Phase One .IIQ raw files

[PJ]

Hello everyone.

I was wondering if anybody knows the structure of tags in the Phase One .IIQ raw files and what is possible to write in there. ExifTool reads some IFD0 camera information and basic EXIF. Now the question is if it is possible to write XMP or IPTC in those files at all?

I addressed my question to everyone, but my guess is it is going to be Phil who may know the answer. Phil, ExifTool is amazing. Such a solid piece of software. Thanks a million!

PJ

[Phil Harvey]

IIQ is another TIFF-based format, so in theory you should be able to write anything you can write to a TIFF.  However, I haven't added write support for this format because I don't have the ability to test this.

- Phil

[PJ]

Thanks Phil. Is there a way I could be helpful in this process? Test extensively?

Best, PJ

[Phil Harvey]

Hi PJ,

Yes, you could help.  The first step would be for you to send me a sample IIQ from each camera you want supported.  My mail should be able to handle something that large (philharvey66 at gmail.com), but if this doesn't work you'll have to upload them somewhere for me (rapidshare.com for example if you don't have a better server).

Then I'll analyze the image(s) and get back to you with a plan of action.

- Phil

[PJ]

Great,

I will put a few sample files on a dropbox and will send you a link tomorrow. That sounds fantastic!

Thanks, PJ

[Phil Harvey]

Hi PJ,

Thanks for the sample.  I have had a quick look at it.

Unfortunately Phase One is going to make this more difficult than it needs to be.  There is an unreferenced data segment at the start of the image which looks like it contains the raw image data.

To be able to rewrite this image I will need some way to determine the size of this raw data segment, and analyze it to see if it contains any pointers to other information in the file.

For this I will need IIQ samples from as many different models as you can provide.

Also, do you have Phase One software to use for testing if I provide you with edited versions of these files?  We need to be sure that the OEM software doesn't have any problems with the edited images.

   - Phil

[PJ]

Yes, I will pull some of the files from other backs from our tape archive and will send it to you shortly.

Thanks!!! PJ

[PJ]

Hey Phil,

I have a few new files ready to go. Will send you links shortly. Some of the files are .TIF files. Well, Phase One decided to use this extension for their RAW format. .TIF and .IIQ is exactly the same file format with just different extension. (see: http://forum.phaseone.com/En/viewtopic.php?f=4&t=6749).

To make things more confusing .TIF file is not a TIFF file. I am sure you will be able to notice those nuances in a second.

Hope it helps :-)

PJ

[Phil Harvey]

Thanks.  I got the .TIF.  It is unfortunate that they did this, but I think it should be possible to recognize the IIQ regardless of extension.

- Phil

[PJ]

From what my understanding is .IIQ and .TIF is the same file type that depending on user preferences you can name as .TIF or .IIQ. See the screenshot in the link:

http://dl.dropbox.com/u/5594666/Capture%20one%20prefs.png

.IIQ files from different cameras may look different mostly in the raw data field of the file. Some of the models were performing compression in the back, some were performing compression in the software that operates the back in the tethered mode, older models were not performing any compression at all.

The software that operates backs "Capture One" has the ability to read xmp, IPTC, exif when it works on other file formats like CR2 or jpeg. It's only when it comes down to their own .IIQ files where it forgets it's metadata skills.

Cheers, PJ

[Phil Harvey]

Thanks for this information.

The main difference between the P25 TIFF you sent and the P45+ IIQ is that the P25 image had two IFD's (IFD0 and IFD1) which referenced the same reduced-resolution image (I have no idea why), while the P45+ had only one (IFD0).

- Phil

[Phil Harvey]

I had a closer look at both samples you sent.

The raw data is actually part of the maker notes, so it would be preserved when the image is rewritten as a TIFF with ExifTool.

Could you test to see if the Phase One software will still recognize the raw data after writing with ExifTool?  First, rename the IIQ to ".TIF", then use a command something like this (note the -m option is necessary to ignore some minor errors):

exiftool -artist=me -m FILE

If this works, then implementing IIQ support will be very easy.  If not, then the Phase One software is sensitive to the location of the raw data in the image (which would require an ugly patch to the ExifTool TIFF writer to put the raw data in a specific location).

- Phil

[PJ]

Well, that's interesting. After renaming .IIQ to .TIF i can use exiftool with the parameter -m. I still get the error "Warning: Maker notes could not be parsed - TIF from IIQ.TIF", but "1 image files updated"!!!

Even Capture One recognizes new tags in it's own RAW files. Images process properly, no errors noticed.

This is awesome, yet another example how genius tool ExifTool is. I will keep on testing on larger scale and will keep you updated regarding results. My goal is to write a complete MWG metadata to 2000 files and process it to jpegs.

BTW what does "Warning: Maker notes could not be parsed" mean?

Best, PJ

[Phil Harvey]

Hi PJ,

This is good news.  I still want to do some work to add proper recognition for this file type, which will remove the "Maker notes could not be parsed" warning.  With the P25 there is another warning about a double-referenced ICC_Profile directory (referenced from both IFD0 and IFD1), which is possibly a more serious problem since ExifTool will delete the reference in IFD1 when the image is rewritten.  I need to apply a patch for this.

Also, we should test with as many different models as possible, so I would appreciate it if you send me samples for any other models you may have available.  Thanks.  So far I have your P25 and P45+ samples, plus I found a P65+ sample that I have tested.

- Phil

Edit: I forgot to answer your last question:  It is a warning which indicates the makernotes were copied as a black box instead of being properly parsed.  This will corrupt any absolute offsets in the makernotes if they exist, but it looks like the IIQ makernotes are relocatable, and if so this is not a problem.

[PJ]

Phil, I just sent you a complete set of RAW files from all of the P-series backs. I don't think I will be able to pull any of the RAWs from the old H-series backs as they were discontinued a few years ago.

Thanks for an update on makernotes. I have no idea what it means :-).

If you think that finding files form H-series backs is important I will keep on digging through archives.

Best, PJ