Quirks with Thetis

NC3Z
Posts: 464
Joined: Sun Oct 29, 2017 8:57 pm
Location: Merritt, NC

Quirks with Thetis

Postby NC3Z » Mon Dec 10, 2018 1:24 pm

Now that the new 13 FW is out I decided to try Thetis (yes I know it is Beta and has not been updated since April 2018). But so far here are some quirks I have found that are not present in 3.4.9.

- No Panafall with RX2

- TX waterfall not adjustable

- When using PL1 Midi the VFO window does not update smoothly, or not at all until VFO stops moving, specifically the 1, 10 and 100KHz. In 3.4.9 this is very smooth.

- When using the Spotting and RX2, the spots are plotted on spacing based on RX1's Sample Rate, so if RX2's Sample Rate is not the same the spots on RX2's display are not in the correct place.


I am sure there will be others, maybe this will help the developers.
Gary NC3Z
User avatar
NJ2US
Posts: 106
Joined: Sun Oct 29, 2017 9:39 pm

Re: Quirks with Thetis

Postby NJ2US » Mon Dec 10, 2018 4:48 pm

Gary and I (NJ2US) have been testing Thetis with some very encouraging results, and our utmost respect and appreciation to the Devs for the effort. Anything that Gary saw, I have been able to duplicate. If you are on the East Coast, we are constantly discussing Thetis and P2 in the evenings on 160m (1855).

(1) Both of us have experienced the following crash, but I only saw it once, and Gary saw it once. It's simply an observation, a data point, and I have no doubt that others may chime in and say, "Not on my radio". But yet, it is worth noting here in case others do see it. We can not re-create the crash on demand. While transmitting, and making no adjustment, (LSB 1.855 mhz) Sample Rate 192000, TX simply stops, no power output, and PTT locked for approx 5 seconds, and then Thetis simply shuts down. Simply restarting Thetis via its windows shortcut. normal operation resumes. It has only happened to me once, and Gary saw it one time...exact same behavior.
(2) Also, sometimes, Pure Signal, just stops. The Feedback and Correcting prompts disappear from the display, and the PS-A radio button still shows as active. It even appears that although the display hints that pure signal has stopped, other people viewing the spectrum indicate that pure signal is still "working". For me, I noticed the anomaly occurred when I switched from Waterfall to Panadapter on RX2.

Feature Request: Now that you can run different sample rates for each receiver, would like to be able to have discreet Zoom and Pan controls for RX2.

It is surprising to me how few ops actually use the second ADC. I think it is invaluable if you have a second antenna to leverage. Not picking on anyone, just an observation, but it sometimes explains why improvements to RX1, don't always make it to RX2. Like Panafall on RX2 in Thetis, but I'm sure that will be stored shortly. Again, thanks to the Devs for all the hard work.

Jeff NJ2US
User avatar
w-u-2-o
Posts: 5539
Joined: Fri Mar 10, 2017 1:47 pm

Re: Quirks with Thetis

Postby w-u-2-o » Mon Dec 10, 2018 5:44 pm

Gary--those functions have simply not yet been carried over into Thetis from PowerSDR yet. Except for the spotting thing, which is a bug.

Jeff--thanks for the crash report!
NC3Z
Posts: 464
Joined: Sun Oct 29, 2017 8:57 pm
Location: Merritt, NC

Re: Quirks with Thetis

Postby NC3Z » Mon Dec 10, 2018 5:55 pm

I see 2.6.1 was just released, looking for release notes.
Gary NC3Z
NC3Z
Posts: 464
Joined: Sun Oct 29, 2017 8:57 pm
Location: Merritt, NC

Re: Quirks with Thetis

Postby NC3Z » Mon Dec 10, 2018 9:18 pm

Another issue. I run my 7000 in the Extended region to operate freqs (legally) outside the ham bands. When Extended is selected and you tune outside the ham bands on VFO B the band buttons for VFO A automatically change to the SWL band, and they refresh/redraw every VFO step.

When VFO A tunes outside the ham bands the VFO A band buttons do try to redraw but not as bad.
Gary NC3Z
NC3Z
Posts: 464
Joined: Sun Oct 29, 2017 8:57 pm
Location: Merritt, NC

Re: Quirks with Thetis

Postby NC3Z » Tue Dec 11, 2018 1:18 pm

Overall I like the improvements, and the wider RX spectrum. And VAC seems to work a lot better.

The VFO smoothness/responsiveness in the out of band area is a real PITA. The other quirks will have to be until the devs get time to bring the 3.4.9 features to Thetis.
Gary NC3Z
K9RX
Posts: 414
Joined: Fri Apr 14, 2017 3:47 pm

Re: Quirks with Thetis

Postby K9RX » Tue Dec 11, 2018 1:59 pm

Jeff,

I've reported an issue with PS stopping ... although in my case it does literally stop working. The button stays on (and I can toggle it off and on) ... the green "correcting" bar is there always but there is nothing in the feedback bar and it is not correcting as witnessed on the radio using DUP. I am unable to get it to come back on again without restarting the program (and the knock on effect of having to restart other programs as well)... however a recent add to my report, sent to the developers (unacknowledged as usual), is that I found a way to get it back on again without resorting to shutting it down. IF this is what you have (or others have) what you can do is pull up the Linearity window, hit Advanced. There is a value there called TINT. Change it from the default of 0.5 to 1.1 ... this seems to reset the system and it comes back to life again, PureSignal that is, then I set it back to 0.5 and close it out.

If this is what you are seeing, if indeed it is not on (turn DUP on to be sure) and it stays off try this to see if this brings it back to life.

Gary
K9RX
User avatar
NJ2US
Posts: 106
Joined: Sun Oct 29, 2017 9:39 pm

Re: Quirks with Thetis

Postby NJ2US » Tue Dec 11, 2018 3:03 pm

Thanks for the observation Gary. I have seen the situation where PS does stop working, but the button stays activated. I will experiment with the Tint setting. Also, I have not been able to get Single CAL to work. When I activate Single Cal, it seems to shut Pure Signal down.
I don't think I have found the detailed instructions for using Single CAL, although it seems straight forward enough.
Other than that, and a couple of the other quirks as noted in the thread, The P2/Thetis experience has been very positive for me.

Jeff NJ2US
NC3Z
Posts: 464
Joined: Sun Oct 29, 2017 8:57 pm
Location: Merritt, NC

Re: Quirks with Thetis

Postby NC3Z » Tue Dec 11, 2018 3:07 pm

Gary, I have also seen this PS issue on Thetis, but did not happen at all last night, but did earlier in the day. I will try your trick when it does.

As far as reporting issues, what is the best mechanism to do this?
Gary NC3Z
User avatar
w-u-2-o
Posts: 5539
Joined: Fri Mar 10, 2017 1:47 pm

Re: Quirks with Thetis

Postby w-u-2-o » Tue Dec 11, 2018 8:38 pm

This is the best place to report issues. The dev's are monitoring, but not jumping in per se. With respect to functionality such as TX waterfall scaling, these are known issues and we all just have to be patient. With respect to firmware issues, it would seem that the Orion (200D) image is in most need of additional work, and getting a new release for the "small FPGA" Hermes variant (aka 10E) is also a high priority.
NC3Z
Posts: 464
Joined: Sun Oct 29, 2017 8:57 pm
Location: Merritt, NC

Re: Quirks with Thetis

Postby NC3Z » Tue Dec 11, 2018 8:45 pm

I do like the reduction in network load, it is down a lot from Protocol 1, plus you get a 10X network speed on top of all that!
Gary NC3Z
NC3Z
Posts: 464
Joined: Sun Oct 29, 2017 8:57 pm
Location: Merritt, NC

Re: Quirks with Thetis

Postby NC3Z » Wed Dec 12, 2018 2:06 am

Weird, last night PS was stable. Tonight I loose PS, along with the Correcting and Feedback indicators disappearing, seems like every other TX without doing anything.
Gary NC3Z
User avatar
NJ2US
Posts: 106
Joined: Sun Oct 29, 2017 9:39 pm

Re: Quirks with Thetis

Postby NJ2US » Wed Dec 12, 2018 3:16 am

Thetis V2.6.2

Oddly enough, I experienced the same problem as Gary. Operating 160 M this evening LSB, in the middle of a transmission, Pure Signal simply just drops off (stops). The Feedback and Correcting prompts disappear. Then, after a few seconds, they re-appear and PS resumes operation. Other times, PS quits, and it takes a recycle of the PS-A button to bring PS back. Only data point I can offer, is that after a full day of operation at 384K sample rate for RX1 and RX2, I switched back to 192K sample, the problem started. Switched back to 384K RX1 and RX2, and the problem persisted. NC3Z also had made changes to sample rate. By the way, after making a sample rate change, I was careful to shut the software completely down, and restart after the sample rate change. Also tried the Tint .5 to 1.5 suggestion, didn't fix it.

Jeff NJ2US
K9RX
Posts: 414
Joined: Fri Apr 14, 2017 3:47 pm

Re: Quirks with Thetis

Postby K9RX » Wed Dec 12, 2018 2:29 pm

Just as a note you're reporting something different than what I'm reporting. For me the "Correcting" green bar remains on. I run with DUP on at all times and it is immediately evident that it is not running as the IMD is just ok at best (-30 or so) and the feedback bar is gone - not there at all. Nothing I do will bring it back again until I discovered the TINT thing. My guess is you're not seeing a condition where it thinks its active and it isn't and thus it doesn't know to respond to say the PS button being turned on or off ... and hitting the TINT is something that requires a complete reset of the PS algorithm, turning it off and on (internally) as is needed to change the conditions of the algorithm vis a vis TINT.

I've had it happen a few more times since discovering the TINT thing and it has always brought is back. Which is great because prior to this I had to shut down PSDR ... wait for several seconds or you get a Windows "2nd copy" warning ... which means 1/2 the time WSJT shuts down ... then I have to restart one or both programs ... then I have to change the affinity for less drop outs ... with TINT it is relatively pain free.

Gary
K9RX
User avatar
NJ2US
Posts: 106
Joined: Sun Oct 29, 2017 9:39 pm

Re: Quirks with Thetis

Postby NJ2US » Wed Dec 12, 2018 2:53 pm

Thanks for the clarification Gary, and I have seen your anomaly as well, where the "Correcting" bar remains. But when I saw that a few days ago, I didn't know about your "Tint" fix. I also wanted to clarify that when I see these issues , I have been running at the 192K sample rate on both RX1/RX2. Understanding that 192K is the preferred setting for PS. But when it does work, it works well at 384K. With the Primary Audio Buffer setting now gone from Thetis, and discreet sample rates for RX1/RX2, I would like to better understand the relationship to Pure Signal. Still, awesome job by Doug and the Devs, with updates almost daily, thank you. Im running v2.6.2

Jeff NJ2US
User avatar
W1AEX
Posts: 425
Joined: Sun Apr 09, 2017 6:17 pm
Location: Connecticut, USA
Contact:

Re: Quirks with Thetis

Postby W1AEX » Wed Dec 12, 2018 7:43 pm

I would echo that I also experience the disappearing PS issue that Jeff has outlined above. My 200D (FW 1.3) running Thetis 2.6.1 will run fine for several minutes and then PS will simply stop correcting. Sometimes it will resume on its own but other times it will get quite stubborn and require intervention such as changing the TINT value parameter.

Generally speaking, PS behaves fairly well on 160 and 75 meters but it's very unstable at 40 meters and above. I can get it to work on 20 meters by changing the TINT value to higher than the default, but the correction is horrible and the AmpView display looks like a piece of modern art. For those who have PS working well on the lower bands I would be interested in how well it works on the upper bands.

73,

Rob W1AEX
"One thing I am certain of is that there is too much certainty in the world."
User avatar
John
Posts: 16
Joined: Sun Apr 09, 2017 2:31 pm

Re: Quirks with Thetis

Postby John » Fri Dec 14, 2018 6:50 pm

Thetis V2.6.0
Firmware 1.3 on 7000DLE
Voicemeeter Banana 2.0.4.4
Sample rate 192000
Database - reset

Thetis crashes when VAC enabled. If Thetis is started with VAC enabled Thetis can crash within a few seconds. No amount of manipulation of values in VAC Monitor will allow VAC to "sync up(?)" or "lock up(?)". Constant restarts of Thetis will sometimes allow VAC to "lock". When locked VAC Monitor is very stable. Have run with a buffer of 256 and 22ms in & out ringbuffer with very good stability - once it "locks" all the way up to 1536000.
A work around for me is to start Thetis with VAC disabled, then open setup and take sample rate to minimum 48000 then enable VAC, VAC will then establish "lock". Sample rate then can be increased as desired.
YMMV
Have also found that record functions under "wave" tab do not function, yet.
Otherwise have been quite been happy with functionality and stability.
Seems in playing with Thetis that databases form older HPSDR is importable is this appropriate?
KRØP
User avatar
w-u-2-o
Posts: 5539
Joined: Fri Mar 10, 2017 1:47 pm

Re: Quirks with Thetis

Postby w-u-2-o » Fri Dec 14, 2018 7:08 pm

If you are having trouble with the resampler stabilizing, simply check off both "Force" checkboxes (with the force parameter left at 1.0000).
User avatar
John
Posts: 16
Joined: Sun Apr 09, 2017 2:31 pm

Re: Quirks with Thetis

Postby John » Fri Dec 14, 2018 7:25 pm

Sorry Scott thats not a fix here. Buffers continue to count "TO" underflows and "FROM" overflows. Both set to 1.000. Buffer set to 2048 and all manual boxes unchecked.
NL7F
Posts: 15
Joined: Sun Apr 09, 2017 8:33 pm
Location: Fairbanks, AK

Re: Quirks with Thetis

Postby NL7F » Thu Dec 27, 2018 4:31 am

Using
200D
FW/Orion_NP_v1.4.rbf
Thetis v2.6.2

I can confirm that recording function either quick record button or under Wave tab is not working.
Windows 7Pro 64 bit throws "Unhandled exception" .
NC3Z
Posts: 464
Joined: Sun Oct 29, 2017 8:57 pm
Location: Merritt, NC

Re: Quirks with Thetis

Postby NC3Z » Thu Dec 27, 2018 12:44 pm

John wrote:If Thetis is started with VAC enabled Thetis can crash within a few seconds.


I have the same exact issue here when starting 2.6.2 with VAC enabled.

However, the Record/Playback seems to work fine.
Gary NC3Z
User avatar
NJ2US
Posts: 106
Joined: Sun Oct 29, 2017 9:39 pm

Re: Quirks with Thetis

Postby NJ2US » Fri Dec 28, 2018 11:29 pm

FW 1.4, Thetis 2.6.2

Sample Rate 192K, A couple Pure Signal blow outs, but more importantly, Single Cal breaks Pure Signal on my setup. Every time. Must restart the application to fix.

Sample Rate 384K, a couple SEQ errors, flash by too fast to see what they are,

Quick Playback plays back, but when I MOX to retransmit, I just get noise.

Just some quick notes,

Jeff NJ2US
NC3Z
Posts: 464
Joined: Sun Oct 29, 2017 8:57 pm
Location: Merritt, NC

Re: Quirks with Thetis

Postby NC3Z » Sun Mar 31, 2019 1:30 am

Well it's been many months so guess time to see if any updates are coming, specifically the issues I flagged in the original post:

- No Panafall with RX2

- TX waterfall not adjustable

- When using PL1 Midi the VFO window does not update smoothly, or not at all until VFO stops moving, specifically the 1, 10 and 100KHz. In 3.4.9 this is very smooth.

- When using the Spotting and RX2, the spots are plotted on spacing based on RX1's Sample Rate, so if RX2's Sample Rate is not the same the spots on RX2's display are not in the correct place.

Anything going on?
Gary NC3Z
User avatar
w-u-2-o
Posts: 5539
Joined: Fri Mar 10, 2017 1:47 pm

Re: Quirks with Thetis

Postby w-u-2-o » Sun Mar 31, 2019 2:22 am

Gary,

The only thing I can tell you for certain that is coming is proper spectral display scaling during TX.

RX2 panafall is definitely low on the list.

Not sure about the other stuff.

73,

Scott
NC3Z
Posts: 464
Joined: Sun Oct 29, 2017 8:57 pm
Location: Merritt, NC

Re: Quirks with Thetis

Postby NC3Z » Sun Mar 31, 2019 11:31 am

Hmmmmm, OK. Was just hoping that we could at least get to what 3.4.9 does and then add to that. I hate loosing things with upgrades.
Gary NC3Z
User avatar
w-u-2-o
Posts: 5539
Joined: Fri Mar 10, 2017 1:47 pm

Re: Quirks with Thetis

Postby w-u-2-o » Sun Mar 31, 2019 1:18 pm

It will come, Gary, you just have to be patient. Remember that right now the only guy working on UI is Doug, as most developers much prefer to work on cool new DSP stuff. And, as luck would have it, Doug is being bombarded with a bunch of cool new DSP stuff as well as a number of bug fixes to integrate into the baseline, which, as a group, the dev's decided to prioritize over RX2 panafall.

And, just to wet your appetite, those things are:

a) a new, highly sophisticated VOX and downward expander
b) Full break-in CW, i.e. "real" QSK for 200D, 7000 and 8000 owners (relays are too slow in older hardware)
c) A bunch of random bug fixes: VAC1 startup bug, quick record/playback bug, VAC2 resampler bug, and probably a few other things I'm forgetting

And don't forget that there is a LOT of firmware test and development also going on to make it more stable and reliable across a larger population of hardware. There's been some good progress made there as well.

If there were 20 developers it would be a lot easier, but remember that there is only a very small handful of these folks. If you can recruit a few folks who are interested in UI work that would be HUGE.

73,

Scott

Return to “Thetis”