Re: [hylafax-users] Pointer to Class1 vs. Class2 discussion?

[Lee Howard <faxguy@deanox.com>]

At 10:39 AM 9/4/01 -0500, Steve Snyder wrote:
>I recently read a message on this list that references a previous 
>discussion on the relative merits of Clas1 and Class2 protocols.  I've 
>looked through the list but I can't find that discussion.  Can someone 
>point me to the approximate date when this discussion took place?

This discussion has come up a lot.  Most of the good discussion, as Bernd
has mentioned was on the -devel list.  But you're probably referring to
this post:

http://www.hylafax.org/archive/2001-08/msg00420.html

>Apparently the consensus is that Class1 is the preferred protocol.  To me 
>this is counter-intuitive, because the Class2 protocol provides hardware 
>support for functions that must otherwise be performed in software (by the 
>host CPU) using Class1.  I would like to understand why, given a device 
>that supports both classes, I should prefer the older protocol.

Older doesn't mean less functional.  And newer doesn't always mean better.

People get worried about CPU usage with Class 1.  Frankly, I don't know why
because any 486 system will adequately handle this type of CPU usage for
several modems.

The biggest problem with Class 2/2.0 is that the protocol code is in the
firmware, and we can't touch it.  So, to fix a bug you either have to
contact the manufacturer and coax them into fixing the firmware (this is
tough, I've been there), or you have to write a workaround in the faxing
software to alleviate the problems of the firmware bug.

So, when you're talking to a bunch of developers who do work on HylaFAX
code for free and their own benefit, you're not likely to see any of us
exert a whole lot of effort for a Class 2 workaround when it's just easier
and as functional to switch to Class 1.

Lee.



____________________ HylaFAX(tm) Users Mailing List _______________________
 To unsub: mail -s unsubscribe hylafax-users-request@hylafax.org < /dev/null