Re: [hylafax-users] note on new firmware of topic chipset (+FTS)

[Lee Howard <faxguy@deanox.com>]

On 2002.05.20 09:32 Giulio Orsero wrote:

> May 20 15:28:51.39: [ 7863]: --> [2:OK]
> May 20 15:28:51.39: [ 7863]: <-- [9:AT+FTS=7\r]
> May 20 15:28:51.41: [ 7863]: --> [2:OK]  <=== OK, but 70 ms have not
> elapsed
> ...
> May 20 15:32:30.26: [ 7863]: SEND end page
> May 20 15:32:33.87: [ 7863]: --> [2:OK]
> May 20 15:32:33.87: [ 7863]: <-- [9:AT+FTS=9\r]
> May 20 15:32:33.89: [ 7863]: --> [2:OK]  <=== OK, but 90 ms have not
> elapsed
> May 20 15:32:33.89: [ 7863]: SEND send EOP (no more pages or documents)
> May 20 15:32:33.89: [ 7863]: <-- [9:AT+FTH=3\r]
> May 20 15:32:33.94: [ 7863]: --> [7:CONNECT]
> ===
> 
> So, modem prototype is still correct because, even if +FTS gives OK, it
> seems it does not really work (it just says OK), but it may be confusing.

I've not got a Topic modem to test (can't seem to find them in the US), 
but my guess would be that we'd need to test things more before we can 
conclusively say that the new Topic firmware is still flawed when it comes 
to +FTS.  The testing would just be to use HylaFAX in a large scale with 
+FTS instead of software pauses and look for MCF reception errors.  
(Unless you have a fax protocol analyzing device to stick on the other end 
which can tell us the actual length of silence.)

I say this because I've seen many modems which do not respond "OK" to 
+FTS=n after n*10 ms.  Lucent chipsets, for example, consistently seem to 
wait 20 ms more than expected.  After talking with MultiTech technicians 
regarding it, I came to understand and believe that +FTS=n is 
"intelligent" in that it only reports "OK" after the DCE has been silent 
for n*10 ms and not specifically n*10 ms after the +FTS=n command was 
sent.  In the Lucent case, the explanation is that the carrier flags and 
such require time before they are completely dropped, thus adding a 
consistent 20 ms to the expected delay.

In the case of the Topic, I suppose that it could be possible, although 
seemingly unlikely, that the modem is not reporting OK after "SEND end 
page" until the carrier flags and such have already been dropped for 50-70 
ms.

Lee.

____________________ 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@hylafax.org < /dev/null