Pages: [1] 2   Go Down

Author Topic: TSE-24 a few thoughts.  (Read 9469 times)

Shakyphoto (Slim)

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 256
    • Shakyphoto
TSE-24 a few thoughts.
« on: June 04, 2013, 11:23:34 am »

Tried out the TSE-24 on my recent trip to Scotland.
I haven't gotten great pictures with it... yet.  I'm still acclimating myself to it.

Couple of early things I noted.

I didn't realize how short the "throw" is from 5 ft to infinity focus.  I screwed up a bunch of times on this one.
When I tried to use the tilt feature to get better depth of focus, I've gotten worse results.

Here is an image I took of our stay at Invergarry Castle in Scotland.  The shift feature is easy to use as you can see here.  All the lines straight up and down.

Logged
Wandering the World, Watching for Wonders
http://www.shakyphoto.com
http://www.fb.com/Shakyphoto

Jim Kasson

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2370
    • The Last Word
Re: TSE-24 a few thoughts.
« Reply #1 on: June 04, 2013, 11:36:53 am »

When I tried to use the tilt feature to get better depth of focus, I've gotten worse results.

If you're just starting out with a tilt/shift lens, the Summerhayes technique is a good place to start.

The key measurement you need to use the Summerhayes method is the perpendicular distance from the lens to the plane that you want in focus. In landscape photography, that plane is usually the ground, and thus all you need to know is how high the camera is. If you’re using the camera at eye height, and you’re of anywhere near average size (male or female – given the accuracy to which you can set a tilt/shift lens tilt angle, it doesn’t make any difference), the angle you want to set your tilt to is one degree for a 24 mm lens, 1 ½ degrees for a 45 mm lens, and 3 degrees for a 90 mm lens.

Jim
« Last Edit: June 04, 2013, 11:39:32 am by Jim Kasson »
Logged

Ellis Vener

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2151
    • http://www.ellisvener.com
Re: TSE-24 a few thoughts.
« Reply #2 on: June 04, 2013, 11:53:10 am »

Tried out the TSE-24 on my recent trip to Scotland.
...

1) I didn't realize how short the "throw" is from 5 ft to infinity focus.  I screwed up a bunch of times on this one.
2) When I tried to use the tilt feature to get better depth of focus, I've gotten worse results.

1) Using Live View "zoomed in" (i.e., at  5x or 10 x magnification) plus a loupe like a  Hoodloupe (http://www.hoodmanusa.com/products.asp?dept=1017) is the best way to very finely establish prime focus, especially with a TS-E lens where the optical  light path is off axis to the groundglass in the viewfinder.

2) The tilt function of a  Tilt/Shift lens basically turns the camera into a miniature front tilt or swing + rise/fall/shift only view camera. Certain principles apply: if you use tilt,  focus the point that the lens (which if you have also used shift is not in the center of the framed composition) is centered on   and then start dialing in the tilt to bring the  top and bottom  of the subject (or the sides  if you are using the tilt function as swing  into focus. You will most likely will then need to  focus a little ways deeper into the scene and stop down  to bring everything into satisfactory focus.  This is how you apply Scheimpflug's principle.

Schiempflug's Principle is a geometric rule that says when a plane through the subject, the plane of the lens, and the film/sensor plane all intersect in a line every point in the chosen subject plane will be in focus at all apertures. By refocusing after you have applied tilt yo uare choosing a different subject plane to focus on than the one you initially focused on. As with all lenses, closing down to a smaller aperture extends the depth of field in front of and behind that subject plane.  Here's an illustration of what is going on http://en.wikipedia.org/wiki/File:Scheimpflug.gif

I only use the tilt function of a TS-E lens when it becomes either absolutely necessary or to create a definite anti-Scheimpflug principle effect.



  
« Last Edit: June 04, 2013, 12:02:07 pm by Ellis Vener »
Logged

Bart_van_der_Wolf

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 8913
Re: TSE-24 a few thoughts.
« Reply #3 on: June 04, 2013, 01:10:28 pm »

1) Using Live View "zoomed in" (i.e., at  5x or 10 x magnification) plus a loupe like a  Hoodloupe (http://www.hoodmanusa.com/products.asp?dept=1017) is the best way to very finely establish prime focus, especially with a TS-E lens where the optical  light path is off axis to the groundglass in the viewfinder.

Hi Ellis,

Yes, Live view with a hooded loupe is indispensable for accurate tilted focusing, the focus confirmation light/beep works fine for modest tilts.

For viewfinder focusing, despite producing a darker viewfinder on lenses with smaller apertures than f/2.8, I use a more detailed Ec-S (Super Precision Matte) focusing screen in my camera.

Cheers,
Bart
Logged
== If you do what you did, you'll get what you got. ==

Shakyphoto (Slim)

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 256
    • Shakyphoto
Re: TSE-24 a few thoughts.
« Reply #4 on: June 05, 2013, 11:49:02 am »

Thanks to everyone for pointing me in the right direction on this one.  I'll have to read up on this.
Little bit disappointed that I wasn't able to do the reading and test out the lens before my trip.
Logged
Wandering the World, Watching for Wonders
http://www.shakyphoto.com
http://www.fb.com/Shakyphoto

Ellis Vener

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2151
    • http://www.ellisvener.com
Re: TSE-24 a few thoughts.
« Reply #5 on: June 05, 2013, 12:42:19 pm »

Another way to extend depth of field, even with large subjects like landscapes and buildings is "focus stacking". There are several programs that can do the processing for you - I use Helicon Focus pro - but the basic idea is the same. Fix your exposure parameters - shutter speed, aperture &  ISO (White Balance too if you are not shootign raw) - to the same settings and focus the camera at different distances moving from close to far. if yo uare usign a manual focus lens like the TS-E series you'll have to do this by guess work but try to move it in small even increments. If using an autofocus lens and have an iPhone, iPad, a Mac running OS X 10.8,  and  soon Android or Windows mobile device handy my favorite way to automate creating the stack is to use the focus stacking function of the CamRanger - http://www.camranger.com
Logged

sdwilsonsct

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3296
Re: TSE-24 a few thoughts.
« Reply #6 on: June 06, 2013, 09:44:25 am »

Another way to extend depth of field, even with large subjects like landscapes and buildings is "focus stacking".

Much cheaper than a T/S lens. Problematic if there's wind.

Shakyphoto (Slim)

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 256
    • Shakyphoto
Re: TSE-24 a few thoughts.
« Reply #7 on: June 06, 2013, 11:35:43 am »

Much cheaper than a T/S lens. Problematic if there's wind.

I have yet to attempt to focus stacking as a technique.  However I prefer to use tools in the field rather than rely on post processing of multiple images.  Even though tools in the field may be more expensive than software, the real expense would be if I took the shots and found out later that the images don't line up or match up and can't do anything about it because my flight home is the next day.  A single plane ticket across an ocean is the price of a good lens.
Logged
Wandering the World, Watching for Wonders
http://www.shakyphoto.com
http://www.fb.com/Shakyphoto

Ellis Vener

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2151
    • http://www.ellisvener.com
Re: TSE-24 a few thoughts.
« Reply #8 on: June 06, 2013, 11:51:24 am »

Much cheaper than a T/S lens. Problematic if there's wind.
True but if you can do it with a minimum number of frames - and the camera is not moved by the wind and do some work on masking the layers, it can work. When creating stitched panoramas i also often have the same  problem (trees, flags and water moving due to wind and current; but also  people and vehicles moving) and solve it by editing the layers that make up the composite using masks on each layer. it is tedious but doable.  But admittedly sometimes not even that works.
Logged

sdwilsonsct

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3296
Re: TSE-24 a few thoughts.
« Reply #9 on: June 06, 2013, 11:56:41 am »

if you can do it with a minimum number of frames - and the camera is not moved by the wind and do some work on masking the layers, it can work

Good point. I shoot a lot of vegetation, so I gave up on Helicon and am trying a T/S.

Thanks for your clear explanation of focusing technique above.

Ellis Vener

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2151
    • http://www.ellisvener.com
Re: TSE-24 a few thoughts.
« Reply #10 on: June 06, 2013, 11:58:30 am »

I have yet to attempt to focus stacking as a technique.  However I prefer to use tools in the field rather than rely on post processing of multiple images.  Even though tools in the field may be more expensive than software, the real expense would be if I took the shots and found out later that the images don't line up or match up and can't do anything about it because my flight home is the next day.  A single plane ticket across an ocean is the price of a good lens.
I agree aboutit being better to solve a technical problem at the time the photographs are being made, and about the expenses of making a miscalculation, plus time and travel relative to the cost of a good lens or a solid tripod/head combination and cable release.

I'd much rather solve a problem when making the photos but I'm glad to have the "after the fact" tools available as well. While software maybe relatively cheap compared to a good lens or tripod/head combination, there is the cost of the time needed to learn and to apply any processing and post-processing tool.
« Last Edit: June 06, 2013, 12:03:54 pm by Ellis Vener »
Logged

CptZar

  • Full Member
  • ***
  • Offline Offline
  • Posts: 157
Re: TSE-24 a few thoughts.
« Reply #11 on: June 06, 2013, 03:43:30 pm »

Right, but honestly, if you take a larger FL than 24mm tilting really gets difficult. Further on, it is not possible to use tilting with large objects close to the  lens, due to the angle of the focus plane. So I do assume it is much easier, with even better results, to use focus stacking.

With focus peeking things change again. But Canon or Nikon don't provide that. Only Sony today. So actually if you are really serious about TS you should use a Sony NEX7 with a metabones speedbooster (to get the angel of view of FF). Then you have an EVF and tilting becomes very easy. However, the NEX7 is a haptical nightmare. So as long as Canon an Nikon do not provide more than a viewfinder from the last century tilting is a PITA with those cameras.

Magical Latern has focus peeking as well, but it doesn't work very good.

Anyway, the time you need to get a sharp image with a 50mm lens and tilting is most probably longer the one need for focus staking. 24mm still works OK though.

It's not about when the problem is solved, it's about how it is solved best. If you go home, and the stone which you put in the very front of your image is only partly sharp, you will wish you did not try to make it all right on location.

But yes, with moving objects tilting is the only way to go.

cheers

Jan
5DMIII
« Last Edit: June 06, 2013, 03:45:38 pm by CptZar »
Logged

Glenn NK

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 313
Re: TSE-24 a few thoughts.
« Reply #12 on: June 06, 2013, 07:00:34 pm »

For a FLAT surface (not necessarily horizontal, but flat), I use the following process:

1)  frame the image,

2)  focus at infinity (or most distant part of image) using focus ring,

3)  zoom in maximum in live view,

4)  using the TILT knob, adjust the foreground into focus,

5)  repeat steps 2) to 4) as required - I usually repeat only once.

I've often shot wide open (f/3.5) with good results.

I completely gave up on charts etc. soon after I learned the above (two years ago).

Glenn
Logged
Economics:  the study of achieving infinite growth with finite resources

Ellis Vener

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2151
    • http://www.ellisvener.com
Re: TSE-24 a few thoughts.
« Reply #13 on: June 07, 2013, 01:14:30 pm »

For a FLAT surface (not necessarily horizontal, but flat), I use the following process:

1)  frame the image,

2)  focus at infinity (or most distant part of image) using focus ring,

3)  zoom in maximum in live view,

4)  using the TILT knob, adjust the foreground into focus,

5)  repeat steps 2) to 4) as required - I usually repeat only once.



I've often shot wide open (f/3.5) with good results.

I completely gave up on charts etc. soon after I learned the above (two years ago).

Glenn

Glenn,

I am surprised, given that the lens is a mount that pivots the lens around its optical axis (that's the line that runs through the center of the lens), that you don't focus in the center of the frame and then dial in the amount of tilt neded to bring the near and far points of that plane into focus.

That is how I use TS-E lenses lenses. Am I missing something?
Logged

Glenn NK

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 313
Re: TSE-24 a few thoughts.
« Reply #14 on: June 08, 2013, 02:09:49 am »

Glenn,

I am surprised, given that the lens is a mount that pivots the lens around its optical axis (that's the line that runs through the center of the lens), that you don't focus in the center of the frame and then dial in the amount of tilt neded to bring the near and far points of that plane into focus.

That is how I use TS-E lenses lenses. Am I missing something?

Ellis:

To be truthful, I don't know, but I do know that some pretty darn good photographers use the method.

Royce Rowland on Naturescapes.net told me about it on a thread a couple of years ago.

Also a good post by Darren K (14 June 2012), which alludes to what you said about focusing in the center of the frame:

http://www.naturescapes.net/forums/viewtopic.php?f=1&t=218217

Glenn
Logged
Economics:  the study of achieving infinite growth with finite resources

Ellis Vener

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2151
    • http://www.ellisvener.com
Re: TSE-24 a few thoughts.
« Reply #15 on: June 08, 2013, 12:12:57 pm »

Glenn: Thank you for that input.  As I come from a background of using view cameras (primarily Sinar P & C and Arca-Swiss F line and Canham DLC) which base tilt designs ( The big Sinars use what Sinar rightly calls their system "off axis tilt" but it is a variation of a base tilt design) the "focus on the far then tilt to the near" approach is one I am very familiar with  but as the TS-E lenses are a tilt on the lens axis design  it might work faster  to try focusing in the middle of the frame and then tilting to find the right angle to bring both the near and far points into focus.
Logged

Glenn NK

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 313
Re: TSE-24 a few thoughts.
« Reply #16 on: June 08, 2013, 12:38:42 pm »

Ellis:

I will probably try your suggestion, but it seem that two or three iterations do the trick with the other method (infinity first).

Glenn
Logged
Economics:  the study of achieving infinite growth with finite resources

Bart_van_der_Wolf

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 8913
Re: TSE-24 a few thoughts.
« Reply #17 on: June 08, 2013, 01:37:00 pm »

[...] but as the TS-E lenses are a tilt on the lens axis design  it might work faster  to try focusing in the middle of the frame and then tilting to find the right angle to bring both the near and far points into focus.

Hi Ellis,

That depends on the particular mechanical lens design.

When I tilt forward with an TS-E 24mm f/3.5 II, I often need to adjust my initial center-focus towards a shorter focusing distance. So maybe it is easier to focus slightly more to the foreground, then tilt, because it will stay closer to what I already had and I can visually judge how much tilt is required for the distance in the upper part of the image. My TS-E 90mm f/2.8 uses a significantly different pivot point, and comes slightly closer to 1/3rd of the distance as a better starting point.

Initial focusing in the distance with those lenses is clearly a less efficient approach. The lesson for Glenn is to not listen to Guru's too much, but do the test yourself.

BTW, it is easy to find out if front/middle/back-ground focusing is the better starting-point for any given Tilt lens.
Just spend some time iterating between foreground and background tilt & focus adjustments until 'the whole enchilada' is in focus. Now reset the tilt to zero, and see where the focal plane is located in the distance. Voila!

Of course that only keeps the ground plane in focus, which is not wanted if there are high structures in the background. One should get focus at halfway up the distant heights if DOF needs to be optimized, so tilt needs to be reduced a bit (and the initial focus point gets a bit closer to the middle distance setting again). So, focusing a bit shy of the middle range, seems more efficient with these lenses.

Cheers,
Bart
« Last Edit: June 08, 2013, 01:49:58 pm by BartvanderWolf »
Logged
== If you do what you did, you'll get what you got. ==

Glenn NK

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 313
Re: TSE-24 a few thoughts.
« Reply #18 on: June 08, 2013, 03:01:46 pm »

In any event, it seems that the process is still iterative.

Glenn
Logged
Economics:  the study of achieving infinite growth with finite resources

Bart_van_der_Wolf

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 8913
Re: TSE-24 a few thoughts.
« Reply #19 on: June 08, 2013, 04:36:26 pm »

In any event, it seems that the process is still iterative.

Hi Glenn,

That's correct, if you want to shoot something else than just the ground-plane in focus. Remember that the focal plane moves up and down as well as closer and further away, after dialing in some tilt. So one iteratively searches for the optimal height in the foreground and background as well. Live View (don't forget a loupe) is heaven-sent for that ...

Cheers,
Bart
Logged
== If you do what you did, you'll get what you got. ==
Pages: [1] 2   Go Up