Just bought NN5 and posting to say hello from Down Under.

[The Shooter]

Hi Everyone,

My name is Andy, aka "The Shooter", although work commitments have meant that I have not gone shooting as much as I would have liked for a while, I am in Perth, Western Australia.

I have just purchased a NN5, RD16 and EZ-Leveller II from Anthony at iPanoramic in Sydney, great guy to deal with, extremely helpfull and great service.

I had intended to use the gear on my "old" trusty Slik 88 tripod, which I have had for 25 years! Unfortunately, the gear arrived and I went to set it up and the cam lock on one of the legs shattered when I locked it down. Upon closer inspection with a magnifying glass I found numerous hairline cracks in the other cam locks, so I guess it was a clear case of it being too "old". So I spent some time on the net that night and bought a Manfrotto 055X PROB tripod leg set at a camera store in the city next day. Much to the annoyance of the wife, I heard mutterings of "Boys and their Toys" on more than one occassion!!

I have viewed the red-door video on setting up your Nodal Point, which seemed foolproof, well as a fool I now have to ask a question.

The video shows using two virtical poles stuck in the ground and ajustments made untill they remain fixed in the viewfinder upon rotation of your camera. They mention using a door way as well

I set up my Nikon D300 with my standard Nikon 18 -70mm zoom, set to 35mm. I am using one jamb of a door way in the foreground with that of another further down the hallway. The frames appear parrallel. For some reason when I move the camera along the horizontal either forward or backward the image does not change upon rotation. The parrallel gap remains constant instead of the jamb in the foreground being the only one I see. I have noticed that although I was careful to set up the leveller, upon rotation the bubble moves out of the inner circle, I am a bit baffled by both these anomalies, perhaps I should go and cookes some eggs as I am sure I am about to have some on my face when I get some replies.

Thanks for your assistance,

Regards
Andy

[The Shooter]

I have just read all the posts regarding the "bubble" issues or peceived issues, however my bubble is moving outside of the black circle, which is the one thing Nick says should not happen to avoid issues with stiching, so what am I doing wrong, if anything/
Thanks
ANdy

[John Houghton]

Parallax shift is going to be small when using a 35mm lens. You will probably be using a yaw increment of 16 degrees or so for shooting a panorama, so swinging the camera around from -8 to +8 degrees with the entrance pupil as much as 10mm away from the rotation axis, say, will give a parallax shift of a nearby pole of only 2.8mm relative to a background a long, long way away (as measured at the pole). Remember that with two relatively close poles, both will be affected by parallax shift, so the shift of the nearest pole relative to the pole a little further away will be less than 2.8mm. To set the no-parallax point to within 1mm would therefore involve detecting shifts of around 0.28mm or less. See http://www.johnhpanos.com/epcalib.htm for further advice.

John

[DemonDuck]

Two things --
1) You really can't use the viewfinder and/or the LCD to determine your nodal point. You have to take pictures of the objects you are using for alignment and look closely at the pictures. Start taking pictures about 6-8 mm in front of your best guess nodal point and move the camera back on the top rail 2mm for each picture until you are about 6-8 mm behind you best guess. Then look at the pictures. You can use jpg so it will be fast. It helps to write down the settings on a piece of paper and photograph the paper along with your alignment features.

2) Make sure the detente plate is sitting flat on the surface and that the rotator knob is tight enough to make a good clunk when you rotate the arm. But not too tight that it's an effort to move.

[The Shooter]

Hi John,

Thank you for your reply. Please be assured there is no sarcasim or disrespect intended in my reply, just a need to understand some of this logic.

Have you seen this video?, http://www.youtube.com/watch?v=k0HaRZi-FWs

I was directed to it by iPanoramic, from whom I purchased my NN5, the video is by the UK distributors of many different panoramic heads including NN. The video is actually using a NN, you will see than this setup is very simple. I fail to see why, having purchased this wonderful equipment, which is designed to produce excellent results with minimal mathematical equations, that is should be harder than this video shows.

The video shows a method that would enable set-up to be done in the field, I don't own a laptop to take multiple pictures to compare one to the other in layers.

Your explanation does make some sense to me, but as a Carpenter with a very good eye for vertical alignment (I can see a 1.5mm difference in plumb over the height length of a doorframe), when I see NO change at all yawing left to right, I have to assume something is wrong elsewhere?

When I looked at the recommended settings for Point A and Point B for my Nikor 18 - 70mm lens it stated 56 and 76 respectively. I know this is a guide only but my reading for A, having used the spot meter in my viewfinder to acurately place the camera vertically over the NN centerpoint, is 61mm. When you are talking about detecting shift differences of 0.28mm surely this must me the equivalent of a photographical mile!

I look forward to hearing your opinions and guidance.
Regards
Andy

[nick fan]

I have just read all the posts regarding the "bubble" issues or peceived issues, however my bubble is moving outside of the black circle, which is the one thing Nick says should not happen to avoid issues with stiching, so what am I doing wrong, if anything/
Thanks
ANdy

Hi Andy,

Please do the following to verify the bubble issue.

1. mount the rotator to the tripod WITHOUT attaching the center column. Edit: should be without attaching the upper assembly.
2. remove the detent plunger knob to avoid movement of tripod due to rotator.
3. level the bubble by looking directly above the bubble.
4. turn the rotator by 360 deg and see if the bubble will go out of the circle.
If the bubble is off, contact your reseller to get a replacement of lower rail.


Nick

[nick fan]

Hi John,

Thank you for your reply. Please be assured there is no sarcasim or disrespect intended in my reply, just a need to understand some of this logic.

Have you seen this video?, http://www.youtube.com/watch?v=k0HaRZi-FWs

I was directed to it by iPanoramic, from whom I purchased my NN5, the video is by the UK distributors of many different panoramic heads including NN. The video is actually using a NN, you will see than this setup is very simple. I fail to see why, having purchased this wonderful equipment, which is designed to produce excellent results with minimal mathematical equations, that is should be harder than this video shows.

The video shows a method that would enable set-up to be done in the field, I don't own a laptop to take multiple pictures to compare one to the other in layers.

Hi Andy,

The method only works if every alignment is perfect. There are many critical alignments in the system. First the rails of NN must be perfectly square which is hard to acheive. Second the weight of camera and lens can cause some small bending of the rails, making the rails not perfectly square. Thirdly, many times the viewfinder of camera and sensor are not perfectly centered on the lens axis. With all these possible deviation from perfect alignment, it can be assured that the above method can only give a crude setting that needs further refinement. Use John's (conventional) method for the refinement. The conventional method mimic the stitching condition. So if you get no parallax in calibration, you get no parallax in stitching. It also put more emphasis on the horizontal than the zenith and nadir.


Nick

[The Shooter]

Hi Andy,

Please do the following to verify the bubble issue.

1. mount the rotator to the tripod WITHOUT attaching the center column.
2. remove the detent plunger knob to avoid movement of tripod due to rotator.
3. level the bubble by looking directly above the bubble.
4. turn the rotator by 360 deg and see if the bubble will go out of the circle.
If the bubble is off, contact your reseller to get a replacement of lower rail.


Nick

Hi Nick,

I am unsure what you mean by without attaching the centre column. The centre column on my Manfrotto tripod is not removable. I have enclosed two photo's to show you my current set up. The raised column photo is only to show you how the underside is scolloped to rest snugly when in the retracted position; which is how I have it when using the NN5.
Thanks Andy

[John Houghton]

Andy, Setting up the panohead in the field is certainly a different proposition to setting up at home, but generally the setting up has only to be done once. The stops on the top and bottom rails enable the head to be assembled quickly and accurately in the preset positions. Of course, for a zoom lens to be used at different focal length settings, you need to determine multiple upper rail settings.

The panohead setup becomes less critical as the focal length is increased. Usually, longer focal lengths are used for landscape subjects, with things so far away that parallax effects become negligible and stitching is easy. Even shooting with a hand held camera is quite practical in such situations. For indoor 360x180 panoramas, fisheye lenses are commonly used and the video demonstrates the setup with what looks like a Sigma 8mm fisheye lens. The setup needs to be done accurately to avoid major parallax effects with things so close to the camera, otherwise major problems will be arise in the stitching.

If you cannot see any parallax effects directly in your tests, then I suggest you take two pairs of shots: one pair with the camera as far back on the top rail as possible and the other pair with the camera as far forwards as possible. Each pair should be taken with the camera focused as it will be for shooting a typical panorama, as the entrance pupil can change position with focus distance. Rotate the head between each shot so that the poles (or door frames) are at each side of the frame. Overlay the pairs of images in Photoshop and align the backgrounds. Switch the upper image on/off and see if there is any shift in the near object. Note the direction in which it moves as you switch the top layer on. If you find that the near object shifts in opposite directions for the two pairs of shots, then it will be possible to find an intermediate position of the camera on the upper rail where the parallax shift becomes zero. If the shift is the same direction for both pairs, then it will probably be because the entrance pupil is always behind the rotation axis and you will need to reverse the top rail so it points forwards rather than backwards.

John

[nick fan]

Hi Nick,

I am unsure what you mean by without attaching the centre column. The centre column on my Manfrotto tripod is not removable. I have enclosed two photo's to show you my current set up. The raised column photo is only to show you how the underside is scolloped to rest snugly when in the retracted position; which is how I have it when using the NN5.
Thanks Andy

sorry, stupid me. I meant "without attaching the upper assembly".
Believe me it is what I said in my mind. :-)

[John Houghton]

Andy, It occurred to me that the reason why you haven't seen any parallax shift when rotating the camera as per the video is that you probably have lens aperture wide open. You need to use a small aperture such as f/22 to minimise the size of the entrance pupil.. This isn't necessary with a fisheye lens (as used in the video), since the entrance pupil is small even at full aperture.

John

[donnnnnny]

I too am having alot of issues finding a nodal point or the point for non parallex movement.Im shooting with a nikon d700 with battery grip and the lens is a 24-85mm.
i have tried many times with no luck yet. It really does not seem to matter where i set the camera, there never is much movement or difference between my two parallel lines.
I have read all the above and will try again
cheers don,

[John Houghton]

Don, Try following these instructions as closely as you can to repeat a test I have just done:

1. Find a window looking out onto a reasonably distant scene.
2. Set you camera on the panohead at a zoom setting of 70mm and aperture f/22. Focus on the background, not the window.
3. Position the tripod on a firm footing and so that the focal plane of the camera is 100cm from the window.
4. As a nearby target object, either stick a short length of sticky tape vertically on the window or use maybe the cord of a venitian blind as I used in the image below. You need to have some clear features in the background behind the tape or cord so you can detect any parallax shift.
5. Rotate the panohead to bring the tape or cord near to the right side of the frame and take a shot.
6. Rotate the panohead to bring the tape or cord near to the left side of the frame and take the second shot.
7. Open both images in Photoshop and drag the second shot onto the first shot, where it will form a new layer.
8. Set the opacity of the top layer to 50% and align the backgrounds of the two images. Zoom in to 100% to display the near object.
9. Restore the opacity to 100% and switch the top layer on/off. The background should not move but the near object will shift sideways like this:

http://www.johnhpanos.com/shift10.gif

In this example, I have deliberately positioned the camera 10mm in front of the optimum no-parallax position. The image is a 100% crop (40D images). However, in order to eliminate some minor distortion and get a perfect match of the backgrounds, I have stitched the images in PTGui, putting control points only on the background features. A layered PSD output was then generated, cropped to the area of interest.

I measured the shift at 26 pixels for this setting on the panohead. At 2mm in front of the NPP, the shift was 5 pixels. As you can see, the shift is very visible. Anybody should be able to reproduce similar results without any difficulty.

John

[Justin H]

Hi Andy,
I also have a NN5 bought from Anthony. I was also getting the same bubble movement, it ended up being as simple as pulling the R16 head apart and tightening the centre screw. It wasnt tight and therefor let the NN5 head move around. In saying that I am yet to get the whole set up right....

OK, so Ive been to Forums, check out the set up methods, done the flower head and wire rack etc etc but I cant get it right. I always end up with my horizontals having vertical misalignments in them ie a wall will have a misaligned join in it. Trying the wire rack set up I couldnt get it perfect. I was down to changing adjustments by 1 mm. Always the same prob with misaligned joins.
Im using my old Nikon D70 with a Sigma 10-20 on it, always set to 10mm. Im lost as to what to do. I have a Nodal Ninja 5 with a R16 head. Oh and Ive tried different rotator degrees.

Am I missing something basic here??

Looking forward to any help. Its driving me nuts. Ive been at this for many many months so if anyone can help??!!!??!!! The only thing I havent tried, which I will do today is shutting down the aperture. Ive been testing at around 5.6

[nick fan]

Hi Andy,
I also have a NN5 bought from Anthony. I was also getting the same bubble movement, it ended up being as simple as pulling the R16 head apart and tightening the centre screw. It wasnt tight and therefor let the NN5 head move around. In saying that I am yet to get the whole set up right....

OK, so Ive been to Forums, check out the set up methods, done the flower head and wire rack etc etc but I cant get it right. I always end up with my horizontals having vertical misalignments in them ie a wall will have a misaligned join in it. Trying the wire rack set up I couldnt get it perfect. I was down to changing adjustments by 1 mm. Always the same prob with misaligned joins.
Im using my old Nikon D70 with a Sigma 10-20 on it, always set to 10mm. Im lost as to what to do. I have a Nodal Ninja 5 with a R16 head. Oh and Ive tried different rotator degrees.

Am I missing something basic here??

Looking forward to any help. Its driving me nuts. Ive been at this for many many months so if anyone can help??!!!??!!! The only thing I havent tried, which I will do today is shutting down the aperture. Ive been testing at around 5.6

what is your horizontal and vertical shooting intervals? what stitcher do you use? Can you upload a set of images for others to diagnose? Pano head calibration is just part of the equation. Lens calibration (parameters in ptgui etc) is just as important.

NIck