Different Modulation Modes -> Different Processing Times!

Roy Laufer

I was just playing around with the new Duplexing feature, to hear what I sound like on different modes, and I noticed something interesting...

Listening to myself on FM was a little disconcerting from the delay, but tolerable, then I switched to USB SSB and the delay became longer, and the difficulty to not sound like an inarticulate idiot become almost impossible (no smart @ss comments about that phrase, please).

It would appear that this little experiment indicates to me that the processing overhead and thereby the time delay from speaking->receiving is not the same for all modes. It is also interesting that different delay times can go from tolerable to almost impossible.

Not that surprising, but interesting nevertheless.

Ken - NM9P

From what I remember reading, the delay is in the routine for the filters.  Sharper skirts on the filters for CW & SSB require more passes through the algorithm and cause more delay (latency).  The filters on AM & FM don't require as sharp a shape factor and do not take as long to process.

BTW:  The more you play with the FDX mode, the more your brain will adapt to the delay.  Eventually you can train your brain to ignore it entirely.  I did. 

BTW.  I have read that delayed feedback in headphones is one of the experimental techniques used to help profound stutterers learn how to talk smoothly.

Ken - NM9P

W7NGA

I am always surprised when hams forget about all the processing that occurs on the voice modes, and that it takes a finite interval to process and migrate from microphone to the antenna. It isn't an issue as long as you don't attempt to monitor yourself real time. It is, apparently, if you are an obsessive contester where milliseconds are a lifetime. <grin>

W0DHB

There were several threads on the topic of latency Ken is correct that the steep filter skirt is the main culprit for the delay. There are two ways you can minimize the delay. - use DIG(U/L) for RX of (U/L)SB - Monitor RX on a DAX Channel The downside is that you lose RX audio processing (Equalizer, etc.) Dave