2013.06.14 17:20 "Re: [Tiff] TIFF Specification Allowance for Multi-Page vs. Reduced Res 2nd IFD", by Douglas Peterson

I've read the Tech Note 1 and explored a TIFF saved with Image Pyramid in PS CS6 with JHOVE and TiffUtil. I see no signs of the SubIFD tag anywhere in the TIFF I saved with Image Pyramid. Am I missing something?

Here is a link to that file: https://www.dropbox.com/s/ziwtxddcirwy4w6/1_120ASPH_f8_2900ppi_99eff-imagepyramid.tif

Based on Chris' comment that I should match the structure created when selecting Save Image Pyramid I must conclude we are "doing it right" as I see no sign of the SubIFD tag in such a file and only a NewSubFileType indicating they are reduced resolution versions of the primary. But based on Joris' comment we are doing it wrong and should make sure the resulting TIFF includes SubIFD references from the primary to the IFD with the reduced-resolution-of-primary.

Any further thoughts would be greatly appreciated. I know you've already been generous to reply at all.

On Thu, Jun 13, 2013 at 3:53 PM, Douglas Peterson < dep@digitaltransitions.com> wrote:

Joris and Chris: thank you both. I'll do just as you suggest as my next avenues of exploration.

Thanks for taking the time to reply!

On Thu, Jun 13, 2013 at 3:27 PM, Chris Cox <ccox@adobe.com> wrote:

Save a TIFF from Photoshop with the “Save Image Pyramid” checkbox enabled.

Look at the structure, and match it.

Also, see the SubIFD definition in TIFF Technote 1.

On 6/13/13 8:52 AM, "Douglas Peterson" <dep@digitaltransitions.com>

>> wrote:

> *Background About Me

> *I have put considerable effort into understanding the proper structure > for TIFFs. I've read an Intro to TIFF Creation <

> http://paulbourke.net/dataformats/tiff/> and the NewSubFileType <

> http://www.awaresystems.be/imaging/tiff/tifftags/newsubfiletype.html> > spec and learned to read hex (at an intro level). I have JHOVE and MAC OSX

TiffUtil at my disposal. I say all this not to brag (as I'm sure I'm one of the least knowledgable members of this mailing list) but only to explain that I have tried very hard to resolve this issue on my own before appealing for help from this community.

> *The Problem

> *Our company is having a problem with several software packages reading

our TIFFs as multi-page documents. That is, upon import/ingestion the software treats the two image IFDs in the TIFF as separate images that should both be imported.

Using JHOVE and TiffUtil I see our TIFFs create two IFDs (plus an EXIF

IFD which is presumably not important to this discussion). IFD#0 is the

primary image at full resolution. IFD#1 is a thumbnail of the primary image.

We want to continue to provide TIFFs with the IFD that contains a thumbnail as some users may have incorporated the thumbnails into their workflows. But we do not want these other programs to see the TIFFs as multi-page documents.

Exploring the structure of the TIFFs we are creating I cannot see

anything wrong. IFD#1 is, as far as I can see, set to [Bit 0 = True][Bit 1

= False][Bit 2 = False][all other bits = False]. Which according to the

spec should mean this IFD should be considered [reduced-resolution version

of another image in this TIFF file] and not [single page of a multi-page image].

*My Questions for the Community

> *1) Can you please verify that our TIFF is following the specification

regarding tagging IFDs representing thumbnails of the primary image?

Can anyone clarify what the resulting JHOVE/TiffUtil/Hex would be in the following cases:

--- Bit 0 is set to 1 and bit 1 is set to 1
--- Bit 0 is set to 0 and bit 1 is set to 1
--- Bit 0 is set to 1 and bit 1 is set to 0
--- Bit 0 is set to 0 and bit 1 is set to 0

3) Is there any consensus on why the specification is written as a series of bit flags (which allows in theory for a given IFD to be flagged as both

>> a multipage document and a reduced resolution version of t