2.6.8

[w9ac]

Edit: in other words, save tx profile to make this 'stick', or at least the tx profile parts

Richie.

Working correctly now after saving the profile. Many thanks!

Paul, W9AC

[Jay]

Hi,
Been playing with C5 and had a lock up on CW with P2 version 2 firmware. Twice in the middle of a pileup. Even resetting the database didn't help, the CW tone and TX would still be locked up. Even after reset and rebooting the computer. So I went back to 1.9 and that seems to help on CW lockup, but now I have a new problem with PS.

When I have PS engaged, there is a PTT delay on TX and then another delay when I release PTT. So I am wondering if there are some PS delay settings I am not aware of? When PS is off, there is no delay. You can even hear the PS relays inside the 8000DLE delayed when doing PTT, which is weird.

I love the new looks! Been learning a lot reading the community subject of 2.6.8.

Thanks everyone! 73 de w5jay/jay..

[cLicari]

Been running c5 with P2 v1.9 but starting to have intermittent problems with Thetis locking up. Sometimes I'll see a few "seq" errors and then it will lock up. Other times no "seq" errors but an audio cutout then it locks up. I switched to P2 v1.7 with no improvement, more lockups, but no "seq" errors. This is same scenario I had with my first try with P2. All was good for about three months then started having intermittent disconnects which forced me back to P1, which runs fine. FYI, the Anan has it's own dedicated network subnet.

Carl
NX5T

[NC3Z]

Have you tried the newer v2.0 firmware?

[cLicari]

Gary... No, does it run on the 8000DLE?
I don't see it listed in GitHub

[NC3Z]

It runs on the Orion II's which are the 7000 and 8000. I use a 7000. It has been posted here in the forum a couple of times before, let me look for a link.

[NC3Z]

Here is a message from N1GP with a link to it:

viewtopic.php?f=32&t=3075&p=7419&hilit=pre1#p7419

[cLicari]

Thanks Gary.
I'll give it a try before going back to P1, which I hate to need to do.

Carl

[NC3Z]

On our eve 160M net there are several of us running 7000 and 8000's with 2.6.8 (currently c5) and 2.0pre1 firmware and none of us are looking back at P1. Thetis is working so much better for all of us.

[Joe-W4WT]

I have an occasional problem and I wonder if anyone else is experiencing it. I'm running 2.6.8 C5 along with v2.0 pre1 firmware on an 8000 board.

I use VAC with Potato and sometimes when I start Thetis, my overflows/underflows just plain run away. I find I can usually fix this by simply unchecking and then rechecking the Buffer Latency Ringbuffer "Manual" check box. This happens regardless of having any other apps running or not. It doesn't happen every time I start Thetis but it does happen frequently. Once I "fix" it, it usually stays fixed, but not always. When running correctly, I usually only see the last two digits changing in the "Var Ratio" under both To Vac and From Vac so things seem pretty stable based on this.

It's not a big problem but it makes me wonder if it is somewhat common or if maybe I have some setting(s) that are incorrect.

Joe W4WT

[cLicari]

Well today I have tried Thetis c5 with P2 v1.7, v1.8, v1.9 (which has been working fine until today), v2.0 pre-beta. None of them are working well enough to use for any length of time, let alone reliably. I'm beginning to think it may be the current Thetis c5 and not the firmware. Don't know what could be causing all these "seq" errors and disconnects all of a sudden. I'll take any and all suggestions.
Thx
Carl
NX5T

[ramdor]

Hi all,

Due to unforeseen gravitational forces, I descended a set of green slimy outdoor wooden stairs somewhat quicker than originally planned. Consequently, coding and 'fiddling' with Thetis is going to have to sit on the back burner until I can comfortably sit down, and use my left arm again

In the mean time, please try d1 exe here : https://www.dropbox.com/s/r0f1ns3dwx39t ... 1.zip?dl=0

Note: there is still work to do on the region stuff, and will not be complete as it stands. Also, c5 introduced a change where txprofile is applied after settings are loaded in, this may result in user not thinking changes are saved. As long as a tx profile is saved, everything will behave as expected.

73 Richie.

(10/7/19) d1
- fix: small mode/filter labels in vfo boxes now reflect state of multirx correctly
- fix: band stack issue with Region3/Japan would prevent Thetis from restarting
- change: 60m slices will highlight if vfo is over them, and in acceptable mode (no check to see if filter is outside permitted area in this highlight)
- fix: if zoom is so large and a filter is unable to fit within available display, dragging it in CTUN mode will perform a reversed spectrum drag
- change: added Region 1/2/3, Germany, to region list. This will effect the band edges shown in red on panadapter and text shown under vfo
- change: selecting a tx profile no longer changes power, power is stored per band and doing both caused inconsistencies

[ramdor]

Well today I have tried Thetis c5 with P2 v1.7, v1.8, v1.9 (which has been working fine until today), v2.0 pre-beta. None of them are working well enough to use for any length of time, let alone reliably. I'm beginning to think it may be the current Thetis c5 and not the firmware. Don't know what could be causing all these "seq" errors and disconnects all of a sudden. I'll take any and all suggestions.
Thx
Carl
NX5T

you can get all previous versions here, https://github.com/w5wc/Thetis/releases/tag/v2.6.8 ... c5, c2, b8, b5, and even builds before that.

I don't think anything has changed under the hood that would make this happen 'all of a sudden'.

edit: personally I would try direct link to radio, and perhaps also different ethernet port/card in computer. I was listening to a station on 80m the other day saying he was having issues with Realtec Ethernet port, as soon as he moved over to an intel one all the problems went away, on same pc (red pitaya and openhpsdr I think it was).

Richie.

[nico]

Due to unforeseen gravitational forces, I descended a set of green slimy outdoor wooden stairs somewhat quicker than originally planned. Consequently, coding and 'fiddling' with Thetis is going to have to sit on the back burner until I can comfortably sit down, and use my left arm again

Wishing you a very speedy recovery and thanks for your fantastic work - you have made me like Thetis again!

[cLicari]

Well today I have tried Thetis c5 with P2 v1.7, v1.8, v1.9 (which has been working fine until today), v2.0 pre-beta. None of them are working well enough to use for any length of time, let alone reliably. I'm beginning to think it may be the current Thetis c5 and not the firmware. Don't know what could be causing all these "seq" errors and disconnects all of a sudden. I'll take any and all suggestions.
Thx
Carl
NX5T

you can get all previous versions here, https://github.com/w5wc/Thetis/releases/tag/v2.6.8 ... c5, c2, b8, b5, and even builds before that.

I don't think anything has changed under the hood that would make this happen 'all of a sudden'.

edit: personally I would try direct link to radio, and perhaps also different ethernet port/card in computer. I was listening to a station on 80m the other day saying he was having issues with Realtec Ethernet port, as soon as he moved over to an intel one all the problems went away, on same pc (red pitaya and openhpsdr I think it was).

Richie.

Richie...
So sorry to hear of your misstep. Hoping you have a speedy recovery. Also, thanks for all your hard work at making Thetis what it is today and will be tomorrow.
Re: the network. The Anan is on it's own subnet and direct to the computer. However, the Ethernet card IS a Realtek TP-Link Gigabit PCI Express card. But it works without issue with P1 and OpenHPSDR 3.6.9. Would you still recommend switching to an Intel card?
Thx
Carl
NX5T

[w2ner]

Richie

Get better, I know what stuff like that feels like, I fell off my roof a few years back. Take it easy and kick back.

[Helmut]

Well today I have tried Thetis c5 with P2 v1.7, v1.8, v1.9 (which has been working fine until today), v2.0 pre-beta. None of them are working well enough to use for any length of time, let alone reliably. I'm beginning to think it may be the current Thetis c5 and not the firmware. Don't know what could be causing all these "seq" errors and disconnects all of a sudden. I'll take any and all suggestions.
Thx
Carl
NX5T

There are two Versions of V2.0pre-beta:

Orion_MkII_Protocol_2_v2.0_pre1.rbf 1.694 KB
Orion_MkII_Protocol_2_v2.0_pre2.rbf 1.692 KB - we got the link now down in the thread - thank you Rick !

I run Thetis 2.6.8 c5 (from today d1) with pre2 and never had such a good tranceiver with the 7000dle-mk2 - it´s a dream !

TNX Richie and get well soon !

73 Helmut dl9eri / oe9eri

[ramdor]

Re: the network. The Anan is on it's own subnet and direct to the computer. However, the Ethernet card IS a Realtek TP-Link Gigabit PCI Express card. But it works without issue with P1 and OpenHPSDR 3.6.9. Would you still recommend switching to an Intel card?

honestly, I dont know, it is just interesting that they they were talking about something similar. A coincidence though perhaps? If you could borrow a nic or try another computer/laptop with intel chipset? I wonder if anyone else has similar findings?

[w-u-2-o]

I have an occasional problem and I wonder if anyone else is experiencing it. I'm running 2.6.8 C5 along with v2.0 pre1 firmware on an 8000 board.

I use VAC with Potato and sometimes when I start Thetis, my overflows/underflows just plain run away. I find I can usually fix this by simply unchecking and then rechecking the Buffer Latency Ringbuffer "Manual" check box. This happens regardless of having any other apps running or not. It doesn't happen every time I start Thetis but it does happen frequently. Once I "fix" it, it usually stays fixed, but not always. When running correctly, I usually only see the last two digits changing in the "Var Ratio" under both To Vac and From Vac so things seem pretty stable based on this.

It's not a big problem but it makes me wonder if it is somewhat common or if maybe I have some setting(s) that are incorrect.

Joe W4WT

Joe,

This is a bug in the resampler algorithm and has nothing to do with Thetis or P2 per se. I experience it here as well with both PowerSDR and Thetis. I suspect you are shaving down your buffer sizes to a minimum because I have found that doing so makes the resampler algorithm more susceptible to this behavior.

73,

Scott

[w-u-2-o]

Well today I have tried Thetis c5 with P2 v1.7, v1.8, v1.9 (which has been working fine until today), v2.0 pre-beta. None of them are working well enough to use for any length of time, let alone reliably. I'm beginning to think it may be the current Thetis c5 and not the firmware. Don't know what could be causing all these "seq" errors and disconnects all of a sudden. I'll take any and all suggestions.
Thx
Carl
NX5T

Carl,

This behavior is typically indicative of firmware not making timing on your hardware and/or some kind of network problem. Firmware is the more likely culprit. Or are you saying this only happens with C5?

73,

Scott

[w-u-2-o]

Hi all,

Due to unforeseen gravitational forces, I descended a set of green slimy outdoor wooden stairs somewhat quicker than originally planned. Consequently, coding and 'fiddling' with Thetis is going to have to sit on the back burner until I can comfortably sit down, and use my left arm again

So, so sorry to read that, Richie Wishing you a very speed recovery. If there is anything you need, please don't hesitate to reach out!

73!

Scott

[n1gp]

Here's a link to the pre2:

https://www.dropbox.com/s/i1fony74jallw ... 2.rbf?dl=0

73's

-Rick / N1GP

[cLicari]

Well today I have tried Thetis c5 with P2 v1.7, v1.8, v1.9 (which has been working fine until today), v2.0 pre-beta. None of them are working well enough to use for any length of time, let alone reliably. I'm beginning to think it may be the current Thetis c5 and not the firmware. Don't know what could be causing all these "seq" errors and disconnects all of a sudden. I'll take any and all suggestions.
Thx
Carl
NX5T

Carl,

This behavior is typically indicative of firmware not making timing on your hardware and/or some kind of network problem. Firmware is the more likely culprit. Or are you saying this only happens with C5?

73,

Scott

Scott...
All was well running P2 v1.9 and Thetis starting with 2.6.7. It wasn't until 2.6.8 c4 that I started to have some "seq" errors and a few disconnects. I upgraded to c5 and started having more "seq" errors and disconnects to the point where it was every few minutes. I then started loading older firmware and finally v2.0 pre1 with the same results. I went back to P1 v2.7 and 3.6.9 later that day and it worked fine. When Richie posted d1 I thought I'd give it a shot. I've seen none of the "seq' errors or disconnects yet but it's only been a few hours of operation. Richie mentioned someone having issues with a Realtek Ethernet card that went away after installing an Intel card so I may try that. It's just odd that everything works well for days/weeks then goes south.

Carl
NX5T

[NC3Z]

Here's a link to the pre2:

Rick, what is the difference between Pre1 and Pre2 ?

[W1AEX]

Re: the network. The Anan is on it's own subnet and direct to the computer. However, the Ethernet card IS a Realtek TP-Link Gigabit PCI Express card. But it works without issue with P1 and OpenHPSDR 3.6.9. Would you still recommend switching to an Intel card?

honestly, I dont know, it is just interesting that they they were talking about something similar. A coincidence though perhaps? If you could borrow a nic or try another computer/laptop with intel chipset? I wonder if anyone else has similar findings?

Richie and Carl,

Initially, when I set up my second subnet for direct connection to my ANAN I used the motherboard's integrated Realtek network interface port. It worked fairly well but I randomly experienced intermittent pauses now and then when running Thetis. At the time there were some discussions about whether the Intel NIC interfaces were a more viable option due to better drivers plus the ability to optimize the connection with all the settings in the Intel NIC's properties. Being adventurous, I disabled the onboard Realtek NIC in my motherboard's bios and bought and installed an Intel NIC and the pauses with Thetis went away. So... that's my non-scientific anecdotal experience with Realtek and Intel NIC adapters. It might help with pauses, freezing and possibly some dropouts. I have not seen any seq errors with Thetis version 2.6.8 while running the Orion (200D) v1.8 firmware from Rich-N1GP so I'm not sure if the Intel NIC will do anything to stop the "seq" errors.

Rob W1AEX

[ea3aqr]

Well today I have tried Thetis c5 with P2 v1.7, v1.8, v1.9 (which has been working fine until today), v2.0 pre-beta. None of them are working well enough to use for any length of time, let alone reliably. I'm beginning to think it may be the current Thetis c5 and not the firmware. Don't know what could be causing all these "seq" errors and disconnects all of a sudden. I'll take any and all suggestions.
Thx
Carl
NX5T

Carl,

This behavior is typically indicative of firmware not making timing on your hardware and/or some kind of network problem. Firmware is the more likely culprit. Or are you saying this only happens with C5?

73,

Scott

Scott...
All was well running P2 v1.9 and Thetis starting with 2.6.7. It wasn't until 2.6.8 c4 that I started to have some "seq" errors and a few disconnects. I upgraded to c5 and started having more "seq" errors and disconnects to the point where it was every few minutes. I then started loading older firmware and finally v2.0 pre1 with the same results. I went back to P1 v2.7 and 3.6.9 later that day and it worked fine. When Richie posted d1 I thought I'd give it a shot. I've seen none of the "seq' errors or disconnects yet but it's only been a few hours of operation. Richie mentioned someone having issues with a Realtek Ethernet card that went away after installing an Intel card so I may try that. It's just odd that everything works well for days/weeks then goes south.

Carl
NX5T

Same problem here with seq errors. Lots of seq errors on c5 and less on c4.

I'm using an Anan 7000dle with P2 2.0 pre1 FW and an Intel NIC. When I tested, some time ago, P2 2.0 pre 2 FW I was getting more seq errors, than using P2 2.0 pre 1.

[n1gp]

@Gary NC3Z

Honestly I don't remember for sure. But I'm pretty sure it was just changing the
timing of the tx/rx data going through the ethernet PHY chip.

That setting alone makes or breaks timing for some radios.

[Joe-W4WT]

"This is a bug in the resampler algorithm..."


Thanks Scott. I'll play with larger buffer settings to see how that might affect it. I run 128 currently. I guess that algorithm is in WDSP.

Joe W4WT

[w-u-2-o]

Joe,

I live with the bug because I love having very like latency. If I have to occasionally beat the resampler into submission it's the cost of doing business right now.

Scott

[vk1hx]

.