I was just trying to work on a set of images, but PL7 consistently offers only the wrong optical module for this camera/lens combination.
The images were taken with a Nikon D5100 and Tamron SP 17-50/2.8 without VC. PL7 offers only the lens with VC - and even after I manually installed the correct optical module for my combination, PL7 still refuses to use that but “insists” on offering the wrong module. I know that these two lenses have quite different behaviour despite of their similar data and names, so it’s important to use the correct module.
Until now, I didn’t find any way to convince PL7 to use the correct and already installed optical module.
Note that I have used that with any other version up to PL5 without any problems.
Ther is no correction for such things, DxO does not provide selecting a module, unless there is a lens ambiguity. DPL 4 and DPL7 are equipped with different module databases, in which the relations are fixed.
Now, if you edited the database as a workaround, DPL should offer a selection because of a (faked) ambiguity. Editing a copy of the image file can work too, but I’d try the database workaround first.
Note that editing the database is only good until the next update. For a real fix, I’d get in touch with support.dxo.com.
If you shared a copy of the original RAW files, I could try to see if the database edit works and how.
“They” did not damage the database. If Tamron (or whoever makes a lens or body) changes metadata, DxO hast to find out first (unless the manufacturers published a list with all the exif tag names and values used in each iteration of each product) and then update the database, which is the reference (because manuf. don’t publish such data) for module selection.
Metadata is mostly kept under the carpet and all who deal with it have to reverse engineer at least some of it. Even exiftool has to deal with Tamron (and other) lens id trouble.
Again, I propose you get in touch with support.dxo.com. This might help them to fix your issue in an official way instead of some workaround database hack that only lasts until the next update…
Note that I have been using this lens for many years. And it has correctly worked with all former versions of OP and PL. So yes, “they” changed something, damaging an existing and correctly working database. Or something else is wrong in PL7 so the correct entry in the database is not found resp. referenced.
BTW, I don’t think that Tamron suddenly changed the metadata of my particular lens without having any access to it…
Yes, I will contact support about this. However, you didn’t mention yet how exactly I can see that metadata a) in the database and b) in the RAW files…
To see what’s in the database, you need a db browser for sqlite. Metadata can be checked with exiftool. In normal operations, exiftool translates tags and tag values, but you can check exiftool’s tables mapping binary to text. Search through exiftool’s documentation.
Could that even be a difference between Windows and Mac? I thought the optical module database should really be the same for both…
Here (with Windows 10), the behaviour is perfectly reproducible: PL 5.x opens the file correctly with the right module, PL 7.x only offers the wrong lens.
