Images from a view camera with Canon 5DMII back

[pinetar]

http://gallery.me.com/nam65#100793

Trying a variety of old lenses. Images tuned in Apple Aperture. No sun last sunday, each lense is very long, seems to work best on distance shots but the soft focus is evident, which is what I was trying to achieve.

Gary

[goamules]

Very nice shots. I don't know what a 5DMII back is, I assume digital. But it shows a good photographer can always make the shot, even mixing old with new.

[Gene McCluney]

A Canon 5D mk II, is a Canon 35mm format digital SLR of 21 mpx. I assume the OP adapted the camera body to the back of his view camera, thus capturing a 35mm area image from the lenses on the view camera.

[dh003i]

I wonder how the sharpness of those photos -- small crops of the image-circle of the LF lens -- compares with the sharpness of a normal Canon lens for the 5D.

Any comparisons?

[pinetar]

Image 12 to the last were shot on the Canon 5DM, manual focus with a Nikon 50 mm f1.4. Bad day, no sun, just nice people and good old tyme music.

The rig and some of the lenses used: http://gallery.me.com/nam65#100496

[Doug Dolde]

Honestly I don't see anything here to write home about.

[rdenney]

The value seems to be in making use of some very old and interesting lenses, though their special qualities might be subverted by using such a small portion of their image circle. I have adapted a range of small and medium-format lenses to my 5D to make similar things possible, but in the end I usually get better shots using high-quality lenses intended for the 35mm format. For example, as big a fan as I am of the Carl Zeiss Jena 180/2.8 Sonnar, it really is no match for the old Nikkor ED AIS 180/2.8 when used on the Canon. Both are classics of their kind, but the Sonnar will cover 6x6, so on medium format it fairly blows the Nikkor away. And the newish Canon 70-200/4 zoom actually is not embarrassed in comparison to the Nikkor.

Where it might be interesting to mount my 5D to the back of a view camera is for completely static subjects, where the digital camera body is moved around using shifts to build an image for later stitching. That restores some of the format though I wonder about the sensor being shadowed by the mirror box and other issues. And, of course, few subjects are really that static in the landscape world. I would think a medium-format back with the sensor mounted close to the mounting plane would work better--no shadowing and fewer component images in a stitch. But it's just a theory--I can't afford one. If a shifting adapter comes my way cheaply enough, however, I might experiment. I already have a 5D.

Rick "not that into the old lens look" Denney

[Greg Lockrey]

If all you want to do is have "that old lens look" you could just as easily stretch some celophane over a stock Canon lens and shoot through that. If you are looking to stitch multible images together it is far easier and superior just to use a pano head with the much better Canon lenses and have virtually no limit to exposure area. Best useful area using this 5D back/large format camera technique I was able to get was about a 3x3" because of the mirror box casting a shadow. You can now do it on the cheap without a pan head with the variety of pano stitching programs available including what is in Photoshop CS and later. Just seems like a waste of effort for the end result to me. :confused:

[bobwysiwyg]

Perhaps he was just challenged to try it. Why not? Lots of folks experiment with various alt. methods of imaging.

[dh003i]

It is interesting to note regarding sharpness, that one of my lenses for my E3, the Olympus 50/2 macro, tests at about 123 lp/mm at f/4. That's an aerial number. This is considered an exceptionally sharp lens, one of the sharpest ever made.

Yet that resolution is still only about 50 lp/mm better than the Nikkor-SW 90/8 or 90/4.5 at f/22 (from back-calculated aerial resolution numbers).

Theoretically, all systems would be equivalent in terms of aerial resolution at least if lenses are perfect, diffraction-limited. From resolution tests I've looked at, LF lenses are much closer to being diffraction-limited than even the best 35mm-format and smaller (i.e., 4/3rds) lenses.

[Frank Petronio]

Why don't you just buy a lensbaby for your dslr?

[jb7]

I sometimes have a look at lens performance on my dslr attached to the Arca-
it's quicker and less costly than shooting film, especially colour.

Forgive me if I'm wrong,
but I'd imagine that the same curiosity was driving these images?

It's interesting (and a quick way) to see how the lenses perform,
and can also give useful exposure feedback for unmarked or experimental optics.

I certainly wouldn't be as quick to dismiss it outright...

[Kurt M.]

one thing about mtf (if shooting color film)...
most film mtf crashes significantly when you go over 50-60 lpmm...so, even if you had a lens that can resolve 100-200, you likely won't see it on the film...
a large format lens can resolve 50-60lpmm, so you are just above that treshold where the film would stop resolving. As you are enlarging significantly less (At 4x5) than with a 35mm frame, you would see a lot 'sharper' images...

[pinetar]

The images posted are from my second attempt at understanding my LF lenses. Since the first shots, I was able to cancel light leaks. The images in no way represent a showcase to be admired. Using this rig is just fun.

I don't see it as habit forming. My Contax/Zeiss lenses keep me happy.

[dh003i]

one thing about mtf (if shooting color film)...
most film mtf crashes significantly when you go over 50-60 lpmm...so, even if you had a lens that can resolve 100-200, you likely won't see it on the film...
a large format lens can resolve 50-60lpmm, so you are just above that treshold where the film would stop resolving. As you are enlarging significantly less (At 4x5) than with a 35mm frame, you would see a lot 'sharper' images...

This is interesting in light of Q.-Tuan Luong's comment on the 1/R approximation breaking down for many large-format uses:


Several sources use the formula 1/r_final = 1/r_lens + 1/r_film to compute the final resolution. This formula is an approximation to the exact calculation consisting of the convolution of the response of the film and the response of the lens. This approximation is most valid when both the film and lens are being used near their resolution limits (spatial frequencies with very low contrast). This corresponds roughly to f-stops up to f16. Thus the formula is pretty good for 35mm and MF work. However, when the frequencies involved are nowhere near the film limits, the formula is a poor approximation which predits a worse 1/r_final than what you actual get. For f-stops of f32 and higher, what you get on film is in fact practically equal to the aerial resolution, and the formula shouldn't be used in that case.

The other thing I'd read is that film's ability to capture 40-50 lp/mm is virtually 100%, while it's much less at higher lp/mm where contrast approaches 0 in MTF charts. Hence, another advantage for large-format.