Thetis 2.8.11-21+++ VAC advanced controls: results and discussion

USB headsets to digital audio workstation software...
K1LSB
Posts: 639
Joined: Wed Feb 05, 2020 5:25 pm

Re: Thetis 2.8.11-21+++ VAC advanced controls: results and discussion

Postby K1LSB » Mon Sep 13, 2021 10:11 pm

DL2XY wrote:We want buffering to be done mostly in the ringbuffer, so set PortAudio buffer to manual 1. Do not use "0", which is a "magic number" telling windows to use an automatic approximation (30 to 60ms), it would be the same as manual off.

73 Walter

I'll ask again....where are you getting that information?

Mark
User avatar
w-u-2-o
Posts: 5540
Joined: Fri Mar 10, 2017 1:47 pm

Re: Thetis 2.8.11-21+++ VAC advanced controls: results and discussion

Postby w-u-2-o » Tue Sep 14, 2021 1:04 am

W4WMT wrote:
w-u-2-o wrote:Still randomly unstable at RB 32. Not worth going bigger, that's already a ridiculously large value.


Do the random instabilities vanish when you Force=1 ?

I'll test that!
User avatar
W1AEX
Posts: 425
Joined: Sun Apr 09, 2017 6:17 pm
Location: Connecticut, USA
Contact:

Re: Thetis 2.8.11-21+++ VAC advanced controls: results and discussion

Postby W1AEX » Tue Sep 14, 2021 1:59 am

After reading through the thread I scrapped my old settings and went with what is in the screenshot below. If I drop below a value of 10ms in the ring buffer fields the underflows and overflows start to accumulate but it is rock solid at 10ms. Today I have been running Thetis for over 9 hours without any underflows or overflows. There have been no audio artifacts or other anomalies in the transmitted audio and the latency when monitoring myself during transmit is greatly reduced from what it was before. Thanks for all the discussion, explanations, and good data!

73, Rob W1AEX

Image
"One thing I am certain of is that there is too much certainty in the world."
User avatar
DL2XY
Posts: 116
Joined: Sun Jul 30, 2017 9:47 am

Re: Thetis 2.8.11-21+++ VAC advanced controls: results and discussion

Postby DL2XY » Tue Sep 14, 2021 11:44 am

K1LSB wrote:
DL2XY wrote:We want buffering to be done mostly in the ringbuffer, so set PortAudio buffer to manual 1. Do not use "0", which is a "magic number" telling windows to use an automatic approximation (30 to 60ms), it would be the same as manual off.

73 Walter

I'll ask again....where are you getting that information?

Mark

Sorry for late answer, i am quite short on spare time...

My statement originated from my experience in coding DSP Application (mostly SONAR testing stuff) using high end audio interfaces on windows-embedded.
The used audio library forces windows to use the user given latency and showed even that behavior: 0=default, too small= garbled audio, high enough=ok.

After a deeper look into PortAudio source i must admit thing are done different:
Portaudio does extensive negotiations with winaudio-api to achive the lowest value of latency windows will permit and uses this suggestion in the end.
What high this value is depends an many things like the audio format (buffersize,samplerate, nbr. of channels...), full or half duplex, exclusive access or not, the OS-version, the kind of API (MME,WDM,KS,DX-AUDIO...) and the system state itself (are there other audiostreams actvive, actual system load and others).

You may type in any low latency value you like, but you will not know what exactly you get.

So 0 or 1 should make no difference, since lowest achievable latency is about 2.5ms for WDS exclusive and may be as high as 120ms for MME multi access.

Sorry for misinformation, i hope it caused not too much confusion.

Walter

PS: The "magic 0" stays true for many other audio applications.
K1LSB
Posts: 639
Joined: Wed Feb 05, 2020 5:25 pm

Re: Thetis 2.8.11-21+++ VAC advanced controls: results and discussion

Postby K1LSB » Tue Sep 14, 2021 12:21 pm

Thank you very much, Walter!

73 Mark
User avatar
ramdor
Posts: 1468
Joined: Wed Jul 03, 2019 3:07 pm
Contact:

Re: Thetis 2.8.11-21+++ VAC advanced controls: results and discussion

Postby ramdor » Tue Sep 14, 2021 12:26 pm

Another full night of tests, around 14hrs or so.

R.

resampler3.jpg
resampler3.jpg (103.46 KiB) Viewed 37194 times
User avatar
ramdor
Posts: 1468
Joined: Wed Jul 03, 2019 3:07 pm
Contact:

Re: Thetis 2.8.11-21+++ VAC advanced controls: results and discussion

Postby ramdor » Tue Sep 14, 2021 6:21 pm

Just a quick vid on something I have been playing around with for the last couple of hours.

VID : https://youtu.be/3KyU9IGTd_M

varRatioGraph.jpg
varRatioGraph.jpg (32.53 KiB) Viewed 37170 times


Cheers, Richie.
User avatar
w-u-2-o
Posts: 5540
Joined: Fri Mar 10, 2017 1:47 pm

Re: Thetis 2.8.11-21+++ VAC advanced controls: results and discussion

Postby w-u-2-o » Tue Sep 14, 2021 8:20 pm

I get the random instability problem even when I have the resampler turned off, i.e. set to "force".

So I guess there is something unique about my OS, driver and hardware configuration that other people, and in particular Richie, is not seeing. Not entirely sure how I'm going to debug this, but I'm thinking on it furiously.
W4WMT
Posts: 325
Joined: Sun Apr 09, 2017 10:12 pm

Re: Thetis 2.8.11-21+++ VAC advanced controls: results and discussion

Postby W4WMT » Tue Sep 14, 2021 8:40 pm

w-u-2-o wrote:I get the random instability problem even when I have the resampler turned off, i.e. set to "force".

So I guess there is something unique about my OS, driver and hardware configuration that other people, and in particular Richie, is not seeing. Not entirely sure how I'm going to debug this, but I'm thinking on it furiously.


At least you have it isolated, and that's a start!

Is the instability bursty or does it just go bonkers forever?
User avatar
ramdor
Posts: 1468
Joined: Wed Jul 03, 2019 3:07 pm
Contact:

Re: Thetis 2.8.11-21+++ VAC advanced controls: results and discussion

Postby ramdor » Wed Sep 15, 2021 2:33 pm

These are the results from around 11hrs testing. There are times when it will run stable for 10-20 mins with hardly any adjustment, and then something will happen and it will start to move around again. Overflows/underflows have started to show up. I unfortunately had PA set to 0/0 in this test where as it was previously 1/1.

Richie.
Attachments
resampler4.jpg
resampler4.jpg (109.86 KiB) Viewed 37115 times
User avatar
w-u-2-o
Posts: 5540
Joined: Fri Mar 10, 2017 1:47 pm

Re: Thetis 2.8.11-21+++ VAC advanced controls: results and discussion

Postby w-u-2-o » Wed Sep 15, 2021 3:35 pm

So I had pointed the finger at my OS configuration in my post above, but I think I've changed my mind again and have swung that finger back around to point at the resampler algorithm in Thetis :D

I've got all my power management turned off, and set the computer into a state where literally nothing changes (that I can control, anyhow), and yet the resampler still goes periodically unstable.

So, just randomly hacking away at it, I found some very curious behavior associated with resampler and/or VAC initialization. If I poke at one of the PortAudio "manual" checkboxes, I can almost always make it the resampler go unstable, and further cycling of that checkbox will not repair the problem.

On the other hand, if I cycle the RingBuffer manual checkbox, this has the opposite effect. It almost always makes an unstable condition fix itself.

So perhaps there is some clue there in terms of different paths through the code that occur for each of those actions.

Nevertheless, how can the algorithm NOT know that under and overflow counts are occurring and NOT act to fix the problem?

I made a short video to illustrate what I'm writing about...

K1LSB
Posts: 639
Joined: Wed Feb 05, 2020 5:25 pm

Re: Thetis 2.8.11-21+++ VAC advanced controls: results and discussion

Postby K1LSB » Thu Sep 16, 2021 12:20 am

I have to admit I'm confused as to what this whole pursuit is all about, at least the "why" of it.

Scott, don't I recall you mentioning some time back in another thread that an occasional random overflow/underflow is inconsequential as long as it's not discernible in the audio?

I've been seeing very infrequent single-increment jumps in the counters ever since I've been using this radio, but I've never paid it any nevermind up to this point.

So what's the driver behind the sudden intense focus on achieving absolute zero-change long-term stability in the counters now?

TIA,
Mark
User avatar
w-u-2-o
Posts: 5540
Joined: Fri Mar 10, 2017 1:47 pm

Re: Thetis 2.8.11-21+++ VAC advanced controls: results and discussion

Postby w-u-2-o » Thu Sep 16, 2021 12:21 am

I've discovered one problem with respect to the resampler instability I'm experiencing: Voicemeeter Potato is doing a "Restart Audio Engine" every six hours. This appears to be associated with a bad program registration. I'm sure I can sort that out.

However, it begs the question: why does that make the resampler run wild? Shouldn't PortAudio automagically reset itself once Voicemeeter finishes its restart?
User avatar
w-u-2-o
Posts: 5540
Joined: Fri Mar 10, 2017 1:47 pm

Re: Thetis 2.8.11-21+++ VAC advanced controls: results and discussion

Postby w-u-2-o » Thu Sep 16, 2021 12:23 am

K1LSB wrote:I have to admit I'm confused as to what this whole pursuit is all about, at least the "why" of it.

Scott, don't I recall you mentioning some time back in another thread that an occasional random overflow/underflow is inconsequential as long as it's not discernible in the audio?

I've been seeing very infrequent single-increment jumps in the counters ever since I've been using this radio, but I've never paid it any nevermind up to this point.

So what's the driver behind the sudden intense focus on achieving absolute zero-change long-term stability in the counters now?

TIA,
Mark

Mark,

The problem I'm experiencing isn't "very infrequent single-increment jumps". The problem I'm experiencing is continuously, and rapidly accumulating, under and overflow counters. It's not an intense focus on achieving perfect stability, although that would be very nice!
K1LSB
Posts: 639
Joined: Wed Feb 05, 2020 5:25 pm

Re: Thetis 2.8.11-21+++ VAC advanced controls: results and discussion

Postby K1LSB » Thu Sep 16, 2021 12:31 am

Scott,

I do understand now, thanks for the explanation (and very quick reply!).

Mark
User avatar
GW0NTM
Posts: 11
Joined: Wed Jun 23, 2021 3:10 pm
Location: Penarth, Wales

Re: Thetis 2.8.11-21+++ VAC advanced controls: results and discussion

Postby GW0NTM » Thu Sep 16, 2021 8:12 am

I just wanted to say that I am also a Voicemeeter Potato user and see exactly the same issues that Scott describes. You are not alone!

Best
Robin
W4WMT
Posts: 325
Joined: Sun Apr 09, 2017 10:12 pm

Re: Thetis 2.8.11-21+++ VAC advanced controls: results and discussion

Postby W4WMT » Thu Sep 16, 2021 10:11 am

w-u-2-o wrote:So I had pointed the finger at my OS configuration in my post above, but I think I've changed my mind again and have swung that finger back around to point at the resampler algorithm in Thetis :D


Please consider this: If your ringbuffers are having fits of instability without the resampler engaged, then there's nothing the resampler can do to help. The underlying problem must be fixed first.

Just on the wild chance, have you tried disabling your VMM swapfile in Windows advanced settings? I have seen the Windows memory manager, during periods when it thinks you're not using your computer (no keyboard or mouse activity), begin to speculatively cache pages in a furious manner, eating up untold machine cycles.

73
User avatar
w-u-2-o
Posts: 5540
Joined: Fri Mar 10, 2017 1:47 pm

Re: Thetis 2.8.11-21+++ VAC advanced controls: results and discussion

Postby w-u-2-o » Thu Sep 16, 2021 11:53 am

Bryan @W4WMT

I suspect our posts may have crossed in the mail, so to speak. Somehow the registration of my Voicemeeter installation became corrupt without me realizing. Voicemeeter was doing its "reset every 6 hours" thing without me knowing, none of the usual advertising pop-up's that also normally come with that appeared. And apparently it's been like this a looooong time :? I reregistered it and now everything ran all night, over 12 hours, with just a few under/overflow counts being registered, all counters were under 50.

So now I've shifted gears to see if by messing with the Feedback Gain I can get some sort of a handle on the startup instability that I observe, and that was demonstrated in the video I posted above, when manipulating the manual checkboxes for both RingBuffer and PortAudio.

Observations so far this morning:

1. There is no setting of feedback gain that will withstand a Voicemeeter restart. It would appear that an external restart of drivers on the Thetis VAC portaudio interface is not handled.

2. Lowering the feedback gain x10^-6 from the default 4 to 1, or even .5, improves but does not completely alleviate the startup instability discussed above and in the aforementioned video.

3. More interestingly, lowering the feedback gain as above allows me to move from RingBuffers = 0 and PortAudios = 1 to RingBuffers = 0 and PortAudios = 0. At the default setting of 4, over/underflows will begin to accumulate at a slow but steady rate. Disclaimer: I've only been messing with this for a little while this morning. I'm currently running at 0.5x10^-6. I need to let it cook all day, or maybe for a few days, to really see if this is a stable condition.

BREAK

Robin @GW0NTM

Is your Voicemeeter installation registered? At the lowest contribution level it's not much of a tithe considering what a wonderful tool Vincent Burel has created for us.
K9RX
Posts: 414
Joined: Fri Apr 14, 2017 3:47 pm

Re: Thetis 2.8.11-21+++ VAC advanced controls: results and discussion

Postby K9RX » Thu Sep 16, 2021 12:30 pm

I offer this humbly for consideration. Might it be Windows changing - and if so it would be unpredictable (until and if what and how Windows is changing is discovered - at a minimum)?

I still run PowerSDR, the last release version. I had tried using the resampler many times and had issues where it would eventually run away. Nothing seemed to make it maintain stability. So I just said the hell with it and stopped using it. I've now seen the following - this IS consistent and repeatable.

I have noted that I can power up the computer - and test using FT8 in transmit. Sometimes I will see what I use to refer to as burps where the signal jumps up and is widebanded (in the passband and sometimes across the whole displayed pan width of maybe 20Khz when on FT8). If I see that - and it shows always within 2 - 3 transmissions max, I simple shut it down (the computer - of course properly closing things down first) and do the process over again. I can then get a stable signal that is rock solid - no jumps - clean transmission. The most I've had to repeat this process is 3 times - usually the 2nd time I reboot it is fine. I run my computer/radio 24/7 and generally its only when updates bug me that I do the reboot (I don't ignore updates from Windows, too risky). Although the times when I have to do it 3 times to get it to be stable are a pita - they are somewhat rare and considering the fact that I'm not doing this daily but rather about every 5 - 8 days makes it acceptable.

Its clear to me that there is something in Windows that impacts the VAC. It is not there all the time - actually mostly not there at all ... but it does show. Once it does - a reboot always takes care of it (based on the process noted above).

If indeed that is the case then I can't see how anything done IN the software, be it Thetis or whatever, is going to be able to KNOW what is happening and correct for it - perhaps by chance - but more than likely not until (and if) whatever is doing it is discovered, analyzed, and then a fault found that can be corrected for.

Gary
K9RX
User avatar
w-u-2-o
Posts: 5540
Joined: Fri Mar 10, 2017 1:47 pm

Re: Thetis 2.8.11-21+++ VAC advanced controls: results and discussion

Postby w-u-2-o » Thu Sep 16, 2021 2:20 pm

Gary,

The idea that Windows audio problems are somehow intractable or insurmountable is far from true.

I have no less than three very well behaved, totally stable, audio app's on my machine right now: Voicemeeter, Studio One, and Reaper. All three will run many days at a time with not so much as a burp, no magic or Windows restarts required. They even happily run right through my aforementioned Voicemeeter registration issue (now fixed). Historically, the same is true (for me, anyways), with Ardour, Audiomulch, Pro Tools, Stereotool, and probably a few other audio app's I've since forgotten.

This in spite of pushing things to a high level of complexity. My Presonus audio interface driver is attached to Voicemeeter. Reaper, Thetis and other app's attach to Voicemeeter. Admittedly I have gone to some trouble to ensure everything is running 48KHz throughout, and have also carefully tested and validated the smallest Presonus buffer size that will work under these conditions (which happens to be 256), but that is not hard to do.

Now I am admittedly pushing things to the limit in terms of latency. The Presonus, Voicemeeter, Reaper and Thetis are all using ASIO. In Thetis I currently have all buffer settings at 0 except for the main VAC buffer size which is set to 256 in order to match up with the Presonus. If I was not using Reaper in the configuration, only Voicemeeter, I'm confident I could get that down to 128. Reaper seems to be the buffer hog!

So with all that stuff working, it is not out of the realm of possibility to achieve the same level of performance in other app's, including Thetis.

Now that I've got that big distraction of Voicemeeter periodic restarts out of the way, looking at finer grain stability of the resampler has been a lot easier. Once you get it running it seems to be pretty good. At my current "extreme" settings, after two hours the worst case under/overflow counter has a 9 in it, which is, for all intents and purposes, an inaudible and undetectable error rate.

The issues seem to be, pending further research:

- A sensitivity to certain combinations of buffer sizes and sample rate.
- No ability to handle disruptions in the state of the attached audio devices.
- Occasional problems achieving stability at startup. Even with relatively conservative Ring and PortAudio buffer settings (for me) of 10ms across the board, I can cause the resampler to run away by cycling the PortAudio manual checkbox and fix it by cycling the RingBuffer manual checkbox.

That said, once running in a stable fashion, if nothing disturbs it it works fabulously well.

The original intent of having the resampler was to solve the problem of buffer under and overruns that were audible to those using VAC for phone and CW. Warren extended this to achieving a high level of frequency accuracy for digi modes as well. And when the resampler is behaving, as I wrote it works fabulously well. There are other threads on the forum discussing this for both FT8 and FME work and the like. I have personally measured it's performance. Those measurements were subject to ionospheric effects but were still measured in tenths of Hertz. With a good signal generator accuracy is probably below what we can measure. Bryan can tell us what theoretical performance is, probably measured in milliHertz.

As it turns out, the resampler can safely be disabled for digi modes as long as under and overrun rates are not excessive. There will be a small frequency offset, but that's just relative, and most digi modes will not notice the occasional glitch even if is audible to our ears. So for those who are having problems with it, or just don't care to be bothered with it, just don't use it. But for those who want buttery smooth audio, it's a must have.

73,

Scott
User avatar
ramdor
Posts: 1468
Joined: Wed Jul 03, 2019 3:07 pm
Contact:

Re: Thetis 2.8.11-21+++ VAC advanced controls: results and discussion

Postby ramdor » Thu Sep 16, 2021 3:43 pm

Well, this was a run for about 10hrs, there has been some overflows/underflows during that time. I can get it to go bonkers seemingly randomly by unchecking/checking some of the latency options, the same as in your video Scott, so perhaps there is some initialisation problem that can happen, not sure. I have also tested routing it through VMp, setting up an asio patch in VMp, an then out to the xair asio driver with the same results, so VMp is not causing any instability it seems.

R.
Attachments
resampler4.jpg
resampler4.jpg (111.51 KiB) Viewed 36991 times
W4WMT
Posts: 325
Joined: Sun Apr 09, 2017 10:12 pm

Re: Thetis 2.8.11-21+++ VAC advanced controls: results and discussion

Postby W4WMT » Thu Sep 16, 2021 5:24 pm

Way back when the resampler was being back-ported into PowerSDR (it was originally developed for Thetis) there were some long-term stability problems when users were selecting "larger" buffer sizes. Is there any way you guys can choose 128 or 64 as your buffer size to see if things run better that way?

73
User avatar
w-u-2-o
Posts: 5540
Joined: Fri Mar 10, 2017 1:47 pm

Re: Thetis 2.8.11-21+++ VAC advanced controls: results and discussion

Postby w-u-2-o » Thu Sep 16, 2021 5:47 pm

ramdor wrote:Well, this was a run for about 10hrs, there has been some overflows/underflows during that time. I can get it to go bonkers seemingly randomly by unchecking/checking some of the latency options, the same as in your video Scott, so perhaps there is some initialisation problem that can happen, not sure. I have also tested routing it through VMp, setting up an asio patch in VMp, an then out to the xair asio driver with the same results, so VMp is not causing any instability it seems.

R.

Pretty much identical results here, Richie!

I've been playing games on and off all day with it. For "all zeros" in the buffers, anecdotally I think I'm finding lower values of feedback gain to be more stable, say 1x10^-6, or 0.5x10^-6. That causes a very gradual accumulation of counters, maybe 10 counts per hour, give or take.

For a more conservative set of buffer settings, e.g. ringbuffers at 10ms and portaudio at 1ms, and the same feedback gains, the counters pretty much sit at zero.

But I've not been running any one particular configuration for more than 1 or 2 hours at a whack, so these results are quite preliminary.

- If the "bonkers" problem could be solved that would make it a lot more robust.

- If the code could recognize and handle a driver reset that would make it more user-proof.

- So far exposing the feedback gain seems valuable. Not sure about the slew time, yet. I'm not really hearing any under/overruns and that's where slew time adjustment would come into play, as it would change the audible characteristic of any such event. And I'm not sure it matters because it is clearly possible to reduce such events to very few per hour. Edited to add: I just intentionally caused it to go "bonkers" in order to introduce intentional under/overruns. I cannot hear the difference between 1, 3 and 10ms slew times.

Cheers!

Scott
User avatar
GW0NTM
Posts: 11
Joined: Wed Jun 23, 2021 3:10 pm
Location: Penarth, Wales

Re: Thetis 2.8.11-21+++ VAC advanced controls: results and discussion

Postby GW0NTM » Thu Sep 16, 2021 7:27 pm

w-u-2-o wrote:Bryan @W4WMT

Is your Voicemeeter installation registered? At the lowest contribution level it's not much of a tithe considering what a wonderful tool Vincent Burel has created for us.


Yes I'm registered. I completely agree, it works wonderfully here. I'd be completely lost without it!

R
User avatar
DL2XY
Posts: 116
Joined: Sun Jul 30, 2017 9:47 am

Re: Thetis 2.8.11-21+++ VAC advanced controls: results and discussion

Postby DL2XY » Thu Sep 16, 2021 10:29 pm

@Richie

I'm still working on optimizing the control loop of the resampler. Made some progress on understanding the interrelations by testing and digging in the sources. It would be extreme helpfull if you could do 2 changes in the current version:

1.)
Please make the parameter ff_alpha accessible in the Advanced tab. I think this could be used to manipulate regulation loop timing. In combination with Feedback Gain we get control over phase margin and stability.
You may use the slewrate controls for this , slewrate is not that important.

2.)
Please transfer the values of the "Force" values (udVACx_Forcex) to the parameters var at (re)inititialization of create_rmatch(). We can put in the values of a succsessfull run and use it for start value, reducing the long primary (re)settling time.

rmatch_parms.png
rmatch_parms.png (21.02 KiB) Viewed 36920 times


With these changes chances are good to develop an algorithm to programmatically optimize the loop for all different setups.

greetings Walter
K1LSB
Posts: 639
Joined: Wed Feb 05, 2020 5:25 pm

Re: Thetis 2.8.11-21+++ VAC advanced controls: results and discussion

Postby K1LSB » Fri Sep 17, 2021 1:38 am

This might be a stupid question but I'm curious -- do any of the VAC tweaks or advanced controls have any influence on PureSignal?

Put another way, if someone is experiencing any issues with PS functioning robustly, is there anything he might try adjusting in the VAC or advanced controls that might improve or stabilize PS's behavior?

Or is PS simply not concerned with those controls in any regard?

Thanks,
Mark
User avatar
w-u-2-o
Posts: 5540
Joined: Fri Mar 10, 2017 1:47 pm

Re: Thetis 2.8.11-21+++ VAC advanced controls: results and discussion

Postby w-u-2-o » Fri Sep 17, 2021 11:33 am

The VAC facility is utilized to transfer digital audio and IQ data into and out of Thetis. It has no other functions and has nothing to do with PureSignal.
K9RX
Posts: 414
Joined: Fri Apr 14, 2017 3:47 pm

Re: Thetis 2.8.11-21+++ VAC advanced controls: results and discussion

Postby K9RX » Fri Sep 17, 2021 12:31 pm

Scott,

Sorry - I didn't say that it was 'insurmountable' ... didn't even think it could be seen as suggesting that. I was only saying that I've found that I might have to reboot, typically once, once a week (and then only as forced by updates) to have a perfectly stable and clean system .... vs. what you seem to say previous to my post which is it goes bat-s#*t crazy randomly .... sorry - I prefer what I'm doing and was suggesting that to others who just want to use their radio and not be a test jockey.

I did play with the resampler - for a good bit doing all the 'tricks' suggested and recommended by you and others ... to no avail. It still, randomly, 'took off'. Then I noticed through normal reboots (again, due to Windows updates) that there were times when it would just work (resampler not active). So I tried, when I had a reboot and it didn't work, to just reboot again and lo-n-behold that worked. To me its less onerous to do this then to worry about my signal going awry all on its own. I have large high gain antenna systems and run power - it is critical to me to have a clean signal, on SSB (where I use PureSignal), on CW and on DIG modes. When the system is not working right I can hear clicks on RX and I'm sure they're there on all modes - but the issue is very prominently seen on DIG (FT8/Q65) transmit. That is not acceptable.

My comment was that whatever is causing the problem is not understood, and I'm sure you'd agree with that. And kudos to those researching it... but the fact that purely with a Windows reboot I can make it work - says it is indeed something IN Windows - and therein lies the need for focus. Also the fact that other apps don't have issues - just says either they've managed to FIND the cause and/or they're only part of the chain and not the part that causes issues. (and indeed it DOES say that it CAN be found and fixed!)

Again: my reason for posting was for those following that just want to have a radio that works and not be an experimentalist - or in most cases be changing parameters with no inkling of what it is doing and only a hope that it will improve things, that's the majority of people I'd bet. It is not to suggest it is "insurmountable" or to detract from those working on it ...

Gary
K9RX
User avatar
w-u-2-o
Posts: 5540
Joined: Fri Mar 10, 2017 1:47 pm

Re: Thetis 2.8.11-21+++ VAC advanced controls: results and discussion

Postby w-u-2-o » Fri Sep 17, 2021 1:25 pm

After 18 hours of soak time, results below. Note that I zeroed the counters after a 5 minute, initial settling period.

I'm pretty sure that I can up the buffers a hair and get the counts down to zeroes, but not sure it's worth it except as an academic exercise.

Walter @DL2XY great stuff! :)

Capture.JPG
Capture.JPG (118.68 KiB) Viewed 36870 times


Capture2.JPG
Capture2.JPG (89.51 KiB) Viewed 36870 times
W4WMT
Posts: 325
Joined: Sun Apr 09, 2017 10:12 pm

Re: Thetis 2.8.11-21+++ VAC advanced controls: results and discussion

Postby W4WMT » Fri Sep 17, 2021 3:45 pm

Purely for the future benefit of those searching this forum for possible solutions: Please don't forget that one can always take the lazy way out (as I have done) and simply clock your radio and your sound device from a GPSDO, thus eliminating clock domain skew and need for a variable resampler.

At least two manufacturers (PreSonus and Lynx Studio) offer sound adapters that accept an external clock source.

You can use a dual-ported GPSDO (like the Leo Bodnar) with one port set for 10 MHz (for the radio) and the other for 48 kHz (for the sound device). Or you can simply use two el-cheapo GPSDOs, one set for 10 MHz and the other set for 48 kHz. Works fine :-)

73, Bryan W4WMT

Return to “Digital ("Virtual") Audio”