Re: [hylafax-users] Hylafax+ With E1 card possible ?

[Steve Underwood <steveu@xxxxxxxxxxx>]

Andrew Rinaldi wrote:
> Hi Steve,
>
> Most of the chipsets used within the V.29/V.17 fax machines are no longer
> available which is why I expect to see an upsurge in new V.34 fax machines,
> with more vendors moving to a V.34 only strategy.  I also find it
> interesting that the latest Davidson report states that in 2005 83.9% of fax
> boards sold were V.34 capable.
>   
Most FAX machines are Japanese. Most Japanese FAX machines use locally 
produced chipsets, by makers most people wouldn't recognise as modem 
suppliers - e.g. TDK/Teridian. They still make plenty of low end modems. 
It wouldn't surprise me to find V.34 models dominating FAX cards, but 
that is almost unrelated to FAX machines. Some FAX modems seem very 
supplier specific - e.g. most recent Canon FAX machines have totally 
broken V.29 transmit, which doesn't seem comparable to any other make of 
FAX machine. They appear to either use a unique modem, or load it with 
special broken DSP code.

> I understand from one of the T.38 contributors that a number of people
> believed that V.34 was unlikely to work reliably over T.38 which is why they
> left V.34 'for further study' in the 2002 specification.  I've also had a
> conversation with one of the IP fax server vendors who said that the Cisco
> V.34 solution was not 'robust' (although to be fair they also reported
> constant change with the Cisco V.17 firmware so maybe it's not V.34
> specific).  One final point, V.34 is a very litigious area and this might
> discourage T.38 vendors from implementing V.34 in the gateways.
>   
I don't know why there would be litigation issues. If you use a licenced 
V.34 modem, you would appear to be in the clear. There are some bogus 
patents related to T.38, including some bizarrely stupid ones 
progressing through the system. E.g. a 2005 Hua Wei application for 
corrupting NSF messages, that describes exactly what every T.38 gateway 
has done since T.38 was first published. The thing about reliability is 
an issue, though. The T.38 2002 spec included messages for V.34, but 
seemed to leave their use a bit vague. More recent versions have become 
more focussed. I'm not clear why people expect V.34 over T.38 to be 
especially flaky, though many T.38 gateways disable ECM as supposedly 
troublesome. ECM isn't troublesome - their networks are.

Steve


____________________ HylaFAX(tm) Users Mailing List _______________________
 To subscribe/unsubscribe, click http://lists.hylafax.org/cgi-bin/lsg2.cgi
On UNIX: mail -s unsubscribe hylafax-users-request@xxxxxxxxxxx < /dev/null
 *To learn about commercial HylaFAX(tm) support, mail sales@xxxxxxxxx*