2023-10-04 11.30.29.mp4
https://www.dropbox.com/scl/fi/kr1kyc3x36u8gdgz3pzug/2023-10-04-11.30.29.mp4?rlkey=l2xav92bu55zk3e41ui3qa9o4&dl=0Shared with Dropbox
After consulting with some friends running SmartSDR, I'm starting to think I have a problem.
Tune to a busy band with several signals. Doesn't seem to matter what bandwidth you have - I'm on 10m now about 30KHz bandwidth.
With your cursor, click left somewhere in the middle. Drag side to side. For me, the waterfall exactly tracks my mouse but the upper half lags behind. Once I stop dragging, they catch up. It's really odd too because the "lagging signals" leave tracks in the waterfall!
This also happens if I'm tuning near either edge of the panadapter and use the mouse-wheel to tune up (on the right) or down (on the left). I'll be tuning this way, see a signal I want to listen to and stop scrolling. The has passed by a bit and I have to scroll back to tune it in.
Here's a link to a video I took:
Anybody see this behavior? I'm running the latest software on Windows 11, Intel gen-12 CPU, 64GB RAM, gigabit network.
Thanks!
73 de Brian n8wrl
I think what you're seeing is the effect of tuning that is happening at multiple points in the DSP chains and not all of them are synchronous with the Panadapter data tap. This means that some of the data shows up prior to tuning and some of after. It has a spreading effect. This effect is compounded by the fact that waterfall data is not one sample, it is many samples over a period of time integrated together. We could attempt to prevent this by not collecting data while inside a tuning window and throwing out any accumulated data (average), but I think that effect would actually be worse visually.
In an ideal world all of the tuning would be done in between frames of incoming data and then we could do a better job of cleaning up these rough edges, but this is a lot harder than it initially appears. Just ask anyone that has ever worked on a synchronous system with multiple receivers/oscillators.
In short, what others have is correct -- this is "normal" behavior with today's software (SmartSDR v3.5.9).
Hi Brian, I do see that same artifact if I move the panadapter quickly, but not so much if I drag it slowly. I have a 6400 running 3.5.9 and a really fast 10th gen i9 computer with the nvidia GeForce RTX graphics board.
I don't find it objectionable, but I do notice it.
I took another look at this just now. 20m, 60Khz bandwidth, I am not moving the cursor that fast. Take a look:
Thank you Ed
73
-Brian n8wrl
Thank you Erik - I appreciate the explanation. 73!
-Brian n8wrl