P1394 Mail Archive: Re: P1394> EMAIL Poll

P1394 Mail Archive: Re: P1394> EMAIL Poll

Re: P1394> EMAIL Poll

Toru Ueda (ueda@slab.tnr.sharp.co.jp)
Fri, 27 Feb 98 18:11:19 JST


1 question concerning the scope of this poll ;
1) What is the scope of the "interest level"? PC printing, Consumer
peer to peer printing, or both ?

The answer is for PC printing.

If I was in charge of this poll, I would add another question to the survey.
3) Opinions on multiple printing protocols;

( ) There should be ONLY 1 printing protocol in any case and any
( X ) 2 protocols are ACCEPTABLE if necessary for different

( ) I PREFER more than 2 protocols
( ) Other (describe)__________________

Of course one protocol which covers all applications is a kind of dream
but in the current situation, it seems too far to me.
We have (at least) two protocols for printing, one is for
direct printing and another is for PC printing.

On Mon, 23 Feb 1998 18:35:51 -0800
"Gregory A. LeClair" <gleclair@agentz.com> wrote:

> ===================================================================

> 1. Please rank the following proposals according to your interest level using the following scale ( 10 = High level
> of interest vs. 0 = No interest).
( 2 ) 1284.4 Over Data FIFO Architecture (DFA)
( 0 ) 1284.4 Over SBP-2
( 5 ) Direct Printing Protocol - (current PWG-C proposal 0.71)
( 10) SBP-2 Native - (current PWG proposal 0.1c)
( ) Other (describe)__________________
> 2. Please provide background comments on your ranking.
> Why do you prefer to use the given solution?
SBP2 is a excellent protocol for initiator-target model.
This covers memory allocation, flow control and more.

> Why should others consider the given solution?
DPP's focus is simple direct printing. This might not be suitable
for PC printers.
1284.4 is a quite good protocol but is not well-suited for the
protocol over SBP2.

> Does the the given solution meet the existing requirements?

> What issues are you aware of (if any) with the given solution?
The following is my technical impression.
Because I'm not an expert of PC printing, this opinion might be
ridiculous. I don't intend to disturb PWG's symmetric protocol

SBP2 with cross logins seems complicated idea.
To support both a target function and a initiator function of
SBP2 in one device is no problem.
However these should be used separately. If one application (or
one logical connection) uses both, many additional sequences
must be defined.

For example, in the login sequence, if login from node A to node
B is OK but login from B to A is not OK, one sample sequence is
the following.

1) Application in node A calls connect routine.
1-1) A --- SBP2 login --> B
1-2) A <-- Login Ack <-- B
1-3) A <-- SBP2 login <-- B
1-4) A --> No response--> B
1-5) A --> SBP2 logout<-- B

To get a symmetrical bi-directional channel, this may be

Toru Ueda
Sharp Corp.
Software Labs.