Page 11 of 21

Re: 2.6.9

Posted: Thu Dec 12, 2019 3:18 pm
by W1AEX
Beautiful job with the enhancements in the b3 release Richie. Monitoring the CPU utilization of Thetis alone is way more useful to me so that's a great change for my operating. The new framerate and uV options are working perfectly as well. I count myself lucky that I don't see any SEQ errors with v2.6.9 so I can't report on the changes made to the SEQlog reporting of -ve sequence errors.

Thanks again for every release you have put out there!

73, Rob W1AEX

Re: 2.6.9

Posted: Thu Dec 12, 2019 3:29 pm
by w4cm
ramdor wrote:UPDATE

Hi all,

Please try : https://www.dropbox.com/s/150bv7vdxfjsn ... 3.zip?dl=0

This version includes a work in progress historic meter data. It can be enabled in setup>display>general in the multimeter group. Colour/alpha is set in appearance> meter. Readings are currently spread over 2 seconds.

Also, the regular frame timing code has had a work over, and you will notice that fps displayed in top left of spectrum will be closer to main display fps configured via setup>display>general.

Use the drop down/click the cpu% in the status bar to select displayed value.

NOTE: seqlog error warning on status bar will only show for -ve numbers. There is now an option to control this in the seqlog window.

73 for now, Richie.

(11/19/19) b3
-change: frame timing improvements, and removed waterfall late changes made in b2
-add: cpu% can now show thetis/system cpu%, use the drop down on the cpu% in status bar
-add: option in SeqLog to only show status bar warning when -ve seq errors detected (default)
-add: rx signal history over two seconds shown on smeter. Options to turn on/off and set colour/alpha




Thank you very much for all the goodies,

W4CM/ Cecil

Re: 2.6.9

Posted: Thu Dec 12, 2019 3:35 pm
by VA2EY
Here is my problem with the Auto-Attenuate (PS-A) since version Thetis_2.6.9_B1, 2.6.9_B2, 2.6.9_B3. Note that with the previous versions everything worked correctly. Test with old database or new and same problem

Drive 15 or low value ok (Fig. 1), but when I increase the drive (Fig .2) the Auto-Attenuate does not work. I have to put the S-ATT manually (fig.3). The Autto-Attenuate only works very rarely, so it can not be used.
Anan 100D
Firmware Angelia_Protocol_2_v11.7_7-15-18
1.JPG
fig 1
1.JPG (144.9 KiB) Viewed 4895 times

2.JPG
fig 2
2.JPG (141.42 KiB) Viewed 4895 times

3.JPG
fig 3
3.JPG (142.66 KiB) Viewed 4895 times

Re: 2.6.9

Posted: Thu Dec 12, 2019 4:34 pm
by dl6eat
Thanks Richie for the "again" great update!

I had the chance to test "b3" for 2 hours again today with no issues. Simply great to see that super fast progress. Love it....

May be I can just try to throw a wish into your project funnel? Since I love to have a "live" screen I'm using most likely low averaging.
The display feels "up-to-date" and shows SSB-signals following the voice of the operators right away…..no delays….like on an analog spectrum analyzer.
In that mode I reduced "Peak Blobs" to 1......but on crowded bands you still see that red circle Jumping around very quick....too quick in my mind.
Also if you are talking to a specific person and enable "Only Inside-Filter" the peak readings change too quick.

Would it be possible to have an optional function like "Peak-Hold"?
May be the peak value can be displayed in addition on top of the screen right above the Jumping red circle?

It's simply hard to read values when averaging of the panadapter is low.

Just an idea…….

Thanks for reading - 73,

Andy

Re: 2.6.9

Posted: Thu Dec 12, 2019 5:11 pm
by ULTIMAX
dl6eat wrote:Thanks Richie for the "again" great update!

I had the chance to test "b3" for 2 hours again today with no issues. Simply great to see that super fast progress. Love it....

May be I can just try to throw a wish into your project funnel? Since I love to have a "live" screen I'm using most likely low averaging.
The display feels "up-to-date" and shows SSB-signals following the voice of the operators right away…..no delays….like on an analog spectrum analyzer.
In that mode I reduced "Peak Blobs" to 1......but on crowded bands you still see that red circle Jumping around very quick....too quick in my mind.
Also if you are talking to a specific person and enable "Only Inside-Filter" the peak readings change too quick.

Would it be possible to have an optional function like "Peak-Hold"?
May be the peak value can be displayed in addition on top of the screen right above the Jumping red circle?

It's simply hard to read values when averaging of the panadapter is low.

Just an idea…….

Thanks for reading - 73,

Andy



I'm second in motion to what Andy Suggest, is so quick you get motion sickness :lol:

Al
N3YV
73

Re: 2.6.9

Posted: Thu Dec 12, 2019 6:35 pm
by wa1oxt
I seem to have a level issue with the 2tone test.
Anybody else ?

garyradio//wa1oxt

Re: 2.6.9

Posted: Thu Dec 12, 2019 7:06 pm
by wa1oxt
Please ignore my last post.
Problem solved with a reboot of software.

tnx oxt

Re: 2.6.9

Posted: Thu Dec 12, 2019 8:29 pm
by w-u-2-o
VA2EY wrote:Here is my problem with the Auto-Attenuate (PS-A) since version Thetis_2.6.9_B1, 2.6.9_B2, 2.6.9_B3. Note that with the previous versions everything worked correctly. Test with old database or new and same problem

Drive 15 or low value ok (Fig. 1), but when I increase the drive (Fig .2) the Auto-Attenuate does not work. I have to put the S-ATT manually (fig.3). The Autto-Attenuate only works very rarely, so it can not be used.
Anan 100D
Firmware Angelia_Protocol_2_v11.7_7-15-18
Have you tried going back to a Thetis version before 2.6.9_b1 recently? In other words, now that it does not work for you, if you go back now does it work for you again? It may not be a Thetis problem. This behavior is more indicative of a firmware problem or a problem with excess current through the F3 PTC device.

Re: 2.6.9

Posted: Thu Dec 12, 2019 10:44 pm
by ramdor
WIP with the peak stuff



Richie.

Re: 2.6.9

Posted: Thu Dec 12, 2019 10:50 pm
by ULTIMAX
Wow Richie,

Thank you for taking that into account,,that's going to sweet brother,,,thanks can't wait :D

Re: 2.6.9

Posted: Thu Dec 12, 2019 11:01 pm
by VA2EY
[/quote]Have you tried going back to a Thetis version before 2.6.9_b1 recently? In other words, now that it does not work for you, if you go back now does it work for you again? It may not be a Thetis problem. This behavior is more indicative of a firmware problem or a problem with excess current through the F3 PTC device.[/quote]

I just checked (many more times) Thetis_2.6.8 any alpha version
Thetis_0700_171119_269_a7
Thetis_2240_181119_269_a8
all these version works properly

problem only with
Thetis_2320_211119_269_b1
Thetis_1622_091219_269_b2
Thetis_1905_111219_269_b3

Re: 2.6.9

Posted: Thu Dec 12, 2019 11:30 pm
by ramdor
VA2EY wrote:I just checked (many more times) Thetis_2.6.8 any alpha version
Thetis_0700_171119_269_a7
Thetis_2240_181119_269_a8
all these version works properly

problem only with
Thetis_2320_211119_269_b1
Thetis_1622_091219_269_b2
Thetis_1905_111219_269_b3


Issue found. New version out sometime over weekend.

Thanks for the version info, helped me track it.

Richie.

Re: 2.6.9

Posted: Thu Dec 12, 2019 11:55 pm
by KB1YFD
Hi Richie,

Thanks for all the effort put into this software. I have been following and upgrading with every release. I'm happy to report that everything has been working fine with my 7000DLE S/N-13. I also own a Flex-6500 and have been comparing the two radios. The UI for the Flex ig pretty good, but your improvements have been closing the gap. I look forward to every new release. Do you have a PayPal account or another way I could show my appreciation?

Cheers, John KB1YFD

Re: 2.6.9

Posted: Fri Dec 13, 2019 12:50 am
by ramdor
KB1YFD wrote:Hi Richie,

Thanks for all the effort put into this software. I have been following and upgrading with every release. I'm happy to report that everything has been working fine with my 7000DLE S/N-13. I also own a Flex-6500 and have been comparing the two radios. The UI for the Flex ig pretty good, but your improvements have been closing the gap. I look forward to every new release. Do you have a PayPal account or another way I could show my appreciation?

Cheers, John KB1YFD


your message alone is enough appreciation. There are many people that have put way more time into all of this, all done for the love of it in most cases, including all the community feedback . Just glad you are enjoying it as it progresses along :D

73 Richie.

Re: 2.6.9

Posted: Fri Dec 13, 2019 6:20 am
by dl6eat
Thanks Richie for picking that one up so quick!

Your preview looks great - "motion sicknesss" should be solved.... :lol: :lol:

@all Have a great Christmas time!

73,

Andy

Re: 2.6.9

Posted: Fri Dec 13, 2019 6:45 am
by ramdor
UPDATE

Hi all,

Please try : link removed, see post #1

This is a pre b4 release, as some of the items are still work in progress. Hopefully will fix the issue Michel is having with PS attenuation.

73 for now, Richie.

(11/19/19) b3_1
-add: vsync option for directx screen updates, initial pass (wip still work to do)
-change: directX provided with fps setting during initialisation, was previously always maxed to 144
-fix: issue with PS NeedToRecalibrate check, did not consider existing TX attenuation if the sampled signal was too low. This was an error introduced in recent refactor
-add: peak hold delay on peak blobs, initial pass (wip still work to do)

Re: 2.6.9

Posted: Fri Dec 13, 2019 11:20 am
by dl6eat
Quick feedback Richie,

First of all "Many thanks for the averaging update on the blobs!"….looks great now.


Using b3_1 riight now under test. My 7000 freezes from time to time - lots of red SEQ Errors underneath the waterfall.
Two tone test gets interrupted - no stable operation at present.

Using an Intel Core i5 (NUC) with direct ethernet connection to my ANAN (no router)

Anybody else observing this behaviour?

Thanks 73,

Andy

Re: 2.6.9_b3_1

Posted: Fri Dec 13, 2019 12:11 pm
by DL8LAQ
I am testing b3_1 on 10m calling cq with PS-A on. For the very 1st time I have got -v on the SEQ log.

DCC0
s0=0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 -7 0 0 0 0 0 0 0 0 r:210985 l:210991 12/13 13:06:58:007

I believe I have installed fw pre4, not sure :-)

Thanks you Richie for all the nice features!

Re: 2.6.9

Posted: Fri Dec 13, 2019 4:10 pm
by VA2EY
ramdor wrote:UPDATE

Hi all,

Please try : https://www.dropbox.com/s/sjzmbx14w0oyy ... 1.zip?dl=0

This is a pre b4 release, as some of the items are still work in progress. Hopefully will fix the issue Michel is having with PS attenuation.

73 for now, Richie.

(11/19/19) b3_1
-add: vsync option for directx screen updates, initial pass (wip still work to do)
-change: directX provided with fps setting during initialisation, was previously always maxed to 144
-fix: issue with PS NeedToRecalibrate check, did not consider existing TX attenuation if the sampled signal was too low. This was an error introduced in recent refactor
-add: peak hold delay on peak blobs, initial pass (wip still work to do)


PS attenuation works well now and I like the peak hold delay on peak blobs
Thanks Richie

Re: 2.6.9

Posted: Fri Dec 13, 2019 7:13 pm
by ramdor
dl6eat wrote:Quick feedback Richie,

First of all "Many thanks for the averaging update on the blobs!"….looks great now.


Using b3_1 riight now under test. My 7000 freezes from time to time - lots of red SEQ Errors underneath the waterfall.
Two tone test gets interrupted - no stable operation at present.

Using an Intel Core i5 (NUC) with direct ethernet connection to my ANAN (no router)

Anybody else observing this behaviour?

Thanks 73,

Andy


1) reboot power/cycle everything, if still happening try 2)
2) check cpu load, as a shot in the dark perhaps try to reduce the work Thetis is doing :

main DisplayFPS: 30
Fill Panadaptor: OFF
DriverEngine: DirectX
DriverThread: Nornmal
Anti Aliased: OFF
Vsync (wip): OFF
Accurate Frame Timing: OFF
Peak Blobs: OFF
RX1/2 SampleRate: 192000
FFT size: 5.859
Waterfall Update Period: 20 or greater
disable picDisplay: Checked
Turn off spotting
Disable Cat control & midi

as a final test set the spectral display of OFF also try with VAC's turned off

If it is still happening then perhaps some hardware/firmware timing issue? Has it just started happening with b3_1? If so then I am at a loss.

Also, check in C:\Users\<user>\AppData\Roaming\OpenHPSDR\Thetis to see if there is an ErrorLog.txt file. If so attach to reply post.

73 Richie.

Re: 2.6.9

Posted: Fri Dec 13, 2019 8:35 pm
by w4cm
PS working great with b3_1 on 7000DLE.
My short wish list would be the ability to choose a different color for SPOT stations than being the same color as grid text color.
Implement on main display Rx/Xmtr meter for VFO A / Rx 1 and additional functional meter for VFO B/ Rx2 that works with function of multirx and Rx2 activation.

Thanks Richie

W4CM/ Cecil

Re: 2.6.9

Posted: Sat Dec 14, 2019 10:21 am
by dl6eat
Thanks Richie for your in-depth reply.

I tried a lot this morning - but realized a stupid behaviour which couldn't have anything to do with Thetis and its current Version at all.
Thetis did not behave stabil but when removing the PTT line to my PA-set the problem disappeared :roll:
So there were obviously some RF inrush problems and Thetis stopped working because of that.

I have to say that I moved an old PC out the shack yesterday and installed various new cables round the new one - but that new PC was never running Thetis....strange.
So funny behaviour…..

I can only apologize since my own stuff kicked me into orbit...…
Without the amp ptt cable I tested again Version 3_1 this morning and used manual ptt to key the PA.
All fine - all smooth...….

Thanks anyway - especially for your good work . Enjoy some good days off business over Christmas!

73,

Andy

Re: 2.6.9

Posted: Sat Dec 14, 2019 10:27 am
by ramdor
ok Andy, good stuff, glad you found the issue as it was a bit of a headscratcher, like I was scratching my head thinking what I could have broken :lol: :D

Richie.

Re: 2.6.9

Posted: Sat Dec 14, 2019 2:57 pm
by W1JA
Here's another instance of negative sequence error.
DCC0
s0=0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 r:718870 l:718868 12/14 08:56:09:057
s1=0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 -4 0 0 0 0 0 0 0 0 r:208268 l:208271 12/14 08:41:51:609

Should we continue posting these here? :)

Re: 2.6.9

Posted: Sat Dec 14, 2019 4:09 pm
by w-u-2-o
IMHO at this point, unless you are attempting to use the built-in Wireshark capture utilities to look at these events for further analysis, there really isn't much benefit in the seq error log other than to tell you if you've got some sort of horrible problems with your LAN or PC delivering data. UDP data is lossy, and therefore there will always be a certain amount of positive seq errors. Similarly, there appears to be some sort of strange issues with either the firmware and/or the PC networking drivers such that there are a certain amount of negative seq errors. In either case, if the results are not audible, they can be safely ignored.

Re: 2.6.9

Posted: Sun Dec 15, 2019 5:44 am
by vk1hx
Gary_Sitton wrote:Anyone with a 200D having startup issues with latest Thetis 2.6.9 beta,
I have my own version which I built that I and others use without any issues
on the 200D and my 7000DLE. I can send you a .zip file with my (unofficial)
version 2.6.9.1 which I am developing and using for my SDR DSP research.


Hi Gary,

I've been trying to send you a PM without success, it sits in the outbox here on the form for some reason.
I'm interested in your development of the Thetis for the 200D. I run the 200D Rev.24 and would like to test.

As the PM doesnt seem to be working, my email address on QRZ is good.
Regards
Phil - VK1HX

Re: 2.6.9

Posted: Sun Dec 15, 2019 2:36 pm
by w-u-2-o
Finally back in the saddle for a few days. My report follows...
ramdor wrote:(11/19/19) b3_1
-add: vsync option for directx screen updates, initial pass (wip still work to do)
Did not note any changes on or off, understand this is WIP.
-fix: issue with PS NeedToRecalibrate check, did not consider existing TX attenuation if the sampled signal was too low. This was an error introduced in recent refactor
Never saw this problem myself, still don't see it.

Also:

The "Signal History" setting strangely compelling. It provides a nice presentation of the dynamic range of a signal. This is also an indication of "loudness". And the ability to set the color and transparency is brilliant. I'm using the "Edge" meter, BTW. Having user accessible settings in milliseconds for min. and max. hold time would be most welcome.

Accurate Frame Timing cleans up the waterfall beautifully, but almost exactly doubles CPU usage (which remains evenly distributed over all cores).

Re: 2.6.9

Posted: Sun Dec 15, 2019 7:47 pm
by ramdor
Hi Scott,

The vsync stuff is part of DirectX, and I am still testing. Certainly on a 60hz display, if you select 100fps desired and then click it on it will drop the fps count (top left) to 60. I have noticed oddities regarding vsync on my 4k display when I turn off DP1.2 support and can only reach 30Hz.

I have also noticed that selecting 30fps desired on a 30Hz display gives a much smoother waterfall. A waterfall update period of 33ms or less (probably quite a bit less) should also be used if you want a jitter free wf, as this will ensure the wf is updated every single frame. Waterfall is updated at the fundamental frame rate, however it is delayed if your waterfall update period is greater than that fundamental frame rate timing. Using a waterfall delay of anything less than 1000/fps will always ensure the waterfall is drawn each frame. There is no way as far as I can tell to apply a waterfall update delay accurately, and imo should be changed to be based as a multiple of frame delay.

Yes there was an issue with the PS code. It would only show up if attenuation applied had been previously set to 0, or if it had reached 0 at some point. Resolved now though.

Ref the signal history. It is exactly that. It matches and follows the signal, and plays out over 2 seconds. The number of records stored is currently 2000ms / meter delay, where meter delay is minimum of analogue/digital refresh. I can put in a configurable play time, but it is 'recording' more than just a min/max so having a hold for each end does not really apply. Unfortunately when a signal moves from low>high the weighting on the average is such that it moves very quickly (to give responsive smeter) and as such history on that side seems to move 'quicker' than the history when the signal is falling. However, I can add a far simpler min/max version that just stores each end, with hold times etc. at some point.

Yep accurate frame timing, is somewhat of a cpu burner ;)

Richie.

Re: 2.6.9

Posted: Sun Dec 15, 2019 9:10 pm
by W1JA
Hi Richie,

Re Signal History
There's a transient on T/R transitions that causes the S meter to start either at 0 or full scale for an instant at the start of each receive period. This makes the history shading drag up from 0 or down from full scale. Perhaps there could be a delay of some milliseconds before the history starts tracking and displaying.

73, John

P.S. I gotta thank you again for all you're doing for Thetis. Your work is awesome.

Re: 2.6.9

Posted: Sun Dec 15, 2019 9:24 pm
by ramdor
Hi John,

Yes I know this issue is there, but have not had a chance to investigate. It could be the averaging of the meter data, or it could be the data coming out of WDSP.CalculateRXMeter. Not sure yet, but is on my annoyance list ;) A delay could hide it, but would be nice to find the underlying cause.

Cheers,

Richie.