How are the new line of architectural cameras going to influence your work?

[sanking]

Good point about file size. I have worked a lot over the past several years with faily high resolution scans of 6X7cm and 6X9cm files, with file size in the 60-120 mp range. With this size file I have found it possible to do rather extensive perspective control, quite a bit more than the 1/3 width area Kirk mentions, not surprising perhaps since my file sizes are at least 4X more than what one would get with the best DSLR with file size of 20 mp or so.

My route is probably not a viable professional path because it involves scanning, but for high quality results, when one has the time, it works.

Sandy King

Christopher, I see what you are asking. Again in terms of DSLR (or scanned film) the larger the native file size in relation to the final print size, the less obvious artifacts are generated by stretching. My rule of thumb when teaching students, who do not have shift lenses, depending on the detail in the stretch area, if you have to stretch it more than 1/3 the width of the frame you will probably generate obvious artifacts (but this is related to native file size vs. final print size as mentioned earlier. This is also related to lens quality sharpening etc.). 1/3 is really not much in terms of trying to correct serious perspective problems so this is not really a viable alternative sometimes. I hope this makes some sense.

[Kirk Gittings]

Sandy, I think I learned that from a discussion with you awhile back and tried it myself.

[Michael Mutmansky]

I think there are two factors in this that weigh in Sandy. B&W vs. color could be a considerable impact, as color smear may be much more noticeable than B&W grain smear. I think the tolerance for smear would be much higher in B&W for this reason.

Also the reproduction ratio is a factor as well, but I think that this is implicit in the conversation. It hasn't been explicitly discussed, however.

One other thought is the minimum amount of information needed to express minute details in an image. A line or building corner needs a certain minimum amount of 'pixels' to be expressed properly, and the larger the file, the greater the amount of data that is actually used to express minute details. So when the interpolation (smearing) happens, it is much less noticable in a situation where the data exceeds the minimum amount necessary to express a minute detail.

I think this is functionally the main reason that the file size permits much more manipulation.

[sanking]

Michael,

Your points are well taken, both as regards reproduction ratio and the greater latitude with B&W film compared to color.

In fact, I am aware of someone who did have some trouble with perspective control artifacts in having very large color prints (30X40 inches) made from scans of Mamiya 7 negatives where quite a bit of perspective control was applied. I have personally not seen any such artifacts, either with color or B&W, but I rarely print from MF negatives larger than 20X24.

Sandy

[David A. Goldfarb]

Aside from the issue of perspective control artifacts, I think it would be important that you can only easily correct something like a simple building facade either with Photoshop perspective transformation, which is the equivalent of tilting the enlarging easel, or using rear tilt and/or swing. It's a powerful tool, but it won't fix everything. When a tall structure has surfaces at different distances that need to be square, it's just easier to use front rise.

I had a chance to play a bit with the ArTec, Techno, and the Alpa at Photoplus. They're all fantastic cameras, and the choice among them (and the Cambo Wide DS) would surely be one of individual working preference. Myself, I liked the ruggedness and smooth control of the Techno, and the fact that it's a bellows camera that can handle lenses as long as 210mm without requiring helical mounted lenses. In some ways it's like one of the non-folding Ebony 2x3" cameras, but with the rigidness and precision of a metal camera.

[Henry Ambrose]

I'd love to have such a beast. Camera movements are definitely to be preferred to PS especially in the situations David describes.

But I think I read the new Rodenstock lenses for those cameras are in the neighborhood of $7,000.00 each. That is scary.

[sanking]

Clearly the superior method of doing perspective control is in the camera, whether it be a view camera or another type with shift lenses. However, it is pretty remarkable what one can do with a Photoshop if you have a large file to work with. I actually started doing perspective control in Photoshop on 5X7 negatives that I made with a view camera with limited movements. I was pretty cautious at first but when working with a file of a 5X7 negative scanned at 2400 spi in 16 bit grayscale you have a file of about 115 mp so it is hard to do much damage.

Sandy King

I'd love to have such a beast. Camera movements are definitely to be preferred to PS especially in the situations David describes.

But I think I read the new Rodenstock lenses for those cameras are in the neighborhood of $7,000.00 each. That is scary.

[archivue]

But I think I read the new Rodenstock lenses for those cameras are in the neighborhood of $7,000.00 each. That is scary.

it's much cheaper for arca swiss R...

http://www.galerie-photo.net/contents/fr/d178.html

[Bob Salomon]

I'd love to have such a beast. Camera movements are definitely to be preferred to PS especially in the situations David describes.

But I think I read the new Rodenstock lenses for those cameras are in the neighborhood of $7,000.00 each. That is scary.

No, they will not be in that price range. But the price has not been released yet. As soon as the factory ships the prices will be available.

[Henry Ambrose]

Yes, Sandy I do that as well. Sometimes I miss the precise corrections in camera and then tweak the scan to make it exact. But the thing that is difficult in PS is what David referred to - simultaneous corrections of more than one object in the scene.

The other thing to keep in mind is that, more often than not, commercial files aren't going to be printed really large. I find that clients rarely ask for that. So corrections in PS done on a DSLR file are of no consequence for an 11X14 on the office wall. And of no consequence in small reproductions.

I see a tendency here to think that everything we shoot needs to be able to be printed at 3 or 4 feet tall. Sometimes it might and extra large might be your personal preference but I find in the strictly commercial world little call for such extremes. Its natural that if you're using a large format camera that you'd want to exploit its every capability. I share that yearning. But again, in commercial work there's not so much demand.

Bob, its nice to hear that those new lenses may not be hyper priced. How about you talk them into some great shift lenses for DSLRs?