Jerry Jones wrote:
without ecm -
line errors will cause slight imperfections (dots) on transmitted
image
with ecm -
retry, retry, retry, fail
In the ECM retransmissions only the frames that were not received
properly are resent. Once the data to send is assembled by the sender
it is transmitted in exactly the same way as with non-ECM communications.
So as an example let's say that an ECM page consisting of a single block
of 100 frames is sent, and that a "slight" imperfection is detected, so
maybe 10 frames are bad: frames 5, 15, 25, 35, 45, 55, 65, 75, 85, and
95. ECM protocol will cause there to be three more retransmissions
before the sender has a deliberate chance to abort retransmissions. You
can probably see that the same "slight" imperfection would have a
difficult time persisting through three more iterations. However, for
the sake of argument, let's say that frames 5, 55, and 95 still aren't
received properly by the 3rd retransmission (4th attempt). The sender
then has the choice to change the speed of transmission and continue
another set of 4 retransmissions or to end retransmissions and (in
theory) move on to the next block.
Most fax machines will abort the call if they do not choose to continue
retransmissions... even if they do signal that they're going to move on
to the next block. So at that point the receiver is left with a page of
image data with some "very slight" imperfections in frames 5, 55, and
95. Now it's up to the receiver to be able to make something out of
that image. If the image type is MH or MR then the receiver is going to
be able to print out the full page... with maybe some dots as you say.
If the image type is MMR then the image will be truncated at the data
where frame 5 occurs (pretty much the top of the page). If the image
type is JBIG then the end result will depend upon the decoder... but
generally the decoders are lightly tolerant of some slight data
corruption, but too much corruption will be a problem.
So at the end of all of this contrived scenario the problem in not
getting a "slightly imperfect" image out of an ECM-using fax machine
would be due to the data type more than it woudl be due to the usage of
ECM. If you truly are experiencing the kind of, perhaps contrived,
situation where "retry, retry, retry, fail" occurs then I would expect
that disabling MMR+JBIG would be more effective than disabling ECM.
(Yes, disabling ECM does also disable MMR and JBIG, but you can use ECM
with MH and MR and thus benefit of a tolerant image type *and* a
tolerant protocol.)
If you continue to experience problems with ECM without MMR/JBIG then
you will (guaranteed) experience problems without ECM.
Lee.
_______________________________________________
--Bandwidth and Colocation provided by Easynews.com --
asterisk-users mailing list
To UNSUBSCRIBE or update options visit:
http://lists.digium.com/mailman/listinfo/asterisk-users
