2.6.9

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

Re: 2.6.9

Postby w-u-2-o » Sun Dec 15, 2019 9:34 pm

Hi Richie,

Ah, vsync, as in sync with the monitor frame rate automagically. I shall turn it on! However I already have the rate set to match my monitor, which is 30Hz.

Copy all on the waterfall update rate. Again, I had that set to 32ms to match the monitor rate, but setting it to 16ms does smooth it out a bit. Not running with the "CPU burner" :D

Thanks!

Scott
User avatar
W3MMR
Posts: 145
Joined: Thu Jul 11, 2019 8:36 am
Location: Springfield, PA
Contact:

Re: 2.6.9

Postby W3MMR » Mon Dec 16, 2019 12:09 pm

One thing I did notice, and Im not sure if it is a setting that I had in my PC. Im running a 2k 144hz monitor and a 1080p 60hz monitor. The 144hz monitor is set to 144hz in the monitor menu and in my Nvidia control panel, using a DP 1.2 output/input. When I selected Vsync, and mind you Thetis and the setup menu are both open on the 144hz monitor, it limited the FPS to 60. What I had to do was unplug the HDMI input to the 60hz monitor and the FPS went right up to 144hz. I plugged the other monitor back in, and now its staying at 144hz. Not sure if thats an issue on my PC's end, or what was going on. But never the less, its working now. And i thought the FPS looked low for the last few days, but it was still saying 144 so I didnt think anything of it. At least now I know that with Vsync enabled, its for sure running at 144hz.

Thanks for the hard work Richie!

Perry
W3MMR
User avatar
w-u-2-o
Posts: 5540
Joined: Fri Mar 10, 2017 1:47 pm

Re: 2.6.9

Postby w-u-2-o » Mon Dec 16, 2019 2:43 pm

Hi Richie,

It looks like the "Wheel Tunes Outside Spectral" option may be broken.

These are my current settings:

Capture.JPG
Capture.JPG (101.9 KiB) Viewed 15869 times


With those settings, and CTUN turned on, and zoomed in pretty far on the spectral display, when tuning with the mouse wheel (or MIDI, etc.) the display scrolls when the passband gets to the edge instead of the passband disappearing off of the edge of the spectral display.

Thanks!

Scott
User avatar
w2ner
Posts: 98
Joined: Wed Sep 11, 2019 12:14 am
Location: SC

Re: 2.6.9

Postby w2ner » Mon Dec 16, 2019 2:54 pm

HI Richie

Low priority item. I'm not sure this is a fault of the coding or not. When you record a signal it does not record with the settings you have in the receiver, for example. If you have the AGC set so the noise floor is not heard, the recording has it included. It makes any recording very nasty. Not sure this can be fixed or not. Just thought I would bring this up.
User avatar
ramdor
Posts: 1468
Joined: Wed Jul 03, 2019 3:07 pm
Contact:

Re: 2.6.9

Postby ramdor » Mon Dec 16, 2019 8:12 pm

Mouse wheel scroll outside spectral window means exactly that. If enabled mouse wheel tune will work in the blue+red area. If disabled mouse wheel tune will only work when over red spectral area. It does not change the behaviour of how the passband/filter scrolls. It was primarily put in to prevent frequency change if someone just missed a slider and was using mousewheel.

s1.png
s1.png (158.21 KiB) Viewed 15804 times


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

Re: 2.6.9

Postby ramdor » Mon Dec 16, 2019 8:20 pm

w2ner wrote:HI Richie

Low priority item. I'm not sure this is a fault of the coding or not. When you record a signal it does not record with the settings you have in the receiver, for example. If you have the AGC set so the noise floor is not heard, the recording has it included. It makes any recording very nasty. Not sure this can be fixed or not. Just thought I would bring this up.


Just tested using the record button on front end with AGC gain set so I could barely hear the noise. Recorded, played back through windows player, same level of noise. Then I did a record with agc set so that noise was blowing my ears out. Played back through windows player and same loud level heard. Are you using another recording system or the Wave window (which I myself have not used).

Richie.
User avatar
w2ner
Posts: 98
Joined: Wed Sep 11, 2019 12:14 am
Location: SC

Re: 2.6.9

Postby w2ner » Mon Dec 16, 2019 8:22 pm

ramdor wrote:
w2ner wrote:HI Richie

Low priority item. I'm not sure this is a fault of the coding or not. When you record a signal it does not record with the settings you have in the receiver, for example. If you have the AGC set so the noise floor is not heard, the recording has it included. It makes any recording very nasty. Not sure this can be fixed or not. Just thought I would bring this up.


Just tested using the record button on front end with AGC gain set so I could barely hear the noise. Recorded, played back through windows player, same level of noise. Then I did a record with agc set so that noise was blowing my ears out. Played back through windows player and same loud level heard. Are you using another recording system or the Wave window (which I myself have not used).

Richie.



No I was using the player built into Thetis, just hitting the play button. Someone asked me to record them and the playback noise was high as if I did not set the AGC level.
User avatar
ramdor
Posts: 1468
Joined: Wed Jul 03, 2019 3:07 pm
Contact:

Re: 2.6.9

Postby ramdor » Mon Dec 16, 2019 8:35 pm

I have not used the playback, but after a quick test on mine of an LSB signal, when played back I see what looks like a DSB signal, both sidebands, and at the centre of the panadapter. Not sure if it is actually being tx'ed at the frequency of where the vfo is, but did sound ok when using MON (on LSB and USB). All a bit odd.

I don't have a dummy load connected up so a bit of a pain to test, and shack is up the death trap steps which I refuse to go up until it all dries out and I can get some grip paint on them. The recent fall down them was not something I want to repeat any time soon :roll:

Richie.
User avatar
w2ner
Posts: 98
Joined: Wed Sep 11, 2019 12:14 am
Location: SC

Re: 2.6.9

Postby w2ner » Mon Dec 16, 2019 8:38 pm

ramdor wrote:I have not used the playback, but after a quick test on mine of an LSB signal, when played back I see what looks like a DSB signal, both sidebands, and at the centre of the panadapter. Not sure if it is actually being tx'ed at the frequency of where the vfo is, but did sound ok when using MON (on LSB and USB). All a bit odd.

I don't have a dummy load connected up so a bit of a pain to test, and shack is up the death trap steps which I refuse to go up until it all dries out and I can get some grip paint on them. The recent fall down them was not something I want to repeat any time soon :roll:

Richie.


That's what I see, DSB but the sound is much louder and my AGC has no effect on the playback. Thought I would bring this up and I have tried this a few times when someone requests a recording and playback.
User avatar
w2ner
Posts: 98
Joined: Wed Sep 11, 2019 12:14 am
Location: SC

Re: 2.6.9

Postby w2ner » Mon Dec 16, 2019 8:42 pm

I just tested the record and playing it back with the windows player, it sounds like it should. But the playback in Thetis is not doing that. So something is up with the playback in Thetis which makes it a big PITA to playback for someone.
wa1oxt
Posts: 73
Joined: Mon Apr 10, 2017 7:12 pm

Re: 2.6.9

Postby wa1oxt » Mon Dec 16, 2019 9:44 pm

Richie , you have the same problem I have.
I never had any luck with the playback feature in Thetis....
Always worked with P1 and MRX.

Nice feature to have , kind of miss it.

Turn your power down to zero watts output to test it.

garyradio / wa1oxt
VA2EY
Posts: 20
Joined: Wed Apr 24, 2019 11:57 am

Re: 2.6.9

Postby VA2EY » Tue Dec 17, 2019 1:09 am

I made a recording by trying the following functions successfully during the recording.
RX1 AF = ok
NB = ok
NR = ok
NR2 = ok
ANF = ok
AGC Gain = ok
all these functions work well and the playback (in TX) is the same, do not see anything abnormal.

In recording mode, you have an AF frequency. If you playback the recording without TX, you have double sideband appearance. Because you are not recording IF. You must switch to TX with LSB, USB, AM, DSB or FM mode



Anan 100D Thetis v2.6.9 b3_1
(google translate)
Michel
User avatar
ramdor
Posts: 1468
Joined: Wed Jul 03, 2019 3:07 pm
Contact:

Re: 2.6.9

Postby ramdor » Tue Dec 17, 2019 2:01 am

Ah ok, so just hitting play does not tx? You first need to be tx'ing then hit play?

Honestly I have not used the playback feature or looked at the code behind it, so have no idea what it it did/does.

Richie.
VA2EY
Posts: 20
Joined: Wed Apr 24, 2019 11:57 am

Re: 2.6.9

Postby VA2EY » Tue Dec 17, 2019 2:09 am

yes, you have to tx before, and hit play

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

Re: 2.6.9

Postby ramdor » Tue Dec 17, 2019 5:16 am

UPDATE

Hi all,

Please try ; see post #1

If there is nothing seriously broken in this build then code will go off to Doug in a couple of days.

73 for now, Richie.

(11/19/19) b4
-change: if wdsp.getpixels() returns no data available then data is not copied to display buffer
-add: peak blobs added to GDI+
-add: option to fade peak blobs when in hold mode
-fix: cw pitch shift code changed in vfoa/b lost focus so that multiple DDSFreq’s are not set, hopefully removing the artefact noticed by Chris W2PA
Last edited by ramdor on Wed Dec 18, 2019 12:50 pm, edited 1 time in total.
User avatar
dl6eat
Posts: 45
Joined: Wed Feb 14, 2018 1:31 pm
Location: Ellerhoop - JO43VR
Contact:

Re: 2.6.9

Postby dl6eat » Tue Dec 17, 2019 9:24 am

Hi Richie,

Just tested b4 for 30 minutes - all stabil.

BTW, I agree with W2NER that the "record and play" isn't useful as it is right now when not using VAC (just within Thetis)...…
The noise level isn't even close to what I get while receiving.

Anyway - wish all paricipants here in our nice forum "Merry Christmas" - thanks for pushing Thetis forward!

73,

Andy
No knobs, Pure Signal and Thetis (open source)
ANAN 7000, ANAN 10, Intel NUC 8i7 32 GB + 16 GB Optane
User avatar
Tony EI7BMB
Posts: 651
Joined: Sun Apr 09, 2017 2:31 pm
Location: Dublin
Contact:

Re: 2.6.9

Postby Tony EI7BMB » Tue Dec 17, 2019 11:07 am

Hi Richie, thanks for the update to B4 . Just testing here and it does not seem to change TX profile on band change . Think this might have been fixed on a previous release if memory serves?
User avatar
ramdor
Posts: 1468
Joined: Wed Jul 03, 2019 3:07 pm
Contact:

Re: 2.6.9

Postby ramdor » Tue Dec 17, 2019 11:17 am

Tony EI7BMB wrote:Hi Richie, thanks for the update to B4 . Just testing here and it does not seem to change TX profile on band change . Think this might have been fixed on a previous release if memory serves?


I thought it was fixed for mode change. Didn't think it did anything for band change related to profile. Mind is slowly starting to go fuzzy now :lol:

Richie.
User avatar
Tony EI7BMB
Posts: 651
Joined: Sun Apr 09, 2017 2:31 pm
Location: Dublin
Contact:

Re: 2.6.9

Postby Tony EI7BMB » Tue Dec 17, 2019 11:29 am

You are right Richie . For some reason all my bands changed to DIGU mode but as each band remembers the mode changing back to USB etc fixes it back to normal behavior . Sorry i did not notice what was really happening. Thanks again for the update
User avatar
W3MMR
Posts: 145
Joined: Thu Jul 11, 2019 8:36 am
Location: Springfield, PA
Contact:

Re: 2.6.9

Postby W3MMR » Tue Dec 17, 2019 11:34 am

Another thing I noticed to go with my post from yesterday. When I have Vsync enabled, and I have OBS open, and move it from my 144hz monitor to my 60hz monitor, the fps plummets to 60fps. When I move it bcack, it comes back up. Now, when Vsync is off and I do the same thing with OBS, I can tell on the panadapter that the FPS drops to 60, but its saying on Thetis its at 144fps. Why is this? Is this a priority thing in the software or is it a setting in my PC?

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

Re: 2.6.9

Postby ramdor » Tue Dec 17, 2019 11:40 am

ok Tony, good stuff.

I am currently working with the waterfall update delay. I have actually removed it all in a test version here, as it makes no sense due to the waterfall code being called at a rate based on FPS. For example if you set a waterfall delay for 20ms, and a desired FPS to 60, then the waterfall update wouldn't happen for 32ms at least.

I have changed it to update every frame, every other, etc, configurable, with the equivalent ms delay shown in setup. The calculated equivalent is based on fps.

Not sure it will be liked by everyone, but it is accurate, where as old method was a bit erratic.

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

Re: 2.6.9

Postby ramdor » Tue Dec 17, 2019 11:51 am

W3MMR wrote:Another thing I noticed to go with my post from yesterday. When I have Vsync enabled, and I have OBS open, and move it from my 144hz monitor to my 60hz monitor, the fps plummets to 60fps. When I move it bcack, it comes back up. Now, when Vsync is off and I do the same thing with OBS, I can tell on the panadapter that the FPS drops to 60, but its saying on Thetis its at 144fps. Why is this? Is this a priority thing in the software or is it a setting in my PC?

Perry


Hi Perry, yes interested in your findings regarding refresh. I can test on 30hz and 60hz only here.

fps displayed top left is actually a display of how many calls per second are made to the draw code, not necessarily how many frames your gpu presents to screen. As soon as VSync is enabled, the DirectX code will block so that a VSync rate is achieved. Also, if you drag from a 60hz to a 144hz the largest rectangle (coverage) on each monitor defines the vsync rate. That said :

assuming pc can do 144 renders per second, then...

@144 fps set in setup
on 60Hz screen without vsync it should show 144 top left
on 144Hz screen without vsync it should show 144 top left

@144 fps set in setup
on 60Hz screen with vsync it should show 60 top left, and you should get red box
on 144Hz screen with vsync it should show 144 top left

So, if you have 144fps set in Thetis, displayed on a 60Hz screen, you are not going to see those other 84 updates. Thetis is actually running the data 144 times a second (and displays 144), but your screen is only showing you it at 60 times a second. Essentially wasted cpu time.

Edit: in your setup, if you drag it from screen to screen then I would use vsync, and ignore the red box when you bring it to the 60hz display. It will be red because the desired 144fps now not achieved. When I have some time I will obtain the refresh rate of monitor and give suitable option to hide red box if achieving refresh rate.

Richie.
User avatar
W3MMR
Posts: 145
Joined: Thu Jul 11, 2019 8:36 am
Location: Springfield, PA
Contact:

Re: 2.6.9

Postby W3MMR » Tue Dec 17, 2019 12:05 pm

ramdor wrote:
W3MMR wrote:Another thing I noticed to go with my post from yesterday. When I have Vsync enabled, and I have OBS open, and move it from my 144hz monitor to my 60hz monitor, the fps plummets to 60fps. When I move it bcack, it comes back up. Now, when Vsync is off and I do the same thing with OBS, I can tell on the panadapter that the FPS drops to 60, but its saying on Thetis its at 144fps. Why is this? Is this a priority thing in the software or is it a setting in my PC?

Perry


Hi Perry, yes interested in your findings regarding refresh. I can test on 30hz and 60hz only here.

fps displayed top left is actually a display of how many calls per second are made to the draw code, not necessarily how many frames your gpu presents to screen. As soon as VSync is enabled, the DirectX code will block so that a VSync rate is achieved. Also, if you drag from a 60hz to a 144hz the largest rectangle (coverage) on each monitor defines the vsync rate. That said :

assuming pc can do 144 renders per second, then...

@144 fps set in setup
on 60Hz screen without vsync it should show 144 top left
on 144Hz screen without vsync it should show 144 top left

@144 fps set in setup
on 60Hz screen with vsync it should show 60 top left, and you should get red box
on 144Hz screen with vsync it should show 144 top left

So, if you have 144fps set in Thetis, displayed on a 60Hz screen, you are not going to see those other 84 updates. Thetis is actually running the data 144 times a second (and displays 144), but your screen is only showing you it at 60 times a second. Essentially wasted cpu time.

Edit: in your setup, if you drag it from screen to screen then I would use vsync, and ignore the red box when you bring it to the 60hz display. It will be red because the desired 144fps now not achieved. When I have some time I will obtain the refresh rate of monitor and give suitable option to hide red box if achieving refresh rate.

Richie.


Richie,
Thanks for the explanation on how it all works. And yes, when I drag Thetis to the 60hz monitor with Vsync enabled, it shows a red box and 60 top left. I wasn't clear in my explaination of whats going on, so I apologize. I was talking about dragging OBS to the 60hz monitor when vsync is enabled and it causes the red light and 60 to show upper left on Thetis, when Thetis is still on the 144hz display, when I do so. And the window size of OBS is much smaller than the window size of Thetis.

Perry
User avatar
W3MMR
Posts: 145
Joined: Thu Jul 11, 2019 8:36 am
Location: Springfield, PA
Contact:

Re: 2.6.9

Postby W3MMR » Tue Dec 17, 2019 12:07 pm

Also noticed that the larger the window I drag to the 60hz display, the more it effects the number displayed on thetis. Smaller the window, the less. OBS drops it to 60, My watt meter window, which is smaller, drops it to 80, and the Bob S Meter which is very small, drops it to 140. But as I sit here and type, I have firefox maximized on the 60hz display and it has no effect on the FPS number... My logging program has no effect either. Seems that some do and some dont. Weird...
User avatar
ramdor
Posts: 1468
Joined: Wed Jul 03, 2019 3:07 pm
Contact:

Re: 2.6.9

Postby ramdor » Tue Dec 17, 2019 12:12 pm

W3MMR wrote:Richie,
Thanks for the explanation on how it all works. And yes, when I drag Thetis to the 60hz monitor with Vsync enabled, it shows a red box and 60 top left. I wasn't clear in my explaination of whats going on, so I apologize. I was talking about dragging OBS to the 60hz monitor when vsync is enabled and it causes the red light and 60 to show upper left on Thetis, when Thetis is still on the 144hz display, when I do so. And the window size of OBS is much smaller than the window size of Thetis.

Perry


Probably some oddity with directX and interaction with OBS.

Quite a lot of posts on the internet that mention it :

https://obsproject.com/forum/threads/14 ... tor.77746/
https://www.reddit.com/r/Twitch/comment ... rame_rate/
https://linustechtips.com/main/topic/64 ... streaming/

etc

Also, window ui size issues probably related to UI scaling on each monitor being different?

Cheers, Richie.
User avatar
W3MMR
Posts: 145
Joined: Thu Jul 11, 2019 8:36 am
Location: Springfield, PA
Contact:

Re: 2.6.9

Postby W3MMR » Tue Dec 17, 2019 12:39 pm

After reading those links, it seems like a windows issue. Unfortunately its not just an issue with obs. OBS just seems to have the most effect. And some programs have an effect and some don't. Weird. Oh well. Not a big deal since I keep the watt meter and s meter on the 144hz monitor anyway.

Thanks for the help and the work you've done Richie..

73

Perry
K2GX
Posts: 1
Joined: Thu Apr 13, 2017 12:09 am
Location: Redding, Connecticut

Re: 2.6.9

Postby K2GX » Tue Dec 17, 2019 2:31 pm

Hi Richie,

Just tested b4 – the palette setting LinAuto (Display>RX1>Waterfall>Palette) Waterfall displays no data (White), with Driver Engine running DirectX.

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

Re: 2.6.9

Postby w-u-2-o » Tue Dec 17, 2019 2:34 pm

Is the "Wheel Tunes Outside Spectral" problem fixed in b4?

Cheers!

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

Re: 2.6.9

Postby ramdor » Tue Dec 17, 2019 2:53 pm

w-u-2-o wrote:Is the "Wheel Tunes Outside Spectral" problem fixed in b4?

Cheers!

Scott


not sure there was anything to fix Scott. See : viewtopic.php?f=9&p=9467&sid=b515f3c44a483a638e52b36a0bd62b38#p9435

It doesn't do what you think it does.

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

Re: 2.6.9

Postby ramdor » Tue Dec 17, 2019 3:18 pm

K2GX wrote:Hi Richie,

Just tested b4 – the palette setting LinAuto (Display>RX1>Waterfall>Palette) Waterfall displays no data (White), with Driver Engine running DirectX.

Joe


thanks, fixed. Issue with min/max states, and rgba order.

Richie.

Return to “Thetis”