Talk:Entrance Pupil Database
- 1 Number of picture for 360 / Degrees between picture
- 2 Tripod Mount Length L1
- 3 Third party lenses: Tripod mount to NPP : L, L1 and L2.
- 4 Focal Length / # of pics for 360° / Degree between pics
- 5 Enhance
- 6 Canon 5d Mark III
- 7 Tables
- 8 Panasonic Lumix LX3 Lens Base
- 9 Sigma 8mm F4.0 EX
- 10 Entrance pupil shift
- 11 Distance between the sensor to the nodal point.
Number of picture for 360 / Degrees between picture
Thanks for adding your measurements.
I have a question. I thought the degrees between the pictures would only be applicable to Fisheye lenses ? Does that also apply to normal lenses ? Or is the difference only noticable in Fisheye lenses, and too small have an impact on normal lenses.
-- Korffr 10:40, 22 May 2005 (EDT)
Sorry, I don't know if there is a difference between fisheye and normal lenses. For me it is more an information about the number of pictures needed for 360 degrees with enough overlap. These values are for portrait orientation. Probably we should make a third table with additional things like this then put all the informations in one table, isn't it?
--Pitdavos 11:31, 22 May 2005 (EDT)
Tripod Mount Length L1
The text for this measurment includes "For fixed lens camera's this column is not applicable." Is this really true ? A lot of "Fixed" lens cameras do allow the addition of an adaptor tube which then permits the additional lenses to be added - whilst this implies that such cameras are not "Fixed lens" the way it reads now I suspect puts people off inserting a measurment for cameras which include a Fixed lens.
Unless there is another reason I would propose to change the text to read; "Looking at the bottom of the camera, the distance between the middle of the tripod mount to the base of the lens base, or base for adaptor tube. For fixed lens camera's which do not allow use of adaptors this column is not applicable."
OK well no comments after several weeks so, changes made and the Fuji S7000 is listed with the length of its adaptor tube as well as an example
Third party lenses: Tripod mount to NPP : L, L1 and L2.
Problem : The list includes a number of "third-party" SLR lenses (Sigma, Tamron, Tokina, Zeiss, etc) which are available with different lens mounts (Nikon F, Canon EF, etc). Each distinct mount system usually has a different flange-back-distance (see http://en.wikipedia.org/wiki/Lens_mount#List_of_lens_mounts), ... which means that the L2 distance for nikon-fit Sigma 8mm f/4 will be different from the canon-fit version. Where the mount system is not known, this ambiguity renders the current L2 data less than useful.
Proposed solution A : Following the example of http://olypedia.de/Nodalpunkte_E_System let L2 be measured from "NPP to image plane" (ie the sensor or film plane) because - to the best of my knowledge - this will be consistent for a given lens regardless of the lens mount system. Also let L1 be from "tripod screw to image plane".
- Task 1 : Updating L2 data for all the 'known mount' lenses is a simple job : for Canon-EF lenses add 44mm, for the Nikon-Fs add 46.5mm, for Minolta-AF add 44.5mm, etc. For the 'unknown mount' lenses, as I say, the current data is unusable, so no useful data is lost by switching over the image plane measurement system. Perhaps the 'ambiguous' data could be preserved - in the hope of future clarification - in a separate table, possibly here on the discussion page ?
- Task 2 : Updating L1 is simply a matter of subtracting the appropriate FBD from the current L1 value (eg: Nikon D90 = 40mm - 46.5mm = -6.5mm). The instructions for how to measure L1 could be updated to something like "measure from tripod to mount flange and then subtract the appropriate FBD ..." ?
- Task 3 : Data for H and C are unaffected by this change. (That's an easy one!)
- Task 4: Many cameras with fixed-on lenses will also have an exterior image plane marker (circle with a line through it), but IMO it would be easier to have a separate table for fixed-lens cameras which might as well indicate NPP to tripod socket (L) directly.
Comments most welcome! --Beeswax 02:17, 29 January 2011 (UTC)
I don't have any of these lenses to accurately measure myself, but found some other sources of lens data. I will do some editing and see if I can come up with a suitable table format. --Plattro (talk) 17:24, 7 July 2013 (CEST)
Improved the table accordingly, I think it's better now. Assuming most are with either Nikon or Canon, the difference is just 2.5mm, so the data isn't useless. Most people's measurements accuracy is within this range, and the value is at least a good starting point.
Focal Length / # of pics for 360° / Degree between pics
The number of pictures for 360° and the angle between two pictures doesn't depend only on focal length... It also depends on the size of the sensor. For instance the angle is roughly the same for 28mm on an EOS 5D (24x36 sensor) and 17mm on an EOS 350D (APS sensor).
I suggest to put this information to a separate table. JX 16:46, 10 April 2008 (CEST)
- Good idea. (When setting up this table there where almost no full frame sensors available. Could you contribute values?--Erik Krause 18:19, 11 April 2008 (CEST)
- I will.. when I get time, probably in May. JX 18:13, 14 April 2008 (CEST)
- Perhaps this "Number of pics for 360°" info deserves its own page/article rather than being appended to the already lengthy "Entrance Pupil Database" ? --Beeswax 02:23, 29 January 2011 (UTC)
The Entrance pupil database needs constant enhancement. Please enter the NPP-settings you found for your camera.
Canon 5d Mark III
Does anyone have tested if Mark II and Mark III are the same? - Skal 13:37, 9 January 2013 (UTC)
My suggestion would be to, to split the tables or to mark them with a color according to the manufacturer Dezen 19:48, 10 December 2009 (UTC)
Panasonic Lumix LX3 Lens Base
Just wondering if anyone who's familiar with the LX3 could clarify where the lens base is - is it where the lens section protrudes from the main body of the camera, or is it after the ring that contains the aspect and focus controls (i.e., where the 46mm accessory thread is)? Any help would be greatly appreciated. --Martan74 17:09, 2 April 2010 (UTC)
Sigma 8mm F4.0 EX
I just did some measurements and found an Entrance Pupil Length of 47mm for this lens. Since there are three other measurements I don't know what to put on the page.
--Osiris 11:30, 30 September 2010 (UTC)
- Could well be all of them are right. Fisheyes don't have a single NPP: See this Graph --Erik Krause 18:54, 30 September 2010 (UTC)
- Thanks for the information and the Graph - this explains a lot. I think it would be good to mention this on the page? Would it be possible/allowed to link to this graph? --Osiris 13:06, 4 October 2010 (UTC)
Entrance pupil shift
Fisheyes and wide angles
A number of fisheye and wide lenses show strong entrance pupil shift (See the animations on the web page by Pierre Toscani!!) by angle from center.
The Nikon 10.5 and 16mm fisheye's entrance pupil shifts by around 15mm, the ideal LPP point for panoramas is therefore a bit of a compromise, depending on the amount of image overlap and I don't know what else. The Nikon 14mm f2.8D and other wide angles do this too
I recently noticed some people's measurements conflicted with mine, I am however quite confident in the accuracy of my measurement setup now. The reason I found is that entrance pupil sometimes shifts with the aperture wide open to to internal vignetting away from the aperture,
I now always measure by observing the aperture from the front and with the lens stopped down. Measurements made with the lens wide open (Especially with the parallax method) may therefore differ from mine. This means the ideal NPP could differ slightly if you shoot wide open.
The Nikkor AF 50mm f1.8D does this noticeably. The apparent aperture (50/1.8=27mm) is bigger than the shift, so I expect the difference is smaller than the DOF and not noticeable anyway.
In the above cases I have tried to make the appropriate setting to use clear in the descriptions
Distance between the sensor to the nodal point.
I'am Daniele, firstname.lastname@example.org, I am following the Entrance Pupil database from some time but I found, I think, a little bug in the way we consider the various distances. Mainly I think that is not a proper way of measure that from the lens mount border to the center of the stative mount screw, because the sensor is not located in that point, but little back.
To let all the measures be usable it need to calculate also the distances from the mount screw and the sensor surface.
- The sensor position is of no interest here. The entrance pupil (the no-parallax-point) is a property of the lens only. The sensor position doesn't alter this property in any way. So if you want to adjust your panoramic tripod head, you need the distance from the tripod mounting screw to the rotation axis of the head, that's why those measurements are used here. --Erik Krause (talk) 19:53, 25 October 2016 (UTC)
What I mean is that a Canon EOS 7D using a Canon 15-85 lens at 15mm the Entrance Pupil is at 100mm from the mounting screw to the lens mount border, but in a EOS 550D the position can be different. So, the measure are valid only for the specific lens, specific camera at specific focal lenght. Correct??
- The lens measurements are valid for the specific lens at a given focal length, the camera measurements are for the camera. Add both to get the value for your combination. --Erik Krause (talk) 19:55, 25 October 2016 (UTC)
Today I insert the Entrance Pupil Point position for the Canon 15-85mm lens, that was missing from the list. I would like to know if someome else take this measures, with the 7D and with other Canon body, just to compare the results and better understand the question.