Call Search
     

New to Ham Radio?
My Profile

Community
Articles
Forums
News
Reviews
Friends Remembered
Strays
Survey Question

Operating
Contesting
DX Cluster Spots
Propagation

Resources
Calendar
Classifieds
Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers

Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement
About eHam.net

   Home   Help Search  
Pages: [1] 2 Next   Go Down
  Print  
Author Topic: JustLearnMorseCode  (Read 4224 times)
PA0BLAH
Member

Posts: 0




Ignore
« on: September 01, 2012, 06:38:26 AM »

A  well known program, this topic title.

Thanks to the maker, he makes my day, with his ten long toes and only eight fingers. Because his thumbs are no fingers but his largest 2 toes are toes, nobody denies.  Large toes are a delightfull possession of my companion radio amateurs, because it is easy to jump on them, and I like to do that. Kind of hobby, just like ham-radio, same grade of stupidiness.  I fully agree.

I was questioning what the leading and trailing edges were, as the sound arrives at my ear, from the program "Just LearnMorseCode" adjusted to 40 wpm. Kind of music, I like that, especially when I load the file KamaSutra (Sanskrit: कामसूत्र ) to play in Morse code.

I am not using headphones, in order to stimulate interest in Morse code at NOL  (NOL=now old lady, she was XYL ex young lady once, abt 70 years ago).
Is hard to perceive, she once said, she was interested to learn Morse code, but the condition she formulated was, that I had to learn knitting and embroidery,

That learning the code is important, because when I get a stroke, I can order a cup of coffee just by winking my eyebrows.
Now, when she gets a stroke, I have to keep coffee available and random pouring with a funnel in her mouth, not knowing she likes it or not.

Sometimes people ask: How come you are both growing so old "compos mentis" (lucky for you there is wikipedia and goofle)
I answer then: Well the Dutch: "waterwingebied", kind of source for the municipal water supply, with a controlled quality, that keeps the FICA payable) is according to the map, neighboring field of the previous municipal burying place (grave yard). And, you know, homeopathy is the "science" that pro clamates that every poison diluted enough, is the cure for sickness due to that poison.  Because it is alternative it is cheap and denied, but the non-alternatives know the principle very well by giving you shots against flue, tetanus, small pocks,  and what have you. And In the municipal water is the poison.highly diluted, of all that buried bodies. And indeed We, NOL and I, are no exception in this local area.

So listen to the talk of an old wise  compos mentis man.

Start living close to a grave yard, get your water out of the local soil, and that's it. That means that a lot of people wishing you under the soil, like K8AXW and Stayvertical, are disappointed. Let it be.

Now, the point of this posting: I measured a leading edge that has OVERSHOOT, and a trailing edge, and that makes me formulate this post, that is 7.5 ms BUT IT HAS A JUMP OF 180 DEGREE IN PHASE, at he point it is glued at the trailing of the code element.

Never read about that,

73 PAoBLAH

Logged
LB3KB
Member

Posts: 227


WWW

Ignore
« Reply #1 on: September 01, 2012, 05:12:33 PM »

Now, the point of this posting: I measured a leading edge that has OVERSHOOT, and a trailing edge, and that makes me formulate this post, that is 7.5 ms BUT IT HAS A JUMP OF 180 DEGREE IN PHASE, at he point it is glued at the trailing of the code element.

Never read about that,

So, does it sound good ?

73
LB3KB Sigurd
justlearnmorsecode.com
Logged
PA0BLAH
Member

Posts: 0




Ignore
« Reply #2 on: September 01, 2012, 05:34:29 PM »


So, does it sound good ?


No problem at 40 or 45 wpm. Is it purposedly done, or did I make a blooper in measuring?
Logged
LB3KB
Member

Posts: 227


WWW

Ignore
« Reply #3 on: September 01, 2012, 05:56:40 PM »


So, does it sound good ?


No problem at 40 or 45 wpm. Is it purposedly done, or did I make a blooper in measuring?

The generated waveform is very clean, designed specifically to avoid the blips/clicks/noise some of us are able to hear when lesser software generates Morse code.

When measuring it, you're relying on a number of other factors such as your sound card, drivers and speakers as well as your equipment for and method of measuring.


Are you using the latest version (1.23) ?


73
LB3KB Sigurd
justlearnmorsecode.com
Logged
PA0BLAH
Member

Posts: 0




Ignore
« Reply #4 on: September 01, 2012, 06:23:07 PM »



The generated waveform is very clean, designed specifically to avoid the blips/clicks/noise some of us are able to hear when lesser software generates Morse code.

When measuring it, you're relying on a number of other factors such as your sound card, drivers and speakers as well as your equipment for and method of measuring.


Are you using the latest version (1.23) ?


Yes, 1.23
It is obviously speed dependent, I record the sound coming out of the PC, may be reverberation of the speaker, echo, don't know.
It does not bother me, the sound is excellent, I was just wondering whether or not it was purposely done. At 40 wpm the trailing shows a phase reversal of 180 degrees.

I can construct  a wav file that is error free, by just calculating with a sine function the samples. When I play that file in the PC I can use it as a reference.

Thanks for reply and for the opportunity to use your excellent program.
gn 73 Bob


Logged
PA0BLAH
Member

Posts: 0




Ignore
« Reply #5 on: September 02, 2012, 02:43:14 AM »


I can construct  a wav file that is error free, by just calculating with a sine function the samples. When I play that file in the PC I can use it as a reference.


Done that. realizing afterwards that I made a cosine squared leading and trailing edge of 2 ms each, but in the signalmark time. That means that the mark starts at zero grows in 2 ms to full amplitude and 2 ms before the theoretical end of the dot or dash, start the trailing slope. The theoretical value of the mark timespan is hence measured at zero value of the signal envelope. So it should be better to lengthen the dot time compared with the dot space with 2 ms, with the result that at half amplitude level of the signal envelope the dot or dash time is the correct value. At the bottom 1 ms wider and at the top one   ms smaller.

Result by playing the constructed wav file in mediaplayer and recording the sound at 40 wpm: roughly the same, not a complete phase reversal, but clearly extended zero crossing at the start of trailing edge, and also a pretty long reverberation.

tks es 73 Bob
Logged
PA0BLAH
Member

Posts: 0




Ignore
« Reply #6 on: September 05, 2012, 02:47:53 AM »


I can construct  a wav file that is error free, by just calculating with a sine function the samples. When I play that file in the PC I can use it as a reference.

Done that. realizing afterwards that I made a cosine squared leading and trailing edge of 2 ms each, but in the signalmark time. That means that the mark starts at zero grows in 2 ms to full amplitude and 2 ms before the theoretical end of the dot or dash, start the trailing slope. The theoretical value of the mark timespan is hence measured at zero value of the signal envelope. So it should be better to lengthen the dot time compared with the dot space with 2 ms, with the result that at half amplitude level of the signal envelope the dot or dash time is the correct value. At the bottom 1 ms wider and at the top one   ms smaller.

Sigurd,

When I look at a recorded wav file generated by your program: You make the same blooper.

When Morse is seen as a series of periodic dots, and I limit the bandwidth at the utmost, it degenerates to an 100% amplitude modulated sinewave, reaching the top at the middle of the dot and the bottom at the middle of the dotwide space.
So the cosine squared leading edge is starting in the middle of the signal space and he trailing edge ends in the middle of the space.

Hence the signal timing has to be measured at half the height of the max amplitude; there the timing dot/space is undistorted. What you are doing is taking the ideal time, and let the leading and trailing slope "eat" only from the mark. That is wrong, when you detect the envelope of that signal the mark/space ratio is going lower for higher speeds.

In order to correct that you have to lengthen your ideal marktime with the slope time and to shorten one space dit after the mark with the slope time of your supposedly used 4 ms transition time.

tks es 73 Bob
Logged
LB3KB
Member

Posts: 227


WWW

Ignore
« Reply #7 on: September 05, 2012, 03:39:48 AM »

Your thinking is too complicated.

Did you ever look at what your transmitter's keyer does ?


73
LB3KB Sigurd
justlearnmorsecode.com
Logged
PA0BLAH
Member

Posts: 0




Ignore
« Reply #8 on: September 05, 2012, 04:17:05 AM »


Did you ever look at what your transmitter's keyer does ?


Sure, look at my list of requirements of "the best external keyer". Born by the fact I am a home brewer, not an appliance buyer.

Because your transmitter can not know in advance that a dot or dash is starting, it can not start it in time, such that it is half height at the signalelement starting time.

So a good keyer knows the transmitter is distorting (shortening) his marks with the leading edge time, just like you are doing in your wav files, and for that reason a good keyer has the possibility to lengthen the marks with the slope time of the transmitter and hence shortens the space time with that amount.

To adjust it, you put your antenna output connector of your appliance to a dummyload, measure with a moving coil meter and a diode detector over the dummyload. Adjust the meter to full scale, with carrier on. Give a number of dots with your keyer at 100 wpm or over that speed, and adjust the mark space (weighting) of your keyer such that the needle of the meter is just in the middle.

It is not  nitpicking or (translated Dotch proverb: looking for nails in low water) because with your timing a meter will point at 100 wpm on 30% instead of 50% full scale.
« Last Edit: September 05, 2012, 04:43:15 AM by PA0BLAH » Logged
LB3KB
Member

Posts: 227


WWW

Ignore
« Reply #9 on: September 05, 2012, 04:42:59 AM »

What do your ears tell you ?


73
LB3KB Sigurd
justlearnmorsecode.com
Logged
PA0BLAH
Member

Posts: 0




Ignore
« Reply #10 on: September 05, 2012, 04:45:29 AM »

What do your ears tell you ?


73
LB3KB Sigurd
justlearnmorsecode.com
Translated Dutch proverb: When the owl does not want to see it, glasses and a candle want help.

Bob
Logged
LB3KB
Member

Posts: 227


WWW

Ignore
« Reply #11 on: September 05, 2012, 04:58:59 AM »

Who in their right mind would want to see Morse code ?

Once again - What do your ears tell you ?


73
LB3KB Sigurd
justlearnmorsecode.com
Logged
PA0BLAH
Member

Posts: 0




Ignore
« Reply #12 on: September 06, 2012, 01:27:04 AM »


Once again - What do your ears tell you ?

73
LB3KB Sigurd

My ears tell me that it sounds terrific, indeed. When people are using ur pgm to exercise QRQ, they don't get the machine generated code they expect at 100 wpm.

There exist no complex bugfree software, is a general belief, but never before I have met a programmer that refuses to recognise his bloopers and plans to take adequate action to correct them.

Make, like I did, a wav file with your program at 100 wpm high pitched tone. change in the RIFF header the big ended 4 byte integer samplerate from 22050 in 8000, play the file, the speed is then 36 wpm, so you can decide for yourself.

Bob
« Last Edit: September 06, 2012, 01:45:03 AM by PA0BLAH » Logged
LB3KB
Member

Posts: 227


WWW

Ignore
« Reply #13 on: September 06, 2012, 02:37:27 AM »

I am well aware of how the audio generation works in Just Learn Morse Code, at any speed.

If your concern is how it works with unrealisticly high pitched audio generated at 100 WPM played back at 35 WPM then perhaps you should have said so at the start, rather than going on this wild goose chase.

I do listen to feedback and I do modify the program according to that feedback.  That doesn't mean I incorporate every suggestion I come across, but I have implemented quite a few of them.  It could also be that I would modify the behavior of the software for that tiny group of users that are interested in using it at ridiculous speeds, but I would only consider that after hearing from some of them.

In general, that does not include what anonymous cowards artificially construct and throw out in public on the Internet, though.
« Last Edit: September 06, 2012, 02:41:08 AM by LB3KB » Logged
PA0BLAH
Member

Posts: 0




Ignore
« Reply #14 on: September 06, 2012, 04:51:50 AM »

I am well aware of how the audio generation works in Just Learn Morse Code, at any speed.

If your concern is how it works with unrealisticly high pitched audio generated at 100 WPM played back at 35 WPM then perhaps you should have said so at the start, rather than going on this wild goose chase.

No, I handed a simple and fast way to listen to the quality of the generated code, in the case one  does not master that high speed. I did it because you invited me twice in this thread to tell what my ears say.

When you think 100 wpm is ridiculous, as you say, then I ask, why did you implement it, any higher speed you type in  is reduced to that max, so it is purposely made for 100 wpm max speed.

It is not that ridiculously high, as you say, DJ1YFK listens to e-books converted to that speed in an MP3 file. And look for further information at https://sites.google.com/site/tomw4bqf/copyingcwover70wpm

Better you also implement MP3 recordings, instead of WAV files, that are uncompressed and huge in size, special with 22050 16 bit samples/s, as you implemented,  when 8000 samples/s 8 bit per sample are for Morse code recordings perfect.  

G4FON produces also MP3.

You can use the open source LAME encoder.

Quote
I do listen to feedback and I do modify the program according to that feedback.  That doesn't mean I incorporate every suggestion I come across, but I have implemented quite a few of them.


My comment about the leading and trailing edges is not a suggestion to incorporate a feature, it is the detection of a BUG, especially when you announce that your by your pgm produced soundfiles are of the highest quality compared with others, as you did, it is according to my opinion not to  justify to refuse to clean it up, especially not when I showed you how the signal can be tested in quality on reduced time scale.

Quote
It could also be that I would modify the behavior of the software for that tiny group of users that are interested in using it at ridiculous speeds, but I would only consider that after hearing from some of them.

In general, that does not include what anonymous cowards artificially construct and throw out in public on the Internet, though.

It is not to justify to mention anybody in this forum an anonymous coward, it is perfect allowed behavior to present yourself so, it prevents identity theft, leaking away your privacy, while a call sign is the way you have to announce in a prescribed way by government the radio transmission you perform as allowed in a license. That has nothing to do with the identification here, just as an email address and a telephone number and an exact QTH has nothing to do with that.

In short a call sign is required in the aether, in this forum it is not a call sign but a User-ID that counts. May even be it looks like a call sign, such as ID1OT, but its not.

It is not important WHO indicates you that you make bloopers in software development, but it is important that it is said. That is my humble opinion.

Bob
« Last Edit: September 06, 2012, 05:00:48 AM by PA0BLAH » Logged
Pages: [1] 2 Next   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.11 | SMF © 2006-2009, Simple Machines LLC Valid XHTML 1.0! Valid CSS!