P1394 Mail Archive: Re: P1394> EMAIL Poll

Re: P1394> EMAIL Poll

ALAN_BERKEMA@HP-Roseville-om2.om.hp.com
Tue, 24 Feb 1998 09:52:29 -0800

begin 777 1.txt
M4F5C96EV960Z(&9R;VT@<&%L<F5L,2YH<"YC;VT@*'!A;')E;#$N:'`N8V]M
M(%LQ-2XX,2XQ-C@N,3!=*2!B>2!V96YU<RYR;W-E+FAP+F-O;2!W:71H($53
M3510("@X+C<N,2\X+C<N,R!425,@-2XP($]P96YM86EL*2!I9"!304$Q.34W
M,B!F;W(@/&%L86Y?8F5R:V5M84!H<"UR;W-E=FEL;&4M;VTR+F]M+FAP+F-O
M;3X[($UO;BP@,C,@1F5B(#$Y.3@@,3@Z,SDZ,C$@+3`X,#`@*%!35"D*4F5C
M96EV960Z(&9R;VT@;&ES=',N=6YD97)S8V]R92YC;VT@*'5S8V]R92TQ+FUV
M+F-O;2!;,3DY+C$R-2XX-2XS,%TI"@EB>2!P86QR96PQ+FAP+F-O;2`H."XX
M+C8O."XX+C5T:7,I('=I=&@@15--5%`@:60@4T%!,34P,3$["@E-;VXL(#(S
M($9E8B`Q.3DX(#$X.C,Y.C$P("TP.#`P("A04U0I"E)E8V5I=F5D.B!F<F]M
M(&QO8V%L:&]S="`H9&%E;6]N0&QO8V%L:&]S="D@8GD@;&ES=',N=6YD97)S
M8V]R92YC;VT@*#@N-RXU+S@N-RXS*2!W:71H(%--5%`@:60@5D%!,3`V-#,[
M($UO;BP@,C,@1F5B(#$Y.3@@,C$Z,SDZ,#$@+3`U,#`@*$535"D*4F5C96EV
M960Z(&)Y('!W9RYO<F<@*&)U;&M?;6%I;&5R('8Q+C4I.R!-;VXL(#(S($9E
M8B`Q.3DX(#(Q.C,X.C`Q("TP-3`P"E)E8V5I=F5D.B`H9G)O;2!D865M;VY`
M;&]C86QH;W-T*2!B>2!L:7-T<RYU;F1E<G-C;W)E+F-O;2`H."XW+C4O."XW
M+C,I(&ED(%9!03$P-3`Y(&9O<B!P,3,Y-"UO=71G;VEN9SL@36]N+"`R,R!&
M96(@,3DY."`R,3HS-SHT-R`M,#4P,"`H15-4*0I-97-S86=E+4E$.B`\,#%"
M1#0P.#DN13(V,#0X03`N9VQE8VQA:7)`86=E;G1Z+F-O;3X*1G)O;3H@(D=R
M96=O<GD@02X@3&5#;&%I<B(@/&=L96-L86ER0&%G96YT>BYC;VT^"E)E<&QY
M+51O.B`B9VQE8VQA:7)`86=E;G1Z+F-O;2(@/&=L96-L86ER0&%G96YT>BYC
M;VT^"E1O.B`B)S$S.30@4%=')R(@/'`Q,SDT0'!W9RYO<F<^"E-U8FIE8W0Z
M(%`Q,SDT/B!%34%)3"!0;VQL"D1A=&4Z($UO;BP@,C,@1F5B(#$Y.3@@,3@Z
M,S4Z-3$@+3`X,#`*3W)G86YI>F%T:6]N.B!*87E'964@0V]M;75N8V%T:6]N
M<PI8+4UA:6QE<CH@36EC<F]S;V9T($EN=&5R;F5T($4M;6%I;"]-05!)("T@
M."XP+C`N-#(Q,0I%;F-O9&EN9SH@-#`@5$585`I396YD97(Z('`Q,SDT+6]W
,;F5R0'!W9RYO<F<*
`
end



===================================================================

1. Please rank the following proposals according to your interest level using
the following scale ( 10 = High level
of interest vs. 0 = No interest).

( 1 ) 1284.4 Over Data FIFO Architecture (DFA)

+ Uses existing 1284.4 protocol.
+ Push Push Solution for efficient 1394 unified write transactions.
+ Basic Premise is fairly simple.
- Needs additions for lost, duplicate and out of order packets.
- Needs Fragmentation and re-assembly.
- Effort really turns out to out to be similar to solving the problems that
IP1394 is designed to address.
- Has same host side CPU intensive processing concerns that IP1394 has.

( 0 ) 1284.4 Over SBP-2
+ Uses existing 1284.4 protocol.
- Don't need CBT with SBP-2 pull model.
- Has all the 1284.4 overhead as well as all the things needed to make SBP-2
virtually symmetric.

( 2 ) Direct Printing Protocol - (current PWG-C proposal 0.71)
Answering this question in the context of considering this for the PWG. This is
not intended to detract from the use of DPP by the PWG-C .

+ Has the goal of being thin. Less code and overhead than SBP-2.
+ Up front design to include Peer to Peer.
- Not an existing standard.
- Could grow to be almost as thick (complex) as SBP-2.
- Appears to have similar CPU intensive processing concerns as IP & DFA.
? Does it address multiple host to multiple devices?

( 7 ) SBP-2 Native - (current PWG proposal 0.1c)
+ Existing standard.
+ Designed for 1394 shared memory model.
+ Lends itself to efficient implementation using DMA descriptor based link
controller hardware.
+ Support by Microsoft.
+ Support by Apple.
- Need to abstract symmetry for peer to peer.
- Dual personality Initiator/Target roles also need to be hidden from
applications.

( ) Other (describe)__________________

2. Please provide background comments on your ranking.

Why do you prefer to use the given solution?
Why should others consider the given solution?
Does the given solution meet the existing requirements?
What issues are you aware of (if any) with the given solution?

What is your opinion on the best way to move the discussion forward?
o Decide. Implement. Have interoperability sessions to refine our profile based
upon what we learn.


Other comments?

----

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 ?

o Believe it to have a PWG scope.

Also,
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
application.
( ) 2 protocols are ACCEPTABLE if necessary for different situations
( ) I PREFER more than 2 protocols
( ) Other (describe)__________________

o Would prefer 1 or 2 protocols. As it turns out we will probably have

1) IP1394.
2) PWG protocol
3) PWG-C protocol.
4) AVC? Why was this mentioned as a separate item?