[RTTY] A repository of (poor) RTTY recordings?

[Bill Turner]

On Tue, 20 Apr 2004 13:53:25 -0700, Kok Chen wrote:

>It is perhaps useful for the RTTY community to have available a 
>"standard" collection of digitized sound files of various RTTY signals. 
>  Not the clean loud stuff that we already have on the web as examples 
>of FSK, but some watery fluttery hard to copy, etc. stuff.
>
>All of us could then use the same digitized sound files to play back 
>into whatever system we are testing so there would be some consistency.

_________________________________________________________

Would it be a good idea to come up with a standard text message we could
all use for this?  I'm thinking if we knew in advance exactly what was
sent it would make it possible to count the errors on a particular
transmission and thereby have some hard numbers to compare various
filter and demodulator settings.  If we had numerous examples of the
same text, sent under many different propagation conditions, we could
run them through our different programs and TNC's and begin to come up
with some meaningful statistics.

As a first stab at it how about:

RYRYRYRYRYRYRYRYRYRYRYRYRYRYRYRYRYRYRYRYRYRYRYRYRY
RYRYRYRYRYRYRYRYRYRYRYRYRYRYRYRYRYRYRYRYRYRYRYRYRY
RYRYRYRYRYRYRYRYRYRYRYRYRYRYRYRYRYRYRYRYRYRYRYRYRY
RYRYRYRYRYRYRYRYRYRYRYRYRYRYRYRYRYRYRYRYRYRYRYRYRY

For a total of 200 characters, not including the carriage returns.  This
is visually easy to scan for errors and then just count them.  For a
given file, one might get five errors with one setting, four errors with
a different setting, zero with a different one and so on.  You could
then use a different file with different amounts of flutter and
multipath and repeat the experiment.  Then yet a different file, and so
on.

As an alternative, perhaps blocks of text might be easier to count:

RYRY RYRY RYRY RYRY RYRY RYRY RYRY RYRY RYRY RYRY
RYRY RYRY RYRY RYRY RYRY RYRY RYRY RYRY RYRY RYRY
RYRY RYRY RYRY RYRY RYRY RYRY RYRY RYRY RYRY RYRY
RYRY RYRY RYRY RYRY RYRY RYRY RYRY RYRY RYRY RYRY
RYRY RYRY RYRY RYRY RYRY RYRY RYRY RYRY RYRY RYRY
RYRY RYRY RYRY RYRY RYRY RYRY RYRY RYRY RYRY RYRY
RYRY RYRY RYRY RYRY RYRY RYRY RYRY RYRY RYRY RYRY
RYRY RYRY RYRY RYRY RYRY RYRY RYRY RYRY RYRY RYRY
RYRY RYRY RYRY RYRY RYRY RYRY RYRY RYRY RYRY RYRY
RYRY RYRY RYRY RYRY RYRY RYRY RYRY RYRY RYRY RYRY

For a total of 100 blocks.  Then just count the correctly received
blocks and that would be the score.  100 = perfect.

These are just my first couple of thoughts on the subject.  There are
probably other text groups that would be easier to score.

Comments?

--
Bill, W6WRT
QSLs via LoTW