Home · Register · Join Upload & Sell

Moderated by: Fred Miranda

New fredmiranda.com Mobile Site

New Feature: SMS Notification alert

New Feature: Buy & Sell Watchlist

FM Forums | Canon Forum | Join Upload & Sell

end

Archive 2016 · TS-E Lens Vs. Pano Bracket

jcolwell
Online
• • • • • • •
Upload & Sell: On

p.2 #1 · p.2 #1 · TS-E Lens Vs. Pano Bracket

Tareq wrote:
...Also if that 35 and 50 are more suitable for pano then there is 45TS as well, isn't this a better option then?

I use a 35mm or 50mm (or longer) lens for panos when a two- or three-image shift-pano is not sufficient. I have T-S lenses at 17mm, 24mm, 35mm, 55mm, 80mm, 120mm, and 150mm, and so I have no shortage of options. OTOH, I used a Zeiss MP 50/2 ZE on leveled pano head to take a 14-image panorama that was printed at 16m x 3m. It's possible that a three-image shift pano made with the 17/4L would have covered about the same territory, or a two-image rotation pano using a 11-22mm lens (which I don't own), but there would not have been nearly enough detail for the final print size. I used the 50mm MP lens with fourteen images rather than a 35mm lens with maybe ten images, or 28mm with eight, because that 50 is very sharp and has very little distortion.

Tareq wrote:
...Last point, if i level the camera and the lens so i can eliminate that parallax as much as i an even if not perfectly, will this help to have Pano with shift lens and correct it later if that distortion isn't much noticeable anyway? I mean if i managed to point the lens exactly perpendicular to front scene so there is no parallax and the scene frame as i can see is god enough, where is the issue then? I thought that parallax happen when you shift the camera itself not the lens plane, isn't that shift an advantage...Show more →

I think you're confusing perspective distortion with parallax error. Perspective distortion is introduced when you don't have the lens leveled properly, or especially when you point it up (rising parallel lines rapidly converge), or down (rising parallel lines rapidly diverge). Parallax error is caused when you rotate the lens about a point that's not through its "nodal point". For example, if you take a photo with a fence post in the near foreground (say a couple of metres away), with the fence post in the middle of the image, and then rotate the camera thirty degrees to the left, the post will now be right-of-centre in the next image; unless you rotated the lens through its nodal point, which is what a "pano rig" does for you. Parallax error can also be introduced in a shift-pano when you move the lens relative to the ground, in which case, that fence post is in a different position (relative to the centre of the sensor) for each photo. This difference in relative position for uncompensated rotated or shifted panos is parallax error.

Jan 15, 2016 at 12:24 PM

Tareq
Offline
• • •
Upload & Sell: Off

p.2 #2 · p.2 #2 · TS-E Lens Vs. Pano Bracket

Ah ok.

From where did you get those TS lenses of 35 and 55 and 80 and 120?

Jan 15, 2016 at 03:28 PM

Jeff
Offline
• • • • • •
Upload & Sell: On

p.2 #3 · p.2 #3 · TS-E Lens Vs. Pano Bracket

jcolwell wrote:
For example, if you take a photo with a fence post in the near foreground (say a couple of metres away), with the fence post in the middle of the image, and then rotate the camera thirty degrees to the left, the post will now be right-of-centre in the next image; unless you rotated the lens through its nodal point, which is what a "pano rig" does for you.

I know you've got the concept right, but just for clarification it might help to add that in your example the fence post shifts position in relation to something *behind* it (or in front of it), not just within the composition (which goes without saying). Perhaps I didn't read it carefully enough...

Jan 16, 2016 at 10:50 AM

Gunzorro
Offline
• • • • • •
Upload & Sell: On

p.2 #4 · p.2 #4 · TS-E Lens Vs. Pano Bracket

Tareq -- I don't know about Jim's example with the 50mm lens, but very often longer focal lengths are used in the vertical or "portrait" format, with the longest side in the up/down position. The results in more segments to stitch together, leading to less misalignment and less errors for subject that are mostly at the same distance.

Personally, I don't like doing this vertical format with lenses much wider than 24mm, as the compound distortion "bulginess" shown in Jim's examples becomes very pronounced and errors occur (depends somewhat on software used).

rainshadow -- Nice hand held pano!

I love the freedom of shooting pano's without a tripod.

With today's high MP cameras, we are really getting close to a one body/one lens scenario -- cropping for "tele", and composite pano for UWA. Since composite/pano combines many lower MP images into one large high MP image, even cheaper older camera can be effective. DR is one area that is most benefited by the newer cameras. I would love a FF P&S with an "L" quality zoom lens as my take-everywhere camera. The G1X comes pretty close, but it's still a sub-APS-C sensor. But pretty impressive what can be gotten out of it!

Edited on Jan 16, 2016 at 12:16 PM · View previous versions

Jan 16, 2016 at 12:11 PM

jcolwell
Online
• • • • • • •
Upload & Sell: On

p.2 #5 · p.2 #5 · TS-E Lens Vs. Pano Bracket

jcolwell wrote:
For example, if you take a photo with a fence post in the near foreground (say a couple of metres away), with the fence post in the middle of the image, and then rotate the camera thirty degrees to the left, the post will now be right-of-centre in the next image; unless you rotated the lens through its nodal point, which is what a "pano rig" does for you.

Jeff wrote:
I know you've got the concept right, but just for clarification it might help to add that in your example the fence post shifts position in relation to something *behind* it (or in front of it), not just within the composition (which goes without saying). Perhaps I didn't read it carefully enough...

You're quite right. The apparent shift of the foreground that's displaced by the rotation (or shift) with respect to the background, which shows no apparent shift, is the problem with animage affected by parallax error. I was simply focusing on the way the foreground is changed when you rotate the camera, as opposed to rotating the lens through its nodal point.

Jan 16, 2016 at 12:15 PM

jcolwell
Online
• • • • • • •
Upload & Sell: On

p.2 #6 · p.2 #6 · TS-E Lens Vs. Pano Bracket

Gunzorro wrote:
Tareq -- I don't know about Jim's example with the 50mm lens, but very often longer focal lengths are used in the vertical or "portrait" format, with the longest side in the up/down position...

My fourteen-image pano was with the 50/2 in vertical orientation.

Jan 16, 2016 at 12:16 PM

Tareq
Offline
• • •
Upload & Sell: Off

p.2 #7 · p.2 #7 · TS-E Lens Vs. Pano Bracket

You keep saying about rotating the camera to produce that parallax, but i am talking about not rotating the camera, but you just use the shift option of the lens, so the camera is fixed on a position, and i just shift the lens from one side to another say i take 3 shifted shots without rotating or moving the camera, will that also producing that parallax? i tested and i never see any parallax in foreground of a column/pole or whatever when i shift only the lens, but correct me if i miss something there.

Even if i will use the camera, i also prefer to use something longer than 24mm or 35mm and i have RRS pano kit, and i know how to find that nodal point with it, i did that before, but i was talking about using TS lens, isn't the lens shifting on a nodal axis anyway?

Jan 16, 2016 at 04:38 PM

Peter Figen
Offline
• • • • •
Upload & Sell: On

p.2 #8 · p.2 #8 · TS-E Lens Vs. Pano Bracket

"You keep saying about rotating the camera to produce that parallax, but i am talking about not rotating the camera, but you just use the shift option of the lens, so the camera is fixed on a position, and i just shift the lens from one side to another say i take 3 shifted shots without rotating or moving the camera, will that also producing that parallax? i tested and i never see any parallax in foreground of a column/pole or whatever when i shift only the lens, but correct me if i miss something there."

If you just shift the lens, you absolutely will encounter parallax errors. You have to. The lens is moving so it's seeing the scene from a slightly different vantage point. If you're shooting at a long distance this may not be a factor as the errors are slight and the software can easily correct, but if you have something very close in the foreground, it will be a problem that software can't fix. The closer and wider, the worse it's going to be. Subject matter matters too.

This is why those of use who do these shifted panos prefer to shift the back, just like you would in a traditional view camera, but since that's not the way tilt-shift lenses work on 35mm cameras, you have to figure out a workaround. The easiest is to use an Arca-Swiss compatible plate system like Really Right Stuff, which lets you slide the camera laterally in the same increment that you're shifting the lens. Shift the lens 12mm left and slide the camera 12mm to the right. The lens stays in the same place and there's no parallax and your stitch is that much more perfect.

Jan 16, 2016 at 05:20 PM

jcolwell
Online
• • • • • • •
Upload & Sell: On

p.2 #9 · p.2 #9 · TS-E Lens Vs. Pano Bracket

Tareq wrote:
You keep saying about rotating the camera to produce that parallax, but i am talking about not rotating the camera, but you just use the shift option of the lens, so the camera is fixed on a position, and i just shift the lens from one side to another say i take 3 shifted shots without rotating or moving the camera, will that also producing that parallax? i tested and i never see any parallax in foreground of a column/pole or whatever when i shift only the lens, but correct me if i miss something there.

Even if i will use...Show more →

Actually, I've been talking about both types of parallax. Parallax due to shift and parallax due to rotation are both being discussed.

I took some shift panos in Labrador a couple of years ago, and it was necessary to compensate for lens shift, because I was shooting the edge and expanse of a fish drying surface made up of hundreds of poles. Without correcting the parallax, I probably wouldn't have been able to make a successful pano. Other times, It's not necessary to compensate, but if you always do it as "standard procedure", then you'll never get caught out by it.

Jan 16, 2016 at 05:24 PM

Tareq
Offline
• • •
Upload & Sell: Off

p.2 #10 · p.2 #10 · TS-E Lens Vs. Pano Bracket

Ok, i will give it a more test soon when i and see what you are talking about.

Jan 16, 2016 at 06:18 PM

1 2 end
FM Forums \| Canon Forum \| Join Upload & Sell 1 2 end

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.