latency and the pursuit of levity ...

  • 1
  • Question
  • Updated 4 years ago
As I mentioned in another posting, I am not a contester. Indeed ... I think contesting is of minimal value and perhaps the bane of amateur radio. However, as an open-minded and accepting ham I realize that others may have opposing sensibilities and also might not appreciate the beauty of AM as I have held onto since becoming a ham in 1964. Different strokes as they say.

The latency threads did hold my interest in the comical vein, and as an engineering mathematician, many nights I lay awake exploring ideas at the extrema. And so, I chuckled as I envisioned the likely scenario of two SDR radios with high DSP latencies working each other. With that thought in tow, I headed to my engineering lab to take some measurements.

To ease the burden of calculations, let's assume the DX station has no latency or turnaround delays. As shown in this image, once the spoken word 'over' radiates to our receiving antenna there is 163ms delay before we actually hear the word 'over'. We now must introduce the auditory processing time between hearing the word and formulating our response. We can assume it zero for now .. acknowledging that there is an additional delay factor. Ignoring switching delays and as I have stated earlier, there is a reciprocal ~150ms latency in the other direction. That is, we speak and it takes 150ms before our response arrives at the antenna. Approximately one-third second transpires before the receiving DX station hears the first of our reply after their initial call. At my age with my own personal latencies, methinks .5 second would not be an unreasonable delay.

Greater comedy unfurls when you consider that the DX station may be running a similar configuration with associated SDR latencies. An issue? Certainly not for me and non-contesters, but it does exemplify a potential contribution that may be of concern. That, or simply fodder for discussions over a glass or two of a fine red wine.

73's

dan  W7NGA
San Juan Island, Wa.

Photo of W7NGA

W7NGA

  • 411 Posts
  • 174 Reply Likes

Posted 4 years ago

  • 1
Photo of N9VV

N9VV

  • 4 Posts
  • 1 Reply Like
Back in the "old days" the ARRL had a test for turn around. It was called the "TOR" test and referred to the amTOR and pacTOR modes. Those modes required that the receiving station acknowledge the packet or burst transmission within 20 milliseconds (ms). The TOR test was rigorous and seemed to help all of us CW Operators understand what kinds of T/R speeds to expect. Within the CW crowd this was commonly known as "QSK" operating. With some of the older Ten-Tec rigs, you could expect an 8ms T/R so that you could listen to the other (Contest) station between your own dits <g>  Sadly, the TOR test has fallen into disuse.
73 de Ken (worthless old-timer) N9VV
Photo of Walt - KZ1F

Walt - KZ1F

  • 3040 Posts
  • 644 Reply Likes
A couple glasses of red? I am on board with that. BTW, what is described is nothing compared to the latency of corporate conference call mechanisms. What is incredibly laughable and vertigo inducing is when a speaker is within earshot and about 3/4 to a second later you hear their voice through the headset/handset.  Reminds me of working the old Oscar Phase 3 birds. I found life way more pleasant muting the speaker when I was transmitting.

But points to you Dan for that engineer's description.
Photo of Barry N1EU

Barry N1EU

  • 495 Posts
  • 121 Reply Likes
Hi Dan!  Nice job with the incoming latency.  Can you turn that nifty toolkit and noggin of yours toward measuring the outgoing latency in ssb?  From words being spoken into Flex-connected microphone until emerging from the speaker of a nearby analog receiver?

It's only appropriate that you make these measurements Dan, considering that slow turn around and latency are highly desirable AM attributes   ;-)

73, Barry N1EU
Photo of W7NGA

W7NGA

  • 411 Posts
  • 174 Reply Likes

Barry,

Yesterday I made these measurements of the latency between spoken word at the microphone and modulation appearing at the antenna. I did essentially what you have asked for using my Collins 75S3C receiver.

latency AM   - 135.5ms
latency SSB - 146.5ms

I am fascinated with the QSK spec and going through the mental-gymnastics of 100wpm switching. At 25wpm, the 'off' bit time is 40ms so one has to understand that what they are hearing between characters is relatively old data. 

dan

(Edited)
Photo of Barry N1EU

Barry N1EU

  • 495 Posts
  • 121 Reply Likes
Thanks Dan.  So combined ssb latency rx + tx of 313.5msec.  

I'm sure we haven't heard the last of this . . .  folks tend to be attached to the notion of operating in the realm of real-time.

Barry
Photo of Bill -VA3WTB

Bill -VA3WTB

  • 3012 Posts
  • 663 Reply Likes
The price for such high quality filters.
Photo of W7NGA

W7NGA

  • 411 Posts
  • 174 Reply Likes
It's much like using a cellphone .. you're not aware of the delay until you hear someone talking to you on a nearby phone.

I should mention that you can dramatically reduce the latencies by increasing the filter bandwidth in CW. However, that tends to diminish the effectiveness of the mode.
(Edited)
Photo of Mike W9OJ

Mike W9OJ

  • 88 Posts
  • 29 Reply Likes
I'm aware of it every time I use a cellphone.
That's why I don't like using a cellphone. I'm always doubling with the person on the other end.
Photo of W7NGA

W7NGA

  • 411 Posts
  • 174 Reply Likes
CW operators will be pleased to note that there is essentially zero T/R switching delay, as expected, from pressing the key to RF at the antenna.

 ~7ms on CW transmit.
(Edited)
Photo of Charles - K5UA

Charles - K5UA

  • 317 Posts
  • 88 Reply Likes
Is there pin diode switching in the CW mode and mechanical relay switching in the SSB mode?

If so, could that be part of the turn around time?
Photo of Bill -VA3WTB

Bill -VA3WTB

  • 3012 Posts
  • 663 Reply Likes
From Steves post, Flex Rep

Relays have nothing to do with these measurements.  The measurements are a result of signals passing through the signal chain in the receiver and the transmitter.  As one of my radio mentors was fond of saying "all the delay comes from the quality of the final filter."  This is a hard concept to wrap your head around if you are not familiar with it, but the net-net is that the better filter you provide at the final sampling rate, the more latency is introduced.  The FLEX-6000 is the first radio I am aware of to provide variable-depth filters.  We can significantly reduce the delay by reducing the filtering and we can do this dynamically.  Even the FLEX-5000 with PowerSDR did not have this capability.

Our read when we designed the signal chain for sideband was that a little latency was less important than good filtering.  So we "maxed out" the filtering capabilities for sideband.  The filter dynamically changes width in DIGx modes and in CW.  If you would like to try out a shorter latency with reduced filtering capabilities, try running in DIGU instead of USB and set your filter width to greater than 2kHz.  The latency will go down to what we consider to be a very minimal level.  

The ARRL and others are not used to measuring radios that have complete flexibility in how things like this are controlled and so they put the radio in a given position and took a measurement.  The measurement is accurate for that one setting, but is not a complete picture of what the radio can do (and in this case doesn't explain the tradeoff that we made).  I'm sure I would have done just what they did.  As you guys probably know by now, if you provide a compelling case for making a change in the radio there are very few things we cannot change.  You can try using DIGU and let us know if you prefer it like this.  There is actually a table in the owners manual with this information along with a description of how it works (see 29.6.4 on page 140).  
(Edited)
Photo of Charles - K5UA

Charles - K5UA

  • 317 Posts
  • 88 Reply Likes
Yes, aware of all that. Not asking about latency because of filters, curious about mechanical relay switching time on overall turn around time. Significant or not significant?
Photo of Charles - K5UA

Charles - K5UA

  • 317 Posts
  • 88 Reply Likes
I found the answer to my question. Wikipedia states that RF relays have switching times around 10 msec, whereas PIN diodes have switching times of about 1 microsecond ( about 10,000 times faster. Even 10 msec doesn't seem to be that much in the overall turn around time. So it seems not be significant (EXCEPT in the CW mode where you would want RF output immediately, especially in QSK mode.). Since the 6000 series has RF out 8 msec after key closure, it must be using a fast reed relay or PIN diodes.. I can hear a quiet relay when in the CW mode, and a loud relay in the SSB mode. Why can't we have a quiet relay in the SSB mode?