psk or PSK
Duncan McMahon <zl3jt@...>
de Jungle Telegraph....Gidday!
I notice there is a penchant by some to type their psk text in the upper case(?)
I suppose I could call myself 'experienced' with this mode and I would like to tell
those that do this 'lazy' typing, that it is more prone to faults if this is psk
mode is done this way...
It takes a little bit longer for psk31 to send a CAPITAL LETTER, and if conditions
are marginal (QSB/QRN/QRM) then this method has a greater error rate. That is it
may be the difference between making it or not making it to the DX station.... Watch
carefully when you know a 'macro' is being sent to you....
To test my theory I typed the same macro in both upper and lower cases...it is a
surprising difference in speed!
You really don't need to try to keep ahead of the buffer...
;-)....Also when trying to 'break' into a QSO, just set a macro with only your callsign
once... then TX in the gap! If you send a long call you will probably be 'doubling'
with someone else and computers can't figure that out!
I do it from far, far away with a fair bit of success in the pile-ups! pssst! Don't
happy pskin' in 2k2!
Advertise with ZFREE - to find out more click below
Dave Kjellquist, WB5NHL
Thanks for the tips on psk operation. I learned just a little while back that lower case was faster and a little
less prone to errors. I never run over 5 watts with only indoor dipoles so I need all the help I can get. I have a
couple more tips:
1. Use a spectrum display rather than a waterfall. I can detect and communicate with much weaker signals.
Often, I get a comment that my signal can't be seen on the waterfall yet we will carry on a QSO. Perhaps not
armchair copy, but for errorless copy use IRC.
2. Kill the squelch. I use Hamscope on the "contest mode" only. Yes, I get random characters etc but this is the
same as listening to hiss as your tuning around in SSB or inband noise when your going after that weak one.
Squelch only causes the loss of the weak signals.
Just some thoughts from one of the little guys.
It takes a little bit longer for psk31 to send a CAPITAL LETTER,
To test my theory I typed the same macro in both upper and lowercases...it is a
surprising difference in speed!
Duncan (and others) -
You are exactly correct. The actual speed of characters is dependent
on the distribution of letters in the text, but on average, capital
letters are 2-3 bits longer than lower-case in the Varicode used in
I think lower-case letters average about 5-6 bits each (averaged over
English text), and upper-case about 8-9. This has the two effects
1. At a rate of 31 bits per second, lower case will produce about 5
letters per second, versus 4 letters per second for upper case. This
means your message gets through faster.
2. You also only have to get 5 or 6 bits correct for the software to
decode a lower case letter, three less than for an upper case. For a
fixed bit error rate, this means the letter error rate will be lower
for lower case than for upper case.
Finally, there is the "email rule": capital letters are usually used
for emphasis (i.e., shouting), and are sometimes considered rude for