DP3M and Panorama Head
I am going to manufacture my own panorama head for lightness and compactness sake. I will only be using it for the DP3M, so I can make the bracket without any need for adjustment for parallax changes with different lenses.
Does anyone know the distances (in mm please) I need to move the camera forward/backward of the tripod screw for landscape mode and the distance to the centre of the lens in portrait mode? If not I will make a jury rig to find out. I'm being lazy :-).
If you're shooting at infinity you don't need to make any adjustment for parallax. From my experience parallax compensation is actually only needed for really close shots of around 50' or less. But I'd suggest you experiment to see how close you can shoot without any problems.
In fact for daylight shooting at infinity you may not even need a pano head, or tripod for that matter, if you limit the pano to just the buffer size (7-8 shots).
Most of my panos are handheld and work fine, though I use the SD15 instead of a DP, and have shot up to 30 shots and had them stitch just fine, though admittedly it does take some practice. But you might try it, if you're curious, and see if you can save yourself some hassle.
I plan on using it in some close and some macro work so I think parallax might become a problem.
I must also admit I like the challenge.
For macro work it will definitely be a problem, even with a pano head, as it's difficult to compensate for parallax when you're only a few feet (or inches) away.
In my first attempt at close up stitching, in order to minimize parallax problems, I used a 500mm mirror lens eight feet away. It worked, but was a bit clumsy. In my second attempt I decided to use an xy axis instead (using a 90mm macro at two feet), the x consisting of a slide bar and the y the geared center column of my tripod. This worked much better and is now my preferred method for all my studio work (it obviously wouldn't work in the field due to wind).
Here is a studio shot made from a combination of both pano stitching and focus stacking utilizing the xy axis technique:
You can certainly try making a pano head that will duplicate this, and I'd be curious to see your results if you do, but for really close up work I'd think the xy axis technique would be more accurate.
I determined the so called entrance pupil (i.e. the no-parallax point, incorrectly often called nodal point) of my Sigma DPx Merrill cameras for the hyper-focal setting at f/11. The entrance pupil of the DPx Merrills is different for different focus settings since the front lens of the cameras moves back and forth when focussing. I found the following values for the distance of the entrance pupil to the center of the tripod mounts:
- DP1M: 38mm
- DP2M: 32.5mm
- DP3M: 37mm
The hyper-focal setting at f/11 of the DP1M and DP2M was computed for a very small circle of confusion (0.008mm, slightly more than the size of the diagonal of a pixel), in order to retain most of the sharpness of these cameras. On my camera an easy way to set these two cameras to this hyper-focal setting is to put the camera into manual focussing mode, and move the focus mark beneath the center of the left loop of the infinity sign.
The hyper-focal setting of the DP3M is even further towards infinity, but the movements of the front lens so close to infinity are very small, so the above entrance pupil figure works.
For near-field panoramas, larger offsets need to be used, as the front lens moves forward for smaller focus distances. (This is especially true for the DP3M!)
Hope that helps,
Robert F. Tobler
As someone mentioned, not really needed for landscape mode. I would try some panos without any adjustment just to confirm that. I computed nodal points for three different lenses (50 f2.8, 17-70 and 18-50). I also did the DP1 and DP2. Aside from a few test shots of my kitchen and garage, I never used them but they worked very well for close-in panos. You can google "Panosaurus instructions" and the document (for setup of Panosaurus head) includes directions for nodal point computation. BTW the nodal points for all 3 at 50mm were different. The Panosaurus instructions mentioned that this might be the case.
Here is a recent 3-shot pano horizontal hand-held at 17mm from the 17-70C :
I used the Panosaurus instructions to find the nodal points. I did several lenses and cameras (DP1/2) including the 50mm f2.8, 17-70 older and 18-50 f2.8 but I doubt the numbers are transferable for the DP3M. At 50mm each had different settings.
If you want to check out the instructions, just google "panosaurus instructions". They are part of this setup document and it is quite easy to do. I did two tests i- in my kitchen and in my garage and the images looked very good. I never used the Panosaurus after that because, as was mentioned, it isn't needed for infinity shots.
My measurements are similar to yours (also measured from the center of the tripod mount), differing essentially with the DP3M.
I measured stopped down to f11 at infinity focus, using a vertical laser, checking for lack of movement of the vertical beam on the diaphragm vs the opening, as the camera was rotated about the entrance pupil. (I used a 30 sec exposure to have time to rotate the camera back and forth.)
DP1M = 38mm
DP2M = 32mm
DP3M = 36mm
Well thanks everyone.
I just learned that the job is more difficult than anticipated.
I assume Robert that you mean that since the front lens moves back and forth the parallax point of the DP3M is 37 mm at f11 at one focal distance only, and that is the hyper-focal setting in this case?
Mike's suggestion of an x axis is interesting. It would save me a bit of work. It might work in the field as macro is often close to the ground and often fairly protected.
The Panosaurus instructions are helpful. Thanks Rick.
|DSC_9643 by NOWHITELENS|
from Best Photo of the Week
|Thailand Sunrise by ozziebadger|
from Ships and Boats