Am 10.03.2013 13:29, schrieb Pascal de Bruijn:
>>
>> https://www.dropbox.com/sh/pq9vprixfvco6c5/WZnCjCSAlR
>
> The behavior of the lensfun module is a bit strange indeed, as I can't
> select another lens at all...
>

The problem seems to be that we do a lensfun search with LF_SEARCH_LOOSE 
when we try to find a lens. I don't know what lensfun does exactly then, 
but I assume it will just try to find the closest match, whatever that 
is. In this case it is:

lens: A-Serien Linsen, maker: Carl Zeiss, model: VCL-MHG07a

That's way off. All is still OK in the beginning, as long as the camera 
name is not set manually. The camera is not found, because we are 
looking for an exact camera match. If module lens correction does not 
get an exact match in the database no correction is applied.

As soon as the camera is manually selected, our module assumes all is 
good and a correction is done. However, with the wrong choice of a lens  :(

I was not involved, but we probably do the loose search in order to 
overcome variations in writing when in comes to the lens names. Probably 
it was found that all too often a lens was not detected although it was 
existing in the database with a slightly different spelling.

Not sure how we can overcome this situation. One solution could be a 
blacklist of lens names which libexiv2 will typically report if the 
photo was taken with a pure mechanical lens. If the lens is one of those 
we avoid any automatic detection. I fear that the blacklist might get 
quite long. On my system for the two incumbent files the lens is 
reported with a German string, even if I start darktable with LANG=C. 
Different cameras will have different strings etc.

Ulrich


------------------------------------------------------------------------------
Symantec Endpoint Protection 12 positioned as A LEADER in The Forrester  
Wave(TM): Endpoint Security, Q1 2013 and "remains a good choice" in the  
endpoint security space. For insight on selecting the right partner to 
tackle endpoint security challenges, access the full report. 
http://p.sf.net/sfu/symantec-dev2dev
_______________________________________________
Darktable-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/darktable-users

Reply via email to