Ad
Ad
Ad
Pages: [1]   Bottom of Page
Print
Author Topic: Help needed for special Tipod mount for TS-E 24mm 3.5L II  (Read 752 times)
Christian Arp-Hansen
Newbie
*
Offline Offline

Posts: 12


« on: February 12, 2013, 10:51:14 AM »
ReplyReply

HI
I need a special tripod mounting solution for my 24 ST-E that fixes the front of the lens to the tripod so it does not move with tilt and shift. This will give me the ability to stitch multiple shifted frames without parallax errors.

Does anybody found a usable/nice/sturdy way to do this?
« Last Edit: February 12, 2013, 10:54:57 AM by Christian Arp-Hansen » Logged
uaiomex
Sr. Member
****
Offline Offline

Posts: 934


WWW
« Reply #1 on: February 12, 2013, 11:24:22 AM »
ReplyReply

http://www.hartblei.de/en/canon-tse-collar.htm
I find it too cumbersome and expensive so I keep on using a slide ruler mounted between the camera and the tripod head to move the camera back to original position.
Eduardo
Logged
Christian Arp-Hansen
Newbie
*
Offline Offline

Posts: 12


« Reply #2 on: February 12, 2013, 11:52:28 AM »
ReplyReply

Thanks
Just what I needed. I looks nice, but it is fairly expensive.
 
Logged
MrSmith
Sr. Member
****
Offline Offline

Posts: 744



WWW
« Reply #3 on: February 12, 2013, 01:13:03 PM »
ReplyReply

Have you tried a plate and clamp with mm markings? I use a giottos one that fits the usual arca type plate. I just shift the camera in the opposite direction the same amount of mm. I only do landscape stitching so you would need an L-bracket if you want the camera portrait.
Logged
francois
Sr. Member
****
Offline Offline

Posts: 6443


« Reply #4 on: February 13, 2013, 04:12:41 AM »
ReplyReply

Have you tried a plate and clamp with mm markings? I use a giottos one that fits the usual arca type plate. I just shift the camera in the opposite direction the same amount of mm. I only do landscape stitching so you would need an L-bracket if you want the camera portrait.

In another thread, Bart shows his setup. It's even more foolproof than simple markings.
Logged

Francois
BartvanderWolf
Sr. Member
****
Offline Offline

Posts: 3019


« Reply #5 on: February 13, 2013, 06:14:17 AM »
ReplyReply

In another thread, Bart shows his setup. It's even more foolproof than simple markings.

Hi Francois,

I (obviously) agree, but it only allows to correct in one direction (which it does very well). It offers a small package for light travel if one occasionally could use the additional field of view but doesn't want to use a full pano setup all the time. Of course one could add a similar construction for the orthogonal direction, but that would still not cover the other possible rotations.

I have some reservations, besides its price, with regards to the Hartblei solution or similar ones. I'm not sure how well the lens will support the heavier DSLR bodies, such as my 1Ds3. And of course, the final image will only be as good as the resolution of the edge of the image circle. The TS-E 24mm 3.5 II is quite good, but requires stopping down to f/11 or f/16 to increase in the edge quality, while losing center resolution to diffraction at the same time.

Here is an image that shows the type of reduced quality near the edges, using the single (shift camera right and left opposite to the lens) bar and stops:

Click on the image for a full resolution (11MB !) version

IMHO, much higher quality can be achieved by using the center of the image circle with a rotational stitching method. It also allows to use the lens at f/4 - f/5.6 which offers much faster shutter speeds (may be necessary to avoid vibrations, e.g. when shooting from a bridge with heavy traffic), and even higher resolution. It also allows more creative use of the tilt functionality.

Cheers,
Bart
« Last Edit: February 13, 2013, 06:20:01 AM by BartvanderWolf » Logged
MrSmith
Sr. Member
****
Offline Offline

Posts: 744



WWW
« Reply #6 on: February 13, 2013, 06:42:15 AM »
ReplyReply

If you use a nodal slide instead of shifting to do a 3 shot stitch what are the differences spatially/geometrically between the 2 methods? Do you get different distortion/curvature?

I too wonded about supporting the weight of a camera by the shift mechanism that was designed to only support the weight of the front elements, I would presume it would accelerate any wear/slop in the gearing.

Logged
francois
Sr. Member
****
Offline Offline

Posts: 6443


« Reply #7 on: February 13, 2013, 07:31:04 AM »
ReplyReply

Hi Francois,

I (obviously) agree, but it only allows to correct in one direction (which it does very well). It offers a small package for light travel if one occasionally could use the additional field of view but doesn't want to use a full pano setup all the time. Of course one could add a similar construction for the orthogonal direction, but that would still not cover the other possible rotations.


Hi,
That's how I understood your setup. I use a similar one when I'm traveling by foot or hiking in the mountains.
Logged

Francois
BartvanderWolf
Sr. Member
****
Offline Offline

Posts: 3019


« Reply #8 on: February 13, 2013, 08:55:51 AM »
ReplyReply

If you use a nodal slide instead of shifting to do a 3 shot stitch what are the differences spatially/geometrically between the 2 methods? Do you get different distortion/curvature?

Hi,

The projection 'distortion' for the same FOV will be the same, if a rectilinear projection is used. Any lens distortion will be removed in the rotational stitch, as part of the stitching procedure, and only the center of the image circle is used. The edges will have similar resolution, minus the lens edge distortion in the rotational stitch, but plus some interpolation losses.

All-in-all, probably pretty similar but with better 'quality' edge blur, and higher center quality if a wider aperture is used. But there are many other differences, such as the semi-circular focal plane, and different projection methods which can span wider fields of view.

Cheers,
Bart
Logged
Ellis Vener
Sr. Member
****
Offline Offline

Posts: 1641



WWW
« Reply #9 on: February 13, 2013, 10:12:26 AM »
ReplyReply

I looked into a special device but then decided to simply try  shifting the lens  and stitching to see how bad the results might be attached is an example. It was shot with a TS-E 17mm f/4L on a 1Ds Mark III body and stitched in PTGui Pro. There were a few parallax errors to resolve but they were pretty simple to handle by working with masked layers in a blended + layers master.  As you can see there are two errors in the extreme lower edge were not resolvable but the near / far relationships there were pretty extreme.

The RRS rig Bart uses works very well too but only for lateral shifts.
« Last Edit: February 13, 2013, 10:14:29 AM by Ellis Vener » Logged

Ellis Vener
http://www.ellisvener.com
Creating photographs for advertising, corporate and industrial clients since 1984.
Pages: [1]   Top of Page
Print
Jump to:  

Ad
Ad
Ad