2007.09.24 08:45 "Re: [Tiff] G3 fax: EOL needed at the beginning of each strip?", by Andy Cave

Hi Joris,

Absolutely (and I know both specs pretty well, having written a G3/G4 encoder that is 4x faster than that in TIFFLib).

ITU states that:
4.1.2 End-of-line (EOL)

This code word follows each line of data. It is a unique code word that can never be found within a valid line of data; therefore, resynchronization after an error burst is possible. In addition, this signal will occur prior to the first data line of a page.

Who says that a strip is equiavlent to a page? The TIFF spec doesn't. If it is, then sure each strip is defined as beginning with an EOL. However, if you do not define a strip as equiavlent to a page, then there is no requirement for a strip to being with an EOL.

I personally lean towards the view that you should treat each strip/tile as a separate (part of the) image and therefore each strip should being with an EOL (that's what our Lightning TIFF plugin for the Hqn RIP does), however, since in TIFF you know where each strip starts, there really is no requirement to have an EOL at the start of each strip from a s/w point of view.

I think it's therefore not true that strips have to begin with EOL.

Regards,

Andy.

----- Original Message -----

From: "Joris" <joris.at.lebbeke@skynet.be>

To: "Andy Cave" <andy.cave@hamillroad.com>; "ComSquare Martin Osieka" <martin.osieka@comsquare.ch>; <tiff@lists.maptools.org> Sent: Monday, September 24, 2007 9:32 AM Subject: Re: [Tiff] G3 fax: EOL needed at the beginning of each strip?

I'm curious - where in the spec does it indicate strips have to start with EOL? I don't remember it saying so, and checking quickly, I can't find a reference to that.

The TIFF 6.0 document does not describe the G3 algorithm at all, really. G3 and G4 are instead described in the ITU T.4 and ITU T.6 standards. G3 is also described in RFC 804.