[hylafax-devel] Re: config/usr-xon & usr-rts

[Dmitry Bely <dbely@mail.ru>]

Steve Williams <steve@genie96.com> writes:

> Some modems MUST be in class 0 in order to do adaptave answering.

Perhaps it is really so (could you name any such modem?), but the standard
thinks quite different:

[---cut---]
8.5.2.5 	Adaptive Answer, +FAA
	Write syntax:	+FAA=<value>
	Valid values:	0, 1
	Default value:	0
	Mandatory value: 	0
 	+FAA=0	The DCE shall answer only as a Class 2 facsimile device.
No automatic switching of service class shall occur based on the calling
device type.
	+FAA=1	indicates the DCE can answer and automatically determine
whether to answer as a facsimile DCE or as a data modem.  If a data modem
is detected, the DCE shall operate as described in section 8.3.2.4.

Note: The +FAA parameter controls automatic switching only from Class 2 to
Class 0, for call answering only.  It does not affect call origination,
switching to Class 2 from other classes, or switching to classes other than
Class 0.
[---cut---]

But it has nothing to do with the discussed case. After

ModemSetupAACmd:	AT+FCLASS=0	# leave modem idling in class 0
ModemAnswerCmd:		AT+FCLASS=1A	# answer in Class 1

adaptive answer will anyway *not* work!

Even after 

ModemSetupAACmd:	AT+FCLASS=0;+FAA=1 (or +FAA=1;AT+FCLASS=0)
ModemAnswerCmd:		AT+FCLASS=1A       (or ATA)

it won't work also (the modem *may* switch to Class2, but not to Class1!)

> If they idle in class 1, then they will ONLY answer as a fax, never do 
> the adaptive answer to determin if it's a data or a fax connection.

There is *no* standard way to implement Class0/Class1 answer in the modem
itself (although there are semi-reliable software-only ways), no matter in
which class you are waiting for a call.

> I don't know if that is the case for the USR modems, as they are such
> crap I have never bought one!

They *work* *good* in Class1. 

> > Anyway does anybody understand an idea behind the quoted part?

I still consider this config (and many others with the same technique) as a
bug, and nothing else.

BTW, I believe that adaptive answer should *not* be enabled in the supplied
configs by default. Currently this is not true for many templates:

[---cut---]
bash-2.02$ grep "AT+FAA=1" *
att-dataport-2:ModemSetupAACmd:	AT+FAA=1	# enable adaptive-answer in Class 2
att-dataport-2.0:ModemSetupAACmd:	AT+FAA=1	# enable in current class
exar:ModemSetupAACmd:	AT+FAA=1	# enable in current class
gvc-288-2:ModemSetupAACmd:	AT+FAA=1	# enable adaptive-answer in Class 2
moto-288:ModemSetupAACmd:	AT+FAA=1
mt-1432:ModemSetupAACmd:	AT+FAA=1	# enable adaptive-answer in class 2
ppi-pm14400fxmt:ModemSetupAACmd:	AT+FAA=1	# enable in current class
rc144ac:ModemSetupAACmd:	AT+FAA=1+FCLASS=0 # enable in current class
rc288dpi:ModemSetupAACmd:	AT+FAA=1	# enable adaptive-answer in Class 2
rockwell:ModemSetupAACmd:	AT+FAA=1+FCLASS=0 # enable in current class
sierra:ModemSetupAACmd:        AT+FAA=1    	# adaptive answer
telebit-t3000:ModemSetupAACmd:	AT+FAA=1;+FCLASS=0 # enable in class 2 & reset to data
telebit-wb:ModemSetupAACmd:	AT+FAA=1;+FCLASS=0 # enable in class 2 & reset to data
umc92144:ModemSetupAACmd:        AT+FAA=1
usr-2.0:ModemSetupAACmd:	AT+FAA=1
zyxel-1496e:ModemSetupAACmd:	AT+FAA=1	# enable in current class
zyxel-1496e-1:ModemSetupAACmd:	AT+FAA=1	# enable in current class
zyxel-1496e-2.0:ModemSetupAACmd:	AT+FAA=1	# enable in current class
zyxel-2864:ModemSetupAACmd:	AT+FAA=1	# enable in current class
[---cut---]

A newbie will be confused, while the expert can easily append necessary
parameter manually.

I am going to remove (at least comment out) all these ModemSetupAACmd.

Again, is there any objection?

Hope to hear from you soon,
Dmitry


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