|
Hylafax Mailing List Archives
|
[Date Prev][Date Next][Thread Prev][Thread Next]
[Date Index]
[Thread Index]
Re: [hylafax-users] large multiport systems
Lee,
Your recent in-depth work on ECM, and
discussion/conclusions on ECM vs. non-ECM throughput offers some very
interesting food for thought, thanks! We'll study your comments, and get back to
you in due course. We're also going to spend some time doing real-world
measurements in collaboration with several of our fax broadcast customers, and
see how our practice matches up with your theory.
There are a couple of V.34 advantages I'd
like to highlight/excerpt from that body of "propaganda" you love to hate that
you might want to take into account, since some of them are at least partially
relevant to even single-page faxing:
1. V.8 fast handshaking only occurs when
both sender and receiver support it, so there's no time lost trying to
negotiate this capability.
2. V.8 fast handshaking really is
(theoretically at least) significantly faster. With a 9.6 Kbps or a V.17 modem,
the handshaking is done at 300 bps. With V.8 (ie: when sender and receiver
support V.34) the handshaking is done at a much faster rate of 1,200 bps. The
result is that handshaking time is reduced from approximately 16 seconds with
9.6 Kbps and V.17 to 7 seconds with V.34.
3. V.34 implements a "line probing" procedure
that is much more efficient at retraining on noisy calls. Immediately following
the handshake stage, a signal exchange allows the V.34 receiver to analyze the
characteristics of the connection before beginning the data transmission stage,
and choose several key session parameters. This isn't only performed on every
new connection, but can occur at any time during the connection as part of the
retraining process, which can be a huge win on multi-page faxes.
Consider the following table of average fax
transmission times. I think it's safe to assume the author used ECM throughout,
and that the numbers are for relatively healthy line conditions (Source:
Davidson Consulting, 2003):
9.6
kBps 4-Page Fax V.17 4-Page Fax V.34 4-Page
Fax
(seconds)
(seconds)
(seconds)
Handshake 16 16
7
Page
1 18
12
5
Retraining
6
6 0.25
Page
2
27 18 7
Retraining
6
6 0.25
Page
3
27 18 7
Retraining
6
6 0.25
Page
4 54
36
14
Retraining 6
6 0.25
==========================================================================
Total 166 124
41
Clearly V.34, with the quick handshake, faster data
transfer and super-quick line-probe retrains is very relevant to multi-page
faxing.
One strategy we recommend to our broadcast
customers who aren't too worried about 100% optimum quality on the receiving
end is to try doing V.34 with all capable receivers, but to do non-ECM V.17
or 9.6 KBps to the destinations that don't support it. In other words only do
ECM in V.34 calls that are so quick that it essentially comes for free
;-)
One interesting wrinkle whenever one speaks of ECM
is how hard the ECM implementation actually tries to recover from poor line
conditions (using a sequence of PPR/CTC cycles combined with bitrate reduction).
Clever manipulation of this "persistence" can make a large difference in some
ECM calls, as you have learned. If you're not careful, you can find 1-page ECM
faxes occasionally taking 15 to 20 minutes and still never get that
perfect fax sent! Some times it's just best to drop the call and move to the
next call, as you quite rightly point out.
Anyway, all very interesting and it bears careful
study & measurement. Fun stuff!
-Darren
----- Original Message -----
Sent: Wednesday, September 10, 2003 4:31
PM
Subject: Re: [hylafax-users] large multiport
systems
> On 2003.08.26 10:24 Darren Nickerson wrote:
>
> > I
may be wrong here Lee, but my research tells me you should look for
> >
another culrprit if you're looking to improve your V.34 broadcast
> >
performance.
>
> I have suspicions that V.8 handshaking functions
slightly different
> when using digital equipment than when using analog
equipment. The
> only equipment of which I am aware that supports
V.34 faxing with
> analog equipment comes from either MultiTech or
BrookTrout. As HylaFAX
> (open source) only supports the former, I
have no reason to blame the
> increase in handshaking time when V.8 is
attempted (either by the
> receiver, by the sender, or by both) on the
manufacturer's
> implementation. When HylaFAX truly supports Class
1.0 then maybe I'll
> be in a position to draw conclusions that
way. In any case, the
> increase in handshaking time of which I
have spoken is "only" on the
> order of 1-3 seconds per call.
Almost any significant amount of image
> data will easily offset this
delay. I was wrong in concluding
> otherwise.
>
> The
other culprit of which you speak is ECM protocol.
>
> Usage of V.34
requires the usage of ECM protocol. ECM protocol
> actively strives
for perfect image quality. Non-ECM protocol only
> strives for
near-perfect quality, depending upong the copy-quality
> settings of the
receiver. So the likelihood that retransmission of
> image data
(due to PPR) will occur in an ECM fax is much greater than
> the
likelihood that retransmission of image data (due to RTN) will
> occur in
a non-ECM fax.
>
> It is inaccurate and misleading to compare ECM
faxing against non-ECM
> faxing (or vice-versa) due to this difference in
goals.
>
> I've seen lots of marketing material out there
(propaganda?) that tries
> to sell people on purchasing new hardware,
software, or both because
> making that investment into V.34 and
MMR-supporting equipment will "pay
> for itself over time in toll savings
and necessary lines". This is
> only always true if we are only
considering ECM faxing. It is only
> sometimes true (and in my
experience, more often it is untrue) when
> also considering non-ECM
faxing.
>
> A fax sent without ECM at 14400 bps with MR compression
can, when
> enough noise exists to put image quality less than perfect
but still
> "near-perfect" (and this is a very common situation), be sent
faster
> than a fax sent at 33600 bps with MMR compression (with
requisite ECM,
> of course) because with ECM faxing image data had to be
resent,
> sometimes repeatedly, to acheive perfect image quality while
non-ECM
> faxing completed successfully without image data
retransmission.
>
> I suspect that most people are going to find
the value in ECM faxing to
> be worth it. However, I also suspect
that there are going to be some
> people, particularly bulk faxers, which
will think otherwise - because
> in non-ECM faxing if the receiver sends
MCF rather than RTN, that means
> that the image quality was "good
enough" for them... and the "savings"
> of not using ECM is better spent
sending more faxes out.
>
> Lee.
>
