| Author |
 Replies: 81 / Views: 9,838 |
|
|
|
Valued Member
United States
441 Posts |
Ray,
How many degrees of tilt does your goniometer have and do you ever need more? How's it mounted for use as well as how is your XYZ stage mounted for use?
|
|
Pillar of the Community
2087 Posts |
The one Ray has posted is probably the best bet as far a price goes. as far as Tilt Shift adapters go Kippon produce this: http://www.adorama.com/kasmy645eos.htmlMirex is here but its really way to expensive for just a try out( site is in German), particularly as it is only made to order: http://www.mirex-adapter.de/tilt_shift_adapter.htmI am intrigued with the cheap example Ray has used. the kippon and Mirex approach is to use a medium format lenses on smaller format cameras. This ensures the sensor is covered by the image no matter how much shift or tilt is applied. The Mirex allows for 10 degrees of tilt, and a maximum of 15 mm of shift. At maximum shift and the (approx) 75mm Image circle from the 645 lens still sees some vignetting. I assume the Roxen is only for macro work and that you have to guess how degrees of tilt have been applied Edited by austrokiwi
09/08/2016 12:41 am
|
|
Pillar of the Community
 United States
4038 Posts |
Nice thing about putting the tilt at the camera rather than lens side is there is far less stress on the image circle. In theory, if the tilt were exactly at the sensor, then the image circle effects would be zero. But the adapter I linked to is about as close as you can get to the sensor, so does introduce some shift in the image circle, yet far less (and more predictably) than tilt at the lens end.
Note that because of the distance to from sensor to tilt adapter, the adapter not only introduces tilt, but also shift. The shift is in the correct direction to compensate the distortion produced by tilting the coin. I'm not sure if it's designed that way, but the combined tilt/shift of the "tilt" adapter seems to nearly fully correct both the sensor flatness and distortion at the same time. It's not perfect, as I can still see some distortion when comparing images, but it's not very perceptible.
Andy, I use several types of goniometers. I have some 60x60mm ones that have 7-deg max tilt. The camera tilt adapter has 8-deg max, so they are a good match. I also have a60mmx60mm that has 12-deg of tilt, so can go a bit beyond the camera tilt. Some of the 40x40mm adapters I have go as far as 25-deg, way more than is needed. If you take a look at my website, Example System-3 shows a 40x40mm combo system of goniometer, Z-stage, X-stage, and Y-stage. I believe the goniometer shown goes up to 17-deg.
|
|
Pillar of the Community
2087 Posts |
rmpsrmps: The way you are using tilt is in effect producing exactly the same effect as my use of shift. However getting a shift adapter requires that you think about what lens and which coins you will be imaging.
Effective shift requires a long focal length lens. With the Mamiya 120 at 1-1, shift achieves very little as the close focusing only allows less than 10mm movement. This results in the coin remaining directly under the lens.
Shift, to achieve the same lighting effect works best with long working distances with lenses like the Olympus Zuiko OM 135 Bellows lens (or the mamiya 120) set at low magnification. This also requires a copy stand that allows such working distances.
Edited by austrokiwi
09/09/2016 05:53 am
|
|
Pillar of the Community
United States
4038 Posts |
The methods are similar, but not the same. With pure shift, you distort the image to achieve parallax/perspective shifts. With pure tilt, you affect depth of field. A combo is good for fixing both issues. Here's a simple graphic to show the differences:  |
|
Pillar of the Community
2087 Posts |
Quote:
With pure shift, you distort the image to achieve parallax/perspective shifts. I have used shift numerous occasions, both in coin imaging and in creating large panoramas( also, from time to time, to prevent converging lines when photographing tall structures). I am yet to notice any distortion except where distortion is due to the lens its self( pin-cushion or Barrel) . Could you please expand on this and point out how to identify the shift distortion you are referring to. Edited by austrokiwi
09/09/2016 10:41 am
|
|
Pillar of the Community
United States
4038 Posts |
The whole point of lens shift is to distort the image. The perfect example is architectural photography, where lens shift is used to "straighten" the converging perspective of buildings and such. This straightening effect is intentional image distortion used to compensate perspective so that the building appears to have straight sides.
|
|
Pillar of the Community
2087 Posts |
Quote:
The whole point of lens shift is to distort the image That has not been how I have used it. I use it ( out side of coin photography to correct perspective, and it doesn't involve distortion rather it involves moving the lens so that the sensor can be placed in the correct position to achieve the normal( for the human eye) perspective. Please supply some references as according to my understanding you are describing the use of tilt, not shift. As I understand it Shift allows you to keep the camera sensor parallel to the coin preventing distortion not creating it. Tilt causes distortion so that's why you have to tilt the coin Clearly I don't understand what you are saying: To save confusion please use you diagram to show where shift is causing distortion. I ask as I have never had shift distort a coin image. Please also see this: http://www.cambridgeincolour.com/tu...-lenses1.htmEdited by austrokiwi
09/09/2016 2:34 pm
|
|
Pillar of the Community
United States
4038 Posts |
In the Cambridge article they talk about achieving the shift effect by stretching the image in a trapezoidal fashion. Basically, you stretch the top of the building so that the width at the top is the same as the bottom, so the vertical walls look straight up and down. This trapezoidal stretching is done optically in the shift lens. Any stretching of the image, whether using PS or a tilt lens, is a distortion of the image to achieve the desired perspective effect. Stretching = distortion.
|
|
Pillar of the Community
2087 Posts |
I repeat I have never had shift cause distortion with coins. Let stick to Occams razor: the simplest explanation: The Cambridge article does not use the term distortion. It talks about correcting perspective. I believe creating distortion is your personal way of understanding what is going on and that it is not not a standard photographic point of view. You and I will just have to agree to disagree. {Unless you can provide some clear references that you view is based on.}
Edited by austrokiwi
09/09/2016 2:50 pm
|
|
Pillar of the Community
United States
4038 Posts |
Sure, the Cambridge article calls it "stretching". Stretching one part of an image more than other parts is distortion. It's a different type of distortion than the standard distortions dealt with in lenses, and because it is intentional it's accepted as a photographic technique. But it is still distortion.
By the way, how do you know there is no distortion in your shifted coin photos? With the small movements we make, the distortion may not be very visibly perceptible. To see what's going on, the best way is to overlay a picture taken with shift on top of the same coin and same magnification but without shift. Only then will you clearly see how shift is distorting the coin. Alternatively, you can simply measure the height of the coin compared with its width (in pixels). This only works if the coin is round (or square, I suppose).
|
|
Pillar of the Community
2087 Posts |
I have posted in another forum and have checked in my books and in a number of different places. Tilt and shift are, by the majority( if not all apart from Ray) of sources regarded as means to correct perspective not distort perspective. In my understanding the word distort means, in this context, to pull or twist out of shape.
To take the Cambridge color example. It clearly shows without shift the image will be distorted. Using shift properly will correct that distortion.
I think it is important to make sure that the proper and normal approach is used.
To expand slightly: People wear spectacles to correct vision not distort it. Ray your argument would suggest that it is the other way round. Yes it can said spectacles introduce distortion but it is just not the usual approach to use that perspective and I believe it is very misleading.
Edited by austrokiwi
09/10/2016 01:12 am
|
|
Pillar of the Community
United States
4038 Posts |
|
|
Pillar of the Community
United States
9395 Posts |
Quote:
Tilt and shift are, by the majority( if not all apart from Ray) of sources regarded as means to correct perspective not distort perspective I was always under the impression that a shift lens (without tilt) allowed you to keep the film/sensor parallel to the front of the building, even though you are standing near the bottom. If you use a regular (non-fisheye) lens, the image of the front of the building should have the same shape on the sensor as the building itself. When you have to point the camera upwards, the sensor is no longer parallel to the front of the building, so the usual distortion caused by perspective occurs. The shape of the image is no longer the same as the shape of the building. Wouldn't coins be the same? If you keep the sensor parallel with the coin, wouldn't the image be the same shape as the coin? |
|
Pillar of the Community
 Canada
2784 Posts |
ray your statement applies to your glasses correct.I wear glasses. They focus my vision, but neither cause nor correct any noticeable distortion. I wear glasses as well mind does correct a slight distortion. that is why nobody else can where my glasses. point being people with eye astigmatism . there glasses are corrected for these astigmatism. they correct curvatures.
|
| |
Replies: 81 / Views: 9,838 |
First Page
Previous Page
Posted 09/08/2016 12:34 am
