On 1/31/25 05:25, 'T. Modes' via hugin and other free panoramic software
wrote:
GnomeNomad schrieb am Donnerstag, 30. Januar 2025 um 22:58:50 UTC+1:
Running pto_gen from the command line generates a PTO that also
reports the 35mm focal length as 141.635mm.
Drag-and-dropping the TIFs into a new PTO in Hugin reports the
correct 35mm focal length.
This does not makes sense. Both program use the same code and both
read the fov from the lens database. And this works fine with your
test image on my system.
Maybe you have 2 versions installed: one which called from the command
line and the second one when you call Hugin (I don't know how you
start it.).
Some time ago the path to the lens database changed (from ~/.hugindata
to $XDG_DATA_HOME/hugin).
Of those two, I have only ~/.hugindata. It contains only camlens.db and
expressions.ini.
Then I ran find in my ~ folder and it found two camlens.db files. One in
.hugindata, the other in .local/share/hugin. That's probably the one
from when I compiled 2024.0.1.
I renamed the .hugindata directory and the camlens.db file in
.local/share/hugin, ran pto_gen, and Hugin now reports the proper focal
length.
Possible causes?
At various times, this system has had Hugin 2022 (installed from the
Debian Bookworm repository), 2023 and 2024 (locally compiled from git
source). Maybe during the process, Hugin found calens.db in an existing
directory (.hugindata/camlens.db), created a new camlens.db file in the
new ~/hugin directory, then got confused by having the two copies
around? Or something in Hugin preferences was pointing to one (maybe the
older location)?
I thought of another possible source of confusion. Over the years, I've
changed DSLRs (Minolta Maxxum 7D, Sony SLT-A58, now Sony A7R-IVA) but
have been using the same lenses. That's one reason I moved from Minolta
to Sony. I like my lenses and have a good collection of them. I use
Sony's Minolta-Sony E mount adapter to use them on the A7R. I don't have
any E-mount lenses to test with.
I don't know how lenses are stored in the database, but is it possible
the lens is stored by name, along with attributes derived from the
camera it's on? So the lens was first added to the database (on the
Minolta), maybe its attributes were set based on that camera. Then Hugin
finds the same lens on a different camera (the A58 or the A7R). Does it
update the lens attributes in the database based on the new camera or
reuse what's already there? If it updates, does it do a complete update
or does it only change parts of the attributes (maybe leaving data
that's no longer valid)? If it just reuses existing information, would
those existing attributes still be valid when moving from a 6Mpx Minolta
to a 20Mpx A58 to a 61Mpx A7R?
Check that only one version is installed.
I ran a sudo find -name 'pto_gen' from the / folder. It found this,
./usr/local/bin/pto_gen, along with entries in the source trees.
info pto_gen reports it as version 2024.0.1.
So there was only one version of pto_gen installed, but there were two
copies of the camlens.db in different locations. Maybe the
compile/installation process could warn about that in the future?
I'm migrating to a new system. I'll compile on the new system and see
where it puts camlens.db.
Thanks for the guidance!
--
David W. Jones
[email protected]
authenticity, honesty, community
http://dancingtreefrog.com
"My password is the last 8 digits of π."
--
A list of frequently asked questions is available at:
http://wiki.panotools.org/Hugin_FAQ
---
You received this message because you are subscribed to the Google Groups "hugin and other free panoramic software" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To view this discussion visit
https://groups.google.com/d/msgid/hugin-ptx/aa520dc6-b2c2-4984-a9b1-6b303ee7dd5c%40gmail.com.
