Forum: Q & A

Thread: optimal tilting angle

back to threads list | this thread is closed
Search the forums:
Author  Message 
Nick Fan
[NodalNinja]

Posts: 286
Location: Hong Kong
Registered: 26 May 2006
optimal tilting angle
Posted: 30 May 2008 at 4:06 GMT
updated: 30 May 2008 at 4:08 GMT
Hi,

Many people prefer a compact single row pano bracket with tilting angle. Some popular tilting angle I hear of is 5 deg up and 10 deg down (5 deg up for sigma 8mm and 10 deg down for Nikon 10.5mm on APS-C DSLRs). However, the APS-C sensor sizes are different from different manufacturers (or even among models from same manufacturer), 1.5x, 1.6x and 1.7x. Needless to say the other format such as 1.3x and 2x.
So do these angle work for you? what is your camera and lens? Will other angle work better for you?
Your input will help me find out the ultimate tilting angles. smile
thanks a lot.

Nick
alert moderator
Hans Nyberg

Posts: 856
Location: Denmark
Registered: 28 Aug 2005
Re: optimal tilting angle
Posted: 30 May 2008 at 8:10 GMT
Tilting down 10 degrees is only useable for a fullframe fisheye and a spherical head which you can take a zenith with.
How much you want to tilt up for including zenith with a lens which gives you around 180 degree vertically depends as you say on your camera and lens and how much place the rotator take up at the bottom.

I guess 5 degrees will cover almost all lenses and combinations of 1,5 1,6 and fullframe with shaved fullframe fisheye lenses,

However I must say that besides for action shots I have gone away from doing just 3 or 4 around without zenith even if I can cover the zenith perfect. It is not worth the extra work to get perfect stitching at zenith
If you take a zenith you can also tilt down 10 degrees also with the 180 degree lens and get much better quality at Nadir. To avoid the panohead at the Nadir you just use a smaller crop circle which gives you around 165 degrees.

This solves the problem especially with Enblend that it just ignores the zenith shot if you have it covered by the other shots.

Hans
alert moderator
Nick Fan
[NodalNinja]

Posts: 286
Location: Hong Kong
Registered: 26 May 2006
Re: optimal tilting angle
Posted: 30 May 2008 at 8:42 GMT
Hi Hans,

Thanks for the input and the tip about Enblend.
I guess handheld zenith is good enough for OUTDOOR panos, right? In that case, tilting down is still a valuable option.


Nick
alert moderator
Hans Nyberg

Posts: 856
Location: Denmark
Registered: 28 Aug 2005
Re: optimal tilting angle
Posted: 30 May 2008 at 9:01 GMT

Nick Fan said:

Hi Hans,

Thanks for the input and the tip about Enblend.
I guess handheld zenith is good enough for OUTDOOR panos, right? In that case, tilting down is still a valuable option.
Nick


Yes but I would not recommend doing it with a tilted down fullframe fisheye.
At least not with a 1,6 sensor Canon. The overlap is very small and you need to have a perfect zenith shot.

Hans
alert moderator
Robert Bilsland

Posts: 13
Location: Malvern, United Kingdom
Registered: 24 Oct 2007
Re: optimal tilting angle
Posted: 30 May 2008 at 12:20 GMT
Nick,

I shoot with a Nikon D300 and 10.5mm fisheye lens. This is then mounted on top a Nodal 5 Lite panoramic head and Manfrotto tripod. With this setup I shoot 6 shots round (tilted 15 degrees down to minimise the Nadir hole) and a zenith shot. Sometimes I take a hand held Nadir shot or just cover the hole in Photoshop.

Bob Bilsland.
alert moderator
jeffE

Posts: 54
Location: Chicago, United States
Registered: 23 Mar 2007
Re: optimal tilting angle
Posted: 30 May 2008 at 13:12 GMT
updated: 30 May 2008 at 13:12 GMT
I also use the D300 & 10.5mm w/ NN3. I used to do 15º down, but now that I use a Quick Release plate on the NN3, my base is bigger and 10º is fine. 15º only gives me more tripod base so I prefer more overlap on top.

Also 6 frames around, 1 up, one down on the NN3, then one with a custom arm made by Dennis Stover to do the Nadir patch.
alert moderator
Doug Aurand

Posts: 763
Location: Albuquerque, NM, United States
Registered: 2 Jan 2008
Re: optimal tilting angle
Posted: 30 May 2008 at 14:30 GMT
updated: 30 May 2008 at 14:37 GMT
Nick
I'm not sure everyone got the point of your original question. Maybe I didn't

Are you asking about a simple, single row, non-articlulated rotator, like 360Precisin's recently announced Atome? A rotator that can't shoot a Zenith or Nadir photo at all? Like the propriatary rotator Sunex has for their 5.6mm lens?

IMHO, such a rotator that's designed for anything other than the Sigma 8mm and 4.5mm lenses won't be a very popular seller because of the large Zenith and/or Nadir "holes."

Lenses like the Nikkor 10.5mm and Sigma 10mm just don't have enough Field of View to produce a Full 360°x 360° image with s Single Row of Photos.

I'm not sure why 360Precision chose to offer a Nikkor 10.5mm version of the Atome first, when the Sigma 8mm has long been the lens that the technique of tilting up 5-15° to eliminate the need for a Zenith shot has been used on.

Tilting a Nikkor 10.5mm down 10° reduces or eliminates the need for a Nadir shot, but makes a Zenith shot even more necessary, because the "Zenith hole" gets even bigger

Doug Aurand
Albuquerque, NM
alert moderator
Doug Aurand

Posts: 763
Location: Albuquerque, NM, United States
Registered: 2 Jan 2008
Re: optimal tilting angle
Posted: 30 May 2008 at 14:56 GMT
Nick
To actually answer you question;

The smaller APS-C size sensor Canon (22.2 x 14.8 mm) uses requires more tilt than Nikon's (23.7 x 15.6 mm) sensor to cover the Zenith using a Sigma 8mm f/3.5 .

The Canons with this smaller sensor not only crops the side of the photos taken with a Sigma 8mm, in portait position, but also a little bit at the top & bottom.

I've been slowly learning to use my Canon XTi with a Sigma 8mm f/3.5 on a Nodal Ninja 3 and have found that a 10° tilt up fills the Zenith fine, leaving a "Nadir Hole" that the tips of the "black" square are only a little larger than the "black" circle left by the NN3

I can send you a Top and Bottom Cube face to show you.

While a 10° tilt up is probably more than needed with a Nikon APS-C senor and a Sigma 8mm, there's really no "downside" of tilting the Nikon a little more. The Nadir Hole with the Nikon will be smaller regardless of whether you tilt the camera 5° or 10°.

Using a 10° tilt will work with the 2 most used lines of cameras in Virtual Photography and save the manufacturing cost of producing 2 differently tilted rotators

Doug Aurand
Albuquerque, NM
alert moderator
Nick Fan
[NodalNinja]

Posts: 286
Location: Hong Kong
Registered: 26 May 2006
Re: optimal tilting angle
Posted: 30 May 2008 at 15:50 GMT
Thanks a lot for all the replies. It seems that no single setting can please everyone as I was warned. smile


Nick
alert moderator
Yuv

Posts: 6
Location:
Registered: 18 Dec 2007
Re: optimal tilting angle
Posted: 30 May 2008 at 16:55 GMT
Hi Nick,

it takes a lot to motivate me to post in a forum. congratulations, you made it!

There are two things here that need to be discussed separately: physical photo shooting and software stitching.

In terms of hardware, my request is to tilt the camera *sideway* so that the diagonal of the camera is aligned vertically.

I do tilt my Sigma 8mm by about 2° - the cropped sensor of the Canon 350D is smaller than the one of the Nikons or other Canons where 5° works too. You are right, there is no "one-size-fits-all" solution for this kind of tilting.

In terms of stitching, the software is a moving target.

Currently enblend has a bug at the zenith (well, programmers will call it a new feature request, but for users it is a bug).

Apparently also Adobe's next Photoshop version still has that bug - check their presentation at <admin.connectpro.acrobat.com/_a791863308/p4065271... after minute 16.

We were hoping to have a Google-financed student to work on that bug this summer, but we did not find an appropriate candidate.

The problems with the Zenith (and Nadir) that Hans mentions are real now, but once the software works as expected, they won't be any problem.

Even with the current software, I shoot a single row with my Canon 350D/Sigma 8mm tilted up 2° and sideways for the diagonal alignement and most stitches are acceptable.

I hope this helps. Thanks again for the sponsoring of the pano heads to hugin's Google Summer of Code team.

Yuv
alert moderator
michael medina

Posts: 284
Location: portland, oregon, United States
Registered: 27 Jan 2008
Re: optimal tilting angle
Posted: 30 May 2008 at 18:30 GMT

Nick Fan said:

Thanks a lot for all the replies. It seems that no single setting can please everyone as I was warned. smile


Nick


welll it seems that 5, 10, and 15 degrees would cover the lot wink
alert moderator
Doug Aurand

Posts: 763
Location: Albuquerque, NM, United States
Registered: 2 Jan 2008
Re: optimal tilting angle
Posted: 2 Jun 2008 at 15:49 GMT
Nick
"It seems that no single setting can please everyone as I was warned"

You mentioned both the Sigma 8mm and Nikkor 10.5mm in your original post, since the purpose of the "tilt" is to eliminate the need for a Zenith shot, the different Field of View from different lenses will require a different tilt.

So it without saying that there will be different tilt angles, since we're talking about different lenses.

Like I said before, tilting a Nikkor 10.5mm on any APS-C sensor has very little benefit, since it will take a considerable tilt up to eliminate a Zenith shot. The problem being, that much tilt makes a Nadir shot even more essential. An the problem will just be reversed if the photographer tilts down.

The Sigma 8mm lenses (both the F/3.5 and f/4) are the lenses that "tilting" up is a popular technique.

And since Nikon and Canon cameras dominate virtual photography, you could start out using a "tilt" angle on the single row rotator that that works with both.

Probably 10° like I used with my Canon Xti/Sigma 8mm/Nodal Ninja 3. This would accomodate a lot of other cameras too

If the rotator is sucessful (I think it will be) and there's demand for a 5° tilt for the smaller crop cameras like the Nikon cameras with their DX (APS-C) sensor.

Its like the approach Coastal Optics and Sigma took with the 4.88mm and 4.5mm fisheye lenses respectively. Both produce a full circular image that will fit on the smaller Canon sensor, but both companies offer them with Nikon and Canon mounts.

Looking forward to seeing the new rotator

Doug Aurand
Albuquerque, NM
alert moderator