My CW stinks......

[K7KBN]

What's a Q'?

QTH?  QRX?  QSL?  QSY?

Or do you mean "QSO"?  If so, spell it out; it's only two more keystrokes.

As far as "passing judgement", I can still send good code, and I can certainly recognize it when sent by another operator - and I can recognize bad code.  If I can get the gist of what the other guy is trying to say, I'll give it a go.  He, on the other hand, should be able to tell whether he's QSD.

[K9RFZ]

W3TTT made an excellent suggestion that seems to have been ignored.  Try sending code to a computer program that decodes CW. I use a practice mode on my IC-746 and feed the output through my digital interface to CWGet. I can see the waveform of my keying and make slight adjustments in timing and spacing until the program is decoding 100%. That becomes an objective judge of good code sent to spec.

I have to agree with another post in this thread that MOST fists stink. I've used CWGet to decode W1AW bulletins and practice runs with perfect copy, then swing over to a couple of 1x2 calls in QSO at the same S-meter strength, and the same band conditions totally turn to gibberish that CWGet can't decode. Call signs seem to be the sloppiest part of the QSO. Don't assume that because you've been operating CW for 40 years that your fist is all that good. Test your sending as W3TTT suggested and maybe you'll learn something that can really improve your sending. Sloppy sending won't pass the CWGet decoding self-test.

K9RFZ

[AK7V]

I don't think that producing perfect, machine-copyable code is the definition of a good fist.  I do it -- I've checked, and my sent code is certainly machine-copyable -- but what really matters to the other op (assuming they are an actual person doing the decoding) is whether they can understand easily.  Some variation from "perfect" is acceptable.  That's one of the great things about the human brain -- we can understand people with different accents and we can understand code with slight deviations.

That said, it could be a good "wake up call" to check yourself with decoding software, just to see how it looks.  Even better, record yourself sending some text from a book and then listen to it a few days later.  See if you can copy yourself.

Sometimes I enjoy the challenge of QSOing with an imperfect bug op or a CW newbie.  Works the brain.

[AG6EB]

Is it fair to say that if your CW is 100% copyable by any of the various software programs that translate CW, that you'll at least sound fairly decent to an experienced human?

when trying to increase my sending speed i've been frustrated by a tendency to get the timing off slightly when sending multiple dots - either one too many or one too few, probably from poor coordination (i'm not trying to count them, certainly)

[AK7V]

Is it fair to say that if your CW is 100% copyable by any of the various software programs that translate CW, that you'll at least sound fairly decent to an experienced human?

Yes, I think so.

[W5LZ]

The 'virtual' computer between your ears is much more forgiving than a 'real' computer.  So, if a real computer can understand the code being sent, then that 'virtual' computer shouldn't have any/much problem with it.  That's one way of determining if someone is sending by 'hand' or by keyboard!  Is that a 'biggy'?  Not most of the time, but it can be in particular instances.
 - 'Doc