You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The "file --mime-type" command in Linux gives an "image/tiff" mime type, but when I run the file through ExifTool it says "image/x-hasselblad-fff" so I didn't put anything in the signature file for the mime type since I'm not sure which to use.
Description
Hasselblad FFF Raw Image is a proprietary raw image format generated by some Hasselblad equipment.
Format type
Image (Raster)
Vendor
Hasselblad
File format identification signatures
4D4D002A (which is "MM.*" in ASCII) BOF
48617373656C626C616420483344 (which is "Hasselblad H3D" in ASCII) variable.
Relevant links, documentation, extra information
http://fileformats.archiveteam.org/wiki/Imacon_3F -- I'm not sure if Imacon and Hasselblad-created files are different since I only have Hasselblad files. Presumably, Imacon files won't have "Hasselblad H3D" in the hex and thus won't be identified by the signature as I've written it. But if I take that part out and use only the "MM.*" at the BOF, it collides with some other DNG and TIF files.
There is an existing signature (fmt/1062) for Hasselblad 3FR Raw Image, which is apparently a successor to Hasselblad FFF Raw Image.
We will need to flesh this out a little more. Originally the 3FR puid was disabled because of too many clashes with TIFF's, I fear this will be the same. The string Hasselblad will be in the TIFF's and 3FR files and may clash, so more research will be needed to find a unique difference in the FFF files. Technically they are processed RAW files instead of unprocessed 3FR files. So they will be very similar. Currently all my FFF samples are identifying either as a TIFF or an Exchangeable Image File Format (x-fmt/388)
Ok, after some testing, the FFF format is very similar to the 3FR format. The makernote has one byte different and so far they are all big-endian instead of little. Also the 3FR signature looks for TIFF tag 271 which is a Make tag where the string has Hasselblad in it. This tag exists in the FFF file as well, but jumps around in my samples from near the BOF to near the EOF. Hard to nail it down. So I consolidated the signature to just look for the TIFF header and then the makernote, which seems specific to the FFF format. So here is the pattern I propose.
BOF: 4D4D002A{4-2048}000500040000
There is an older version of the FFF format made by a different software tool called Flexcolor, these are very different, but do contain a unique TIFF Tag, 46275 is a Hasselblad Raw Image tag and can be found in FFF files from Flexcolor. I will add as a separate signature.
Format name
Hasselblad FFF Raw Image
Extensions
.fff
MIME/Media Type
The "file --mime-type" command in Linux gives an "image/tiff" mime type, but when I run the file through ExifTool it says "image/x-hasselblad-fff" so I didn't put anything in the signature file for the mime type since I'm not sure which to use.
Description
Hasselblad FFF Raw Image is a proprietary raw image format generated by some Hasselblad equipment.
Format type
Image (Raster)
Vendor
Hasselblad
File format identification signatures
4D4D002A (which is "MM.*" in ASCII) BOF
48617373656C626C616420483344 (which is "Hasselblad H3D" in ASCII) variable.
Relevant links, documentation, extra information
http://fileformats.archiveteam.org/wiki/Imacon_3F -- I'm not sure if Imacon and Hasselblad-created files are different since I only have Hasselblad files. Presumably, Imacon files won't have "Hasselblad H3D" in the hex and thus won't be identified by the signature as I've written it. But if I take that part out and use only the "MM.*" at the BOF, it collides with some other DNG and TIF files.
There is an existing signature (fmt/1062) for Hasselblad 3FR Raw Image, which is apparently a successor to Hasselblad FFF Raw Image.
Signature XML
Hasselblad-FFF-Raw-Image--signature-file.zip
Credit
The Church of Jesus Christ of Latter-day Saints and Brigham Young University
The text was updated successfully, but these errors were encountered: