Data trace display averaging

  • 6
  • Question
  • Updated 6 years ago
  • Answered
Is averaging on the list for V 1.0 ?

Ned, K1NJ
Photo of Ned K1NJ

Ned K1NJ

  • 319 Posts
  • 83 Reply Likes

Posted 6 years ago

  • 6
Photo of Tim - W4TME

Tim - W4TME, Customer Experience Manager

  • 9198 Posts
  • 3558 Reply Likes
There is a current feature for slowing down the frame rate that will make the display less "jumpy", but actual FFT display averaging is not slated for SmartSDR v1.0
Photo of Ken - NM9P

Ken - NM9P

  • 4239 Posts
  • 1351 Reply Likes
I would imagine this would be implemented after or about the same time as waterfall, pan-fall, etc? (important, but B-list priority, after other crucial performance enhancements?)
Photo of Michael - N5TGL

Michael - N5TGL

  • 308 Posts
  • 103 Reply Likes
Sorry to hear that this is not going to make 1.0. The jumpy/spastic display looks terrible in my opinion, and this is what the brand new owner is going to look at all day. Not only that, it's what the brand new owner's friends are going to look at all day. Lowering the FPS is a solution, albeit a poor one -- it makes it look like you have a computer that can't handle the load and are getting bad frame rates as a result.

To me, the nicest display ever is the 5000 with AVG on and the FPS set to 30. Beautiful, liquid traces that are just sexy. The current display is like staring into a 8kv arc, but without the ozone.
Photo of Paul Christensen, W9AC

Paul Christensen, W9AC, Elmer

  • 323 Posts
  • 138 Reply Likes
I've not spent much time with the PowerSDR/Flex 5K display, but the FFT averaging as used in SpectraVue software with the SDR-IQ is superb. True peaks are shown and the decay is very fluid.

I note that as averaging is increased in SSDR (V. 1.0.24), peak values are compromised.

Just an observation...

Paul, W9AC
Photo of Tim - W4TME

Tim - W4TME, Customer Experience Manager

  • 9198 Posts
  • 3558 Reply Likes
Paul,

This is a interesting idea. You should create a new Idea topic so others can comment on it.
Photo of Steve - N5AC

Steve - N5AC, VP Engineering / CTO

  • 1057 Posts
  • 1097 Reply Likes
Paul, I just tried doing a 4x attack and a 1x decay so that the peaks will be more pronounced and show quicker. The result is interesting -- it looks like the spectrum is on a treadmill and falls behind and then catches up periodically. It's a little distracting to me, but it does accentuate the peaks. I'm not sure I'm going to leave this in the code though...
Photo of Steve - N5AC

Steve - N5AC, VP Engineering / CTO

  • 1057 Posts
  • 1097 Reply Likes
Paul, if you want to drop me an email (steve -at- flexradio dot com), I'll shoot you the video.
Photo of Paul Christensen, W9AC

Paul Christensen, W9AC, Elmer

  • 323 Posts
  • 138 Reply Likes
Thanks, Steve. Much appreciated and I look forward to eyeing the video.

When you have a moment, you want want to take a quick look at Spectravue for its peak response and fluid motion attributes. As a reference, my values are:

FFT Ave = 8
Smoothing = 0
FFT/Blk = 4096

Paul, W9AC
Photo of Steve - N5AC

Steve - N5AC, VP Engineering / CTO

  • 1057 Posts
  • 1097 Reply Likes
Paul liked what I sent in the video. We're going to try this change for v1.1 and see how folks like it.
Photo of Al / NN4ZZ

Al / NN4ZZ

  • 1853 Posts
  • 672 Reply Likes
Steve,
Any chance the 2 averaging algorithms could be options? That would make it easier to compare them. (not sure how we could do it otherwise). Also if there are differing opinions on what is best, a user could pick their favorite.

BTW, the current averaging is nice and better than I'm used to. So if the new algorithms is even better that would be great.

Regards, Al / NN4ZZ
Photo of Paul Christensen, W9AC

Paul Christensen, W9AC, Elmer

  • 323 Posts
  • 138 Reply Likes
Al,

Ask Steve for a link to a video demo. With that you can immediately see the difference. The new algorithm responds to signal peaks but retains the fluid decay of averaging. This has the benefit of seeing true signal peaks that are otherwise reduced through averaging.

Paul, W9AC
Photo of Al / NN4ZZ

Al / NN4ZZ

  • 1853 Posts
  • 672 Reply Likes
Steve & Paul,
I would like to see it.....and maybe we can upload it for others to see. I have a YouTube channel where I have posted a few other Flex videos.

Regards, Al / NN4ZZ
Photo of Ken - NM9P

Ken - NM9P

  • 4239 Posts
  • 1351 Reply Likes
I would love to see it, too, if possible.
Photo of Bill-W9OL

Bill-W9OL

  • 726 Posts
  • 56 Reply Likes
I agree, I use AVG for cw and digital. (don't do much SSB) find the display easy on the eyes, and very easy to precisely select the proper place to transmit.
After long hours of contesting, I find it EZ less tiring then that frenetic display you get when you first start the 6.5K.

First thing I do when starting SSDR is go to 'Display' and move the FPS way lower.

But then what do I know?