be a 1:30 ratio. That is the center of the lenses should be separated by 1/30th of the distance to the nearest subject. Stereo cameras have offset film windows to help eliminate unoverlapped areas of the merged images. Your option would be to slightly toe in the two cameras so that they, like your eyes, are both looking at the subject. Good luck. Link to comment Share on other sites More sharing options...
JDMvW Posted January 10, 2018 Share Posted January 10, 2018 You can also do stereo with a single 4x5 or any size format by shifting the camera the proper distance sideways. This of course won't do for action shots, but for architecture and landscape it can work. Antietam Dunker Church X-eye view stereo made by shifting: 1 Link to comment Share on other sites More sharing options...
rodeo_joe1 Posted January 11, 2018 Share Posted January 11, 2018 Two side-by-side 5x4 cameras could be made to work with a pair of prisms or surface-silvered mirrors forming a small periscope for one of the lenses. Thus bringing the optical axes within 75mm of each other. Pointing the lenses inwards will not work. The required parallax will change with distance and give a 'cross eyed' look to the scene. If a stereo effect could be seen at all, it would quickly give the viewer a headache. Maybe the OP has given up or found a solution by now anyway. Link to comment Share on other sites More sharing options...
chauncey_walden Posted January 11, 2018 Share Posted January 11, 2018 Actually, Joe, parallel lenses without offset image planes (as built in to real stereo cameras) causes more eyestrain than toe in to center the subject. Toe in does introduce keystoning but for most subjects this is not noticeable. Pointing lenses inwards on two cameras or on virtual two cameras (via horizontally shifting) does work. I'm guessing you haven't tried it. Link to comment Share on other sites More sharing options...
rodeo_joe1 Posted January 12, 2018 Share Posted January 12, 2018 But toe-in would cause objects at a distance behind the aimed-at subject to have images displaced in the opposite direction from foreground objects. How can that not look unnatural? Objects at a long distance or infinity should show little or no image displacement at all in a true stereo pair. As you view the (parallel) images, your eyes have to vary in angle to bring various planes of the image(s) into co-incidence, and that's part of what gives the illusion of depth. You only need to artificially alter the inter-ocular distance for very close or very distant subjects. Otherwise the focussing mechanism of a twin lens stereo camera would be very complex indeed, and require a rangefinder-like mechanism. Instead, in most stereo cameras, both lenses are mounted parallel on a single rigid plate. How does that allow for toe-in aiming? Link to comment Share on other sites More sharing options...
chauncey_walden Posted January 15, 2018 Share Posted January 15, 2018 Joe, I just found this. Maybe it will help explain it: vfxio.com/PDFs/Parallel_vs_Converged.pdf Or, hold up a finger at arms length and look at it and notice what happens to subjects in the background. Our eyes toe in to look at any subject. Real stereo cameras have parallel axis lenses but have offset capture windows to reduce the non overlapped subjects. But what do I know, I've only been taking stereo images with stereo, or non stereo one or two cameras for 55 years. 2 Link to comment Share on other sites More sharing options...
GerrySiegel Posted January 15, 2018 Share Posted January 15, 2018 If one wanted to just experiment then a slide bar would be a good place to start and use a 2-1/2 to 3" spacing. I did my first experiments with two box cameras and the spacing was wider than normal. But this is easily tolerated if close up objects are not allowed. Manfrotto makes a bar that is usable. Or look at eBay for some oldie but goodies like the Novoflex original. Or just make one from a couple hunks of wood.. Link to comment Share on other sites More sharing options...
jnanian Posted January 15, 2018 Share Posted January 15, 2018 look for an old polaroid passport or portrait camera all the work is already done for you .. and they cost a lot less than a speed or crown graphic ! Link to comment Share on other sites More sharing options...
rodeo_joe1 Posted January 15, 2018 Share Posted January 15, 2018 Chauncey, the toe-in versus parallel argument is only valid for projected or screen-generated stereo views; where an inter-ocular viewing separation of each image cannot be implemented. Such projections are entirely false, because the screen (or stereo window) has a substantial physical distance from the eye that has to be compensated for. In a traditional stereo print viewer, each eye is separately presented with a field of view, and the stereo window is very close to, or virtually at the plane of the eye(s). Therefore no parallel image compensation is needed. This type of viewing can also be simulated on small or split screens placed close to the eye with focus correction. If you insist on trying to make the unnatural (projected/distant screen) double-image phenomenon look natural, then obviously there will be difficulties to overcome. However, that linked PDF file obviously comes to the conclusion that shifted image parallel projections are the best solution. Link to comment Share on other sites More sharing options...
davidscott Posted January 15, 2018 Share Posted January 15, 2018 Totally agree with Gerry Siegel -- stick with distant subjects and the wider-than-normal separation is acceptable, and may even be preferable. At the risk of seeming like a "miniature", folks exploit wider separations in order to enhance the 3D effect on subjects that would normally all have receded to infinity. Like taking stereo photos of a mountain range from an airplane window, with the movement of the aircraft providing the separation between two exposures. That's separation measured in dozens of metres! I haven't measured my Speed Graphic, but I would guess the shortest distance from the centre of the lens to the edge of the camera body would be downwards. So I would place your two cameras bottom-to-bottom. Of course this will give you two portrait-oriented photos, which isn't a problem for making Holmes-style cards, which use a basically square image for each side of the card. If you haven't already discovered it, I also heartily recommend the freely-available StereoPhoto Maker software. It performs calculations and corrections on stereo pairs and allows output in all of the popular formats, including Holmes cards. The corrections are great, allowing even simplistic stereo cameras like my old Coronet 3D (127 format with meniscus lenses) to make high quality cards. Link to comment Share on other sites More sharing options...
chauncey_walden Posted January 15, 2018 Share Posted January 15, 2018 Slide bars or focusing rails mounted sideways are a great way to work with one camera and static subjects. If the lens axis is perpendicular to the bar you still have the unoverlapped portions of the image to contend with but they can be masked out for slides or cropped off if making prints. In extreme macro work (I have made a lot of 3-D images of insects in amber) I found it easier to move the subject. 1 Link to comment Share on other sites More sharing options...
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now