User name:

Български Português brasileiro
Bosanski Català
繁體中文 Česky
Dansk Deutsch
English Español
Suomi Français
Ελληνικά Hrvatski
Magyar Italiano
日本語 한국어
Bahasa Melayu Nederlands
Norsk Polski
Português Română
Русский සිංහල
Slovenščina Srpski
Svenska ภาษาไทย
Türkçe Українська
Who is online? (14)

LCWO Discussion Forum [Atom LCWO Forum Feed]

This is a simple discussion forum for LCWO users. Feel free to use it for any kind of discussion related to this website.

Thread: HTML5 vs jscwlib

Back to the Forum


Posted: 2021-03-18 02:29
Hi all,
When playing with different settings, I noticed that HTML5 vs jscwlib seem to play at different speeds, despite the same settings.

If I go to:

Character Speed: 20
Effective Speed: 10
Extra Word Spacing: 0

Set CW Player to jscwlib

Hit submit... Now play the sample.

The Sample plays out 17 seconds long and sounds slow (like there is extra spacing between characters).

If I change to HTML5, Hit Submit. Now play the sample...
It's about 12 seconds

And these durations are from the start of the first dit to the end of the last dah. I actually recorded them with Audacity to view.

To further complicate things - if I download the audio files from the jscwlib setting and HTML5 setting - the both create MP3 files that are ~12 seconds (again from first dit to last dah)

Since the same settings are used on both players...and they sound very different - I don't know what I'm getting...

Anyone else experience this? Suggestions?

Thanks all!

Posted: 2021-03-18 10:02
Yes, I reported this some time ago:

Apparently, if you using custom characters, this will break the jscwlib player. I just keep on using the html5 player.

Posted: 2021-03-28 13:38
I did not time it but it seems that HTML is running at slightly less than twice the speed of jscwlib. I reduced the speed but that has messed with my records - not the end of the world though.

Posted: 2021-03-29 13:57
I am now confused. Which speed is correct, those using HTML or those using jscw?

Posted: 2021-03-30 13:15
Amoung other things, the idea was make the speed 'more correct' with the newer player (jscwlib) (please correct me if I am wrong).

As I recall the old player was fine with 'normal' morse speeds but the way it applied Farnsworth extra spacing was not as intended - with the overall effect that the speed of the old player (HTML) was higher than it should have been.

I also recall that this is/was related to the way in which morse speed is measured (PARIS or not etc).

For me the practical effect was that I had to increase the speed setting of my Farnsworth-spaced morse practice to achieve (roughly) the same number of characters in a given time as with the old player.

Posted: 2021-04-06 16:33
Yes, the "jscwlib" player uses correct timing whereas the old player doesn't calculate the timing correctly if you use Farnsworth timing (the higher the ratio between character and effective speed, the bigger the error).

If you download MP3s, they will always be generated by the old program, therefore they will yield the same as the HTML5 player option.

One fine day I will fix the old CW generator so both will generate the same speed with the same settings :-)


Posted: 2021-04-06 21:17
One fine day I will fix the old CW generator so both will generate the same speed with the same settings :-)

This day will be tomorrow. Watch this space!

Posted: 2021-04-07 19:31

Posted: 2021-04-07 22:39

Oops, what has just happened? I use the HTML 5 player and all of a sudden, the speeds have dropped dramatically. Is that because of the changes you've just made?

Is it actually the case that the HTML 5 player was right and the jscwlib is too slow?

Posted: 2021-04-08 07:06
The HTML5 player was always too fast, and jscwlib corrected the timing. So in order to have things consistent, I now corrected the timing of the HTML5 player.

I should maybe adjust the default speeds on this site to account for this...


Posted: 2021-04-08 08:45
I still don't get what "Use REAL speed (not PARIS)" is supposed to do.

If I select "Character Speed = 15 WPM, Effective Speed = 7 WPM" and check " Use REAL speed (not PARIS)", I get a 2:13 m file and it's definitely faster than 15 wpm without spacing.

The corresponding mp3 file is 6:03 m and has fastidious clicks.

Posted: 2021-04-08 09:45
REAL means net characters (5 characters per word) per minute. So if you set 60 wpm REAL for letter groups, the resulting timing (dot length, etc.) will be different than 60 wpm REAL for figures, because jscwlib will squeeze 300 letters / figures into one minute, and figures are on average longer.

REAL speed is only really relevant for HST training, because at the HSTs this is the metric used to measure the speed, as opposed to PARIS. The calculation is only correct without Farnsworth timing.

The MP3 generator doesn't support REAL, so it will generate the text with PARIS timing, thus resulting in something slower/longer.

Back to the Forum

You must be logged in to post a message.