Thanks so much Richie and keep up the good job

Happy weekend to everyone from Sweden!
It is not. Only on the 7 and 8K.ramdor wrote: V/I is only available on 7000/8000's, not sure if that info is available from other hardware.
Hi Scott, Thanks for info ref the V/I, shame that. I was also wondering if there are any temp sensors/data available.w-u-2-o wrote:Hi Richie,
Running 2.6.9 a3. DiretX, normal priority (and also below normal--no real difference noted).
I no longer can make the red flashing indicator appear. Before if I set the display update rate to 60FPS I would get it a lot. And I can no longer set the update rate above 60.
Did you remove the red indicator, or did you change something such that the code is much more efficient? Sorry if I missed something in the discussion.
Thanks!
Scott
all very fluid really... big pinch of salt required, nothing has been done to perform what looks like 'a 1/3 more work'ea3aqr wrote:Increased CPU use in 2.6.9 a3 release?
After installing the latest release, my Thetis CPU use is about 9%. With previous releases it was always 5-6%.
Is that normal?
Good news on the FW. Yep they do. The packet before 396 was the expected packet, ie, a delta of 0. However, after that packet, the next one was 396 ahead of expected, meaning that there had been a whole load lost. -ve values are the important ones to us at the moment, and as far as I can tell no one has reported any. Negative will mean that the packet arriving has a sequence number lower than expected, so completely out of order, which an implementation of PRO could potentially resolve. As you are not seeing -ve's after the 396 everything recovered and no strange sequence numbers arrived.w2ner wrote:Richie
Now that I have all the issues taken care of I was having with the firmware. I'm seeing SEQ errors.. I did have it dedicated to a local NIC card but messing with getting the firmware fixed, I moved it over to the network. I recall this was the reason I moved it to the dedicated NIC. These errors mean anything to you?
I had been running 30FPS on my 4K monitor since you added the red dot. So seeing this post from Scott I decided to run a 60FPS again, no red dot anywhere. So then I decided to run fully screen and 60FPS on my 4K, which has always been a problem, and no red dot and all very smooth.ramdor wrote: Just an example, if you have Thetis expanded to the full width (height doesn't matter) of a 4k display
I'll do that today. I changed a NIC card setting and enabled "Jumbo Packets" and it seems to have fixed it. I have some things to do in the yard but I will leave the systems running and see if it logs anything. I'll post an update later today.ramdor wrote:w2ner, Nicholas, before you move over to dedicated nic, can you run the following commands in cmd window...
The first gives you a list of interfaces, use te number on the left to pass into the next command to get the data. Screen shot that, and repeat on with the dedicated NIC.
Could provide some important info. Cheers, Richie.
Capture.PNG
set display to OFF, instead of panadaptor or something. At the moment the screen is being redrawn even if no radio connection, so you can actually pan around, zoom etc etc even without radio connected. Also, the display arrays are still iterated through. Not ideal but it is how it is atm.ea3aqr wrote: EDIT: Looks like something is wrong... just opening Thetis without starting the connection to the radio, the CPU use is abut 2.5-3% doing nothing
Very interesting, I didnt change that setting when I was testing earlier in the week.w2ner wrote:I'll do that today. I changed a NIC card setting and enabled "Jumbo Packets" and it seems to have fixed it. I have some things to do in the yard but I will leave the systems running and see if it logs anything. I'll post an update later today.
Ok, setting display to OFF, CPU usage goes to 0%.ramdor wrote:set display to OFF, instead of panadaptor or something. At the moment the screen is being redrawn even if no radio connection, so you can actually pan around, zoom etc etc even without radio connected. Also, the display arrays are still iterated through. Not ideal but it is how it is atm.ea3aqr wrote: EDIT: Looks like something is wrong... just opening Thetis without starting the connection to the radio, the CPU use is abut 2.5-3% doing nothing
Richie.
and if you now connect to the radio with display as OFF, that will show you the work done by everything else in Thetis other than the display, give or takeea3aqr wrote:Ok, setting display to OFF, CPU usage goes to 0%.
1) Panafall mode ON and NOT connected to radio 4.5% CPU use.ramdor wrote:and if you now connect to the radio with display as OFF, that will show you the work done by everything else in Thetis other than the display, give or takeea3aqr wrote:Ok, setting display to OFF, CPU usage goes to 0%.
Here ya go Richie. First one is direct and the second one is on the network.. Also, with the "Jumbo packets" enabled, I did see Seq errors again so, not much of a change. The only thing I saw was, the 396 number in the log went from that to I think 63 (forgot to get a screen shot). On direct there is no errors.ramdor wrote:w2ner, Nicholas, before you move over to dedicated nic, can you run the following commands in cmd window...
The first gives you a list of interfaces, use te number on the left to pass into the next command to get the data. Screen shot that, and repeat on with the dedicated NIC.
Could provide some important info. Cheers, Richie.
Capture.PNG
Here it is on my 8 core AMD Ryzen 7 2700X CPU with 16GB DDR4 and NVIDIA 1050 TX. I am also getting some SEQ errors (attached is the SEQ Log).ramdor wrote:UPDATE
Hi all,
Small update that signifies the start of changes to the interface. Clock panel has been moved to a status bar, which also provides a warning triangle if any seq errors occur.
Setup->Appearance->General where you can change back and text colour of status bar.
Please try : https://www.dropbox.com/s/i3g7y6vdrtc6w ... 4.zip?dl=0
Cheers, Richie.
ULTIMAX wrote:https://apache-labs.com/community/download/file.php?mode=view&id=612
Al
N3YV
73's
Hi nico,nico wrote:Hi
Richie thank you again for doing such fantastic work with Thetis ! I have noticed a DirectX bug that that seems to have started roughly after the drop down for Direct X was added in 2.6.8 and still seems to be an issue in 2.6.9 hence why I am raising it