[Asterisk-Users] Which asterisk-friendly cards are fax-capable?

[Lee Howard]

Darren Nickerson wrote:

> I'm curious though, in an earlier message you wrote:
>
> "Sending is also quite  good, however, there are some quirks with the 
> Patton
> firmware which need to be resolved for me to be completely delighted."
>
> Are these issues you refer to just cosmetic 'nice to haves' or do they 
> actually impact the outbound success rate? Is the 2977's outbound in 
> Class 1 as solid as the Mainpine or Multitech boards now? Our 
> experience has shown that is not the case ...


Having worked extensively with modems such as MultiTechs and MainPines 
which use the Lucent/Agere Venus chipset, modems such as Comtrol 
RocketModems (II not I or III) that use the Rockwell/Conexant K56 
chipset, and the Patton 2977 (which along with the Digi AccelePort RAS 
uses an ADI chipset) I can say unequivocally that the former two _can_ 
currently outperform the latter.

But allow me to put that performance difference into perspective ... 
depending upon the locale I can see 99.9% success rates (ratio is pages 
without error divided by total pages received) with MultiTechs, 
Mainpines, and Comtrols.  By comparison, the performance that I see with 
the Patton 2977 averages somewhere between 99.7% and 99.8% success 
ratio.  That said, I have seen deployments of the former two chipsets 
underperform those statistics under "normal" conditions (say 98% to 99%) 
simply because they frequently send to or receive from a very 
problematic fax machine on the other end... and still the owners of 
those fax servers are quite happy with how those perform.  In fact, I 
know of HylaFAX deployments that use USR modems and suffer at around 95% 
success ratio, and the problems even there aren't enough to motivate 
them to buy better modems.  So the performance of the Patton 2977 
shouldn't be considered "bad"... just not as good as I think it should 
be.  Furthermore, when we're talking about digital phone lines and 
digital fax equipment, I should think that could eliminate 50% of the 
potential pitfalls right there... and if the other end is also using 
digtial then communication should be flawless, right?  Well, that's my 
idea anyway... it doesn't work out that way.

I'm a finicky faxer.  One error in a thousand pages may seem 
extraordinary to some, but when thousands upon thousands of pages are 
handled every day the 0.1% is annoying if not frustrating.

I could elaborate on the specific flaws in the Patton cards, I suppose, 
but this isn't the time or place.  I suppose that I could also elaborate 
on flaws in just about any other fax modem hardware that I've used 
extensively, too.  However, the trick isn't so much finding and using a 
flawless piece of equipment (because there isn't such a thing), but 
rather learning to avoid, fix, and deal with the issues as they appear.  
And this is why I have such a hard time favoring fax Class 2 and to a 
greater extent fax Class "proxies" like the Diva Server... because they 
take away so much of the control that is necessary to avoid, fix, and deal.

My hope in IAXmodem is merely to extend that aspect of control into the 
DCE and DSP.

Lee.