Thetis v2.8.11 - UPDATES and current issues

[I2NOY]

...and also remember at startup how was the "TRACK" function when you closed the program. This on DX spotting form.
If possible, also enable a filter of spots only coming from Europe.

[w-u-2-o]

Ernst

There is no way for Richie to do that. The FPGA receives no signal that tells it what the source of the 10MHz is so the firmware can not know and therefore the software can not know either. The circuit that automatically selects the source of the 10MHz reference lies completely outside of the FPGA.

It would be possible to modify your board to add an LED indicator, but I don't know anyone who has actually done that.

73,

Scott

[oe3ide]

Scott, thanks. So we trust our GPSDOs

[FM5GB]

hi

This is easy to check by listening to WWV or any known (good) frequency standard.

[ramdor]

Just a quick question.

The current band stack system has 10 slots say per band stack. You can store whatever frequencies for the band you want in those slots. They can be in any frequency order, so when you cycle, the vfo can jump all over the place, up/down the band. You can hit the sort button to resolve this.

The new system has no slot limit. The stack presented is a filter applied to all entries. Entries have no idea of 'stack'. To display the stack of entries for 80M a filter is defined that selects entries based on band and frequency limits. An entry could technically be included in the result of another filter.

It is far easier for the new solution to present the stack pre-sorted. Anything you manually add will be included in the results of the filter and used as the bandstack in a sorted way. If you want me to implement a user adjustable sorting to the list, that that is going to be a whole bunch of additional work (and pain).

So what are the thoughts, do people use unsorted bandstack lists and jump all over the place? Or having it all sorted is an ok compromise?

bandstack2.jpg (40.19 KiB) Viewed 11492 times

Richie.

[oe3ide]

Good morning Richie,

wow...that looks nice!
For me, the sorted list is 100% ok!

Cheers

[ramdor]

Good morning Richie,

wow...that looks nice!
For me, the sorted list is 100% ok!

Cheers

It is working nicely at the moment Ernst. Just ironing out a couple of oddities, then onto saving/loading the data with the database. Also an import procedure that happens when migrating from old db version.

Then to build a manager so you can add your own filters.... I should have kept on with the bugs for a week more instead of starting this, but I needed a change.

Not sure it will be ready for weekend at this rate..... must.. type..faster !

Richie.

[oe3ide]

Richie, do as some managers (at least some of my managers) suggest. Take this plugin, take another plugin, put it together and you're done! Software development so easy

Ok, back to serious, don't forget the "always on top" option

I totally understand...sometimes finding bugs and try to fix them is like stabing into a wasp's nest or dealing with the Keymaker in Matrix

tenor.gif (445.7 KiB) Viewed 11477 times

[rbduck]

Just a quick question.

The current band stack system has 10 slots say per band stack. You can store whatever frequencies for the band you want in those slots. They can be in any frequency order, so when you cycle, the vfo can jump all over the place, up/down the band. You can hit the sort button to resolve this.

The new system has no slot limit. The stack presented is a filter applied to all entries. Entries have no idea of 'stack'. To display the stack of entries for 80M a filter is defined that selects entries based on band and frequency limits. An entry could technically be included in the result of another filter.

It is far easier for the new solution to present the stack pre-sorted. Anything you manually add will be included in the results of the filter and used as the bandstack in a sorted way. If you want me to implement a user adjustable sorting to the list, that that is going to be a whole bunch of additional work (and pain).

So what are the thoughts, do people use unsorted bandstack lists and jump all over the place? Or having it all sorted is an ok compromise?

bandstack2.jpg

Richie.

The new bandstack looks really good. I also have had no issues with 21c. Plus the gradient really throws a pleasing look to Thetis.

[W1JA]

no worries Ernst. The cancel thing is far from ideal.

I'm glad Ernst posted about the Cancel button problem. It's what I was trying to describe in my post on page 1, where I said

The cancel button resets the entire database back to the end of the previous session, so all operating changes in the current Thetis session are lost. It should just back out of the changes made in the current Setup session, only.

I also made a post in April 2020 ( https://apache-labs.com/community/viewtopic.php?f=9&t=3387&p=10888#p10888 ) called "Delete the Cancel button," which in my opinion would be an acceptable fix for this issue. Just get rid of it!

However, the fix you propose seems like it would work. If you choose this approach I look forward to testing it.

73, John

[ramdor]

However, the fix you propose seems like it would work. If you choose this approach I look forward to testing it.

73, John

Yep has been implemented and to be included in 21d.

Richie

[ramdor]

If region is set to US in the current version Thetis locks you to specific frequencies on 60m each time you press the 60m button. It ignores the stack information for the 60m band totally, It does not allow you to use more entries in the stack. It just cycles through the five channels. Essentially the code is: If region is US then do channel based VFO update on 60m button click, otherwise use stack system.

I can implement this for the 60m band if in the US if you guys want to be locked down and limited to that, but blimey, dont you want to be able to define your own stack entries for the band? I can add the defaults channels already locked easy enough, but they could be changed by unlocking them. If for example, you wanted some extras, or the locked ones moved for different narrow modes.

Essentially, I would have to disable lock/delete/add buttons if region = US and on 60m band, so that the 5 default entries could not be changed.

What say you ?

Richie.

[Joe-W4WT]

"but blimey, dont you want to be able to define your own stack entries for the band? "

I vote for versatility over fixed and locked. You never know when the FCC might decide to change allocations or other rules on 60m and if things are locked like they are now then everyone would be stuck if you or another volunteer programmer wasn't around to change it when it happened.

Thanks for all your work Richie, it is really appreciated!!!

Joe W4WT

[w-u-2-o]

Richie,

This is just my opinion, but 60M channelization for the US is so simple, I don't mind just clicking 5 times to roll through the entire band. Indeed, I've set up the current version so I see all 5 channels on the panadapter in all 5 60M bandstack entries, so I just watch the passband move from one channel to the next. That's how I operate on 60M, no doubt others have their own style.

To summarize: IMHO it's not worth a lot of effort to change the way 60M works now, but again others may differ.

To be my own devil's advocate: on the other hand, having the software function in a common manner on all bands provides a level of consistency that may be worth it by itself.

Thanks!

Scott

[rbduck]

If region is set to US in the current version Thetis locks you to specific frequencies on 60m each time you press the 60m button. It ignores the stack information for the 60m band totally, It does not allow you to use more entries in the stack. It just cycles through the five channels. Essentially the code is: If region is US then do channel based VFO update on 60m button click, otherwise use stack system.

I can implement this for the 60m band if in the US if you guys want to be locked down and limited to that, but blimey, dont you want to be able to define your own stack entries for the band? I can add the defaults channels already locked easy enough, but they could be changed by unlocking them. If for example, you wanted some extras, or the locked ones moved for different narrow modes.

Essentially, I would have to disable lock/delete/add buttons if region = US and on 60m band, so that the 5 default entries could not be changed.

What say you ?

Richie.

I vote for the unlocked and unlimited. This is the reason I chose this type of architecture to purchase over others.
I love the virtualally unlimited functionality and ability chose to make it what I want.Richie, Thank for all the time you take out of your life
to add to the fun this hobby affords.

[cLicari]

Richie...

I'm with Scott on this one.
I think your valuable time could be better spent on things more widely used.
Thanks for all your hard work.

Carl
NX5T

[ramdor]

It is much easier to make it more flexible tbh.

One line of code to make it like all other bands VS a load of bodging, checking this and that to make it how it was

If it needs to add them in, they'll appear pre-locked when region is US. They can be unlocked/deleted/updated however.

21d will drop over the weekend, probably Sunday now. Still working through some stuff.

Richie.

[liam]

HI Richie I also am in favour of Unlocked Thank you for all you time
and expertise
73
Liam EI3GC

[kc2rgw]

Another vote for unlocked.

[EA1DAV]

I am using thetis 21C. During IARU contest today I choose focused to N1MM ( Setup/Cat control/Focus master) and Thetis crash. 3 times.. After that I choose 'select by click' and focus work as expected. There is lot of trafic in port 12060, I don't know is this can be the problem.

73 Jesus

[ramdor]

I am using thetis 21C. During IARU contest today I choose focused to N1MM ( Setup/Cat control/Focus master) and Thetis crash. 3 times.. After that I choose 'select by click' and focus work as expected. There is lot of trafic in port 12060, I don't know is this can be the problem.

73 Jesus

was there a crash exception message?

Check ErrorLog.txt file in C:\Users\<user>\AppData\Roaming\OpenHPSDR\Thetis-x64 or Thetis folder.

Richie.

[EA1DAV]

Hi Richie:

Here the last crash.

Entry : 5:59:22 PM Saturday, July 10, 2021
Solo se permite un uso de cada dirección de socket (protocolo/dirección de red/puerto)
---------stacktrace------------
at System.Net.Sockets.Socket.DoBind(EndPoint endPointSnapshot, SocketAddress socketAddress)
at System.Net.Sockets.Socket.Bind(EndPoint localEP)
at Thetis.Console.<PollN1MMPacket>d__4326.MoveNext()
--- End of stack trace from previous location where exception was thrown ---
at System.Runtime.ExceptionServices.ExceptionDispatchInfo.Throw()
at System.Threading.ExecutionContext.RunInternal(ExecutionContext executionContext, ContextCallback callback, Object state, Boolean preserveSyncCtx)
at System.Threading.ExecutionContext.Run(ExecutionContext executionContext, ContextCallback callback, Object state, Boolean preserveSyncCtx)
at System.Threading.QueueUserWorkItemCallback.System.Threading.IThreadPoolWorkItem.ExecuteWorkItem()
at System.Threading.ThreadPoolWorkQueue.Dispatch()
-------------------------------

[ramdor]

A quick look at the code in that area seems that it parses a bunch of XML stuff that comes in over UDP. There is no exception handling for the XElement or sub element parsing so it will just crash if that fails. Network socket exceptions are caught but not the xml parsing stuff.

I have just slapped a Try/Catch around it all, in the hope it will recover if something goes wrong. At leas it wont crash now. I could log the exception to the error log to get more details, but that thread is ticking every half a second and data keeps arriving and if the error keeps happening you will get a big log file fast.

Richie.

[EA1DAV]

Thanks Richie, there are alternatives in focus, so I can continue working contest. In contests focus is very apreciate.

best 73

[DL8LAQ]

Hi Richie,

I hate colored lines and filter overlay in the waterfall. Since one of the 21 versions I have to check/uncheck "Waterfall - Show Zero Line" and "Waterfall - Show Filter " after every start of Thetis to get rid of it. I did export database and save transmit profile, but every day I have to check/uncheck...

Thanks for your fantastic work

[DO2ZA Erwin]

Hello Norbert,

i have tested this, no Problems here. May be you forgot to press Apply when you make your Changes ??

73 Erwin

[DF2LH]

Hello Richie
first my thanks and respect for your work. You are really moving Thetis forward with your attention to detail!

I have one small point for your list. It concerns the VAC RX level in DIGU mode. The last value used is saved and displayed after restarting thetis, but not applied. It is only applied after a manual change. See a short demo in this video: https://www.youtube.com/watch?v=l2-lkYgkiww

Best regards
Thomas, DF2LH

[DL8LAQ]

i have tested this, no Problems here. May be you forgot to press Apply when you make your Changes ??

"Apply" doesn't make it better.

[DO2ZA Erwin]

Norbert,

here it works, may be your Database is defekt, have you make a test with a fresh Database?

[oe3ide]

I have one small point for your list. It concerns the VAC RX level in DIGU mode. The last value used is saved and displayed after restarting thetis, but not applied

AFAIK the values are stored in the TX-profile. Try to save the corresponding TX-Profile.

73 Ernst