SSDR V1.2.1 - Panafall Speed - Increase the maximum RATE for visual CW copy at higher speeds.

Al_NN4ZZ

Is it possible to INCREASE the maximum speed of the panafall RATE setting?  A faster rate would allow one to visually copy CW at medium to high speed?   At the current maximum rate setting it appears to support visual copy up to about 10 WPM.  

See snapshots below at about 20 WPM and about 10 WPM.

I realize that a higher speed means more processing, faster scrolling, and less storage but for CW that may be a good trade off for some of us.  

Is this idea practical?

Regards, Al / NN4ZZ  
al (at) nn4zz (dot) com




. 


At about 20 WPM characters run together....CLICK for full size image.

Ken - NM9P

Interesting concept, if it can be done w/o problems.

Al_NN4ZZ

Ken,
It could be handy if you are not sure whether the DX busted your call or got it right.

Just scroll back and check it out visually.  And you can screen shot the segment for reference. 

Regards, Al / NN4ZZ  
al (at) nn4zz (dot) com

Jerry Gardner

Why not just use CW Skimmer?

Ernest - W4EG

How about being able to copy code without relying on machine or software.

That would be a great innovation!

Al_NN4ZZ

Hi Jerry,
If CW Skimmer is eventually integrated into the SSDR GUI display that is sent from the radio to the client, that would be great.  But if the only way to use it continues to be as a separate program that runs on the client, it would not be as easy to use skimmer on some of the mobile devices (tablets, etc).    In the long run, a thin client display (on many devices like tablets, phones, etc) and access over the internet would really set a high bar for other to match.  

Hi Ernest,
I'm with you on copying code the old fashioned way, but it sure is nice to see all of the calls (and the 599s)  when there is a pileup spread over 10 KHZ or more.   I mainly use skimmer for those DXpedition pileups.  

Regards, Al / NN4ZZ  
al (at) nn4zz (dot) com

Steve-N5AC

My inclination is to use radio resources for something else. But if this floats your boat, it might be useful to know that it will probably work better if you are zoomed out a little. To copy 20 WPM CW, you'll want at least a 33Hz bin size in the panadapter. Zooming in and cranking up the speed will induce temporal distortion in the waterfall making your activity less productive. As a rough estimation of bin size, look at the width of the panadapter divided by the number of pixels in the display. This will only be rough for a number of esoteric reasons.

Al_NN4ZZ

Steve,
Thanks for the reply and information.  I played with it a bit more but can't get enough to copy at the higher speeds.  Not a big deal as I expect CW Skimmer will be a more  effective solution when the integration is improved.    

I was hoping the rate adjustment would be a simple change as an interim solution.

Regards, Al / NN4ZZ  
al (at) nn4zz (dot) com 

Steve-N5AC

Did you try zooming out where bin size is 33-200Hz or so?

Al_NN4ZZ

Steve,
Yes, as I zoom out more of the signals become readable visually.  The faster CW is still difficult at the current rate as some of the character elements appear to get overwritten.   You can get the same effect on the slower CW by decreasing the rate. 

I don't think zooming out is enough at the higher speeds unless I'm missing something.  I'm assuming a higher rate would make the character elements on the waterfall longer as time scale on the right decreases (for example from 2 seconds to 1 second). And assuming the higher rate would prevent the elements on the faster CW from being overwritten.  

But again I wouldn't want you to spend time on making a change to allow a higher speed unless it is easy, practical, and others also see benefit it doing it.    

In the long run an integrated CW skimmer is probably the best solution.  Thanks for your feedback, it helps me get a better appreciation for the process. 

Regards, Al / NN4ZZ  
al (at) nn4zz (dot) com

Al_NN4ZZ

Video showing waterfall speed of SSDR V1.2.1  and CW Skimmer.  Not sure if this is apples to oranges but shows the speed difference.

Regards, Al / NN4ZZ  
al (at) nn4zz (dot) com

http://youtu.be/TrDKqE2DFs0


https://www.youtube.com/watch?v=TrDKqE2DFs0

Steve-N5AC

30WPM would be 40ms dits. For these to show up with a couple of pixels vertically, you really need to sample at 10ms. This would be 100 lines of waterfall per second. Your eyes would not be able to follow this so you would capture some and then click to halt the waterfall and read it visually? At 10ms per line, we would need to collect 4096 samples in 10ms so the sampling rate must be above 410ksps. Since we are working in I/Q, you would need to be looking at greater than 410kHz of bandwidth. At 410kHz of bandwidth, each pixel will be 100Hz of bandwidth. For me, I would far prefer to decode CW with my ears and brain than my eyes and brain. If I was going to use my eyes, I would prefer to use a blinking light rather than a waterfall, which I think would be very distracting while I was trying to focus on decoding with my eyes. The FFT in the radio can run several THOUSAND FFTs per second., but to really do what you want, I think you would also want to incorporate a filter and sample even faster than the 100 per socond I mentioned above. By the time we've done all this, it's almost as easy to just decode the CW and tell you what is being sent...

Al_NN4ZZ

Steve,
That final thought from above would be the holy grail --- "By the time we've done all this, it's almost as easy to just decode the CW and tell you what is being sent..."

Maybe some day....

I really don't want to copy visually but thought it might be nice to look back through the waterfall if I wasn't sure the DX station got my call right or to spot a 599 in a pileup.  The CW skimmer type decoding integrated in the display would be much better. 

Regards, Al / NN4ZZ  
al (at) nn4zz (dot) com