Increase cpu load causing mount disconnect?

[KenHarpster]

Robin,

I have attached a Windows created compressed .zip of the entire nights sessions for your use. I see several that are large, but one in particular. I included the others in case there was an incident that started the problem but did not actually crash.

I have installed the latest version and will do some testing on my end.

Thank you!

Ken

[Beepa]

Robin, thank you. I am very excited to exercise your modification. Unfortunately, it's expected to be cloudy here for several nights, but I will report back as soon as I get results. It's so great to get support so quickly.
Cheers, Regards,
Dave

[admin]

Hi folks,

@Dave - if you wanted to test in advance, leaving the camera running for several hours at 4s exposure should (I'm fairly confident) show the problem with the increased CPU usage if it is going to still happen (It should also show it in the old version). That's the sort of thing that can be done on the tabletop indoors without waiting for clear skies. I think that all of the sequence code was just a way to make that long term imaging happen with a high frame turnover but a long enough exposure to show the frame progress.

@Ken - you had a different issue - at 22:44:08 or thereabouts, the display rendering package that SharpCap uses to draw the user interface on screen failed. What happens in this situation is that one part of the display software crashes, and that causes SharpCap to freeze up the next time it has new stuff ready to draw onto the screen. Fortunately this situation is pretty rare, but unfortunately there is no definitive cure. Things that may help are making sure that you have all available windows updates installed on your SharpCap PC, along with any updates available to the graphics card driver (go to the chip manufacturer site - AMD, NVidia or Intel - for the latest drivers, rather than PC manufacturer site).

The graphics crash bug is annoying - it's not a direct response to something that SharpCap does (as far as I can tell), and it happens randomly on a small number of machines. If I could make it happen to me and had steps to make it happen repeatedly I'd have a chance of working around it, but I only ever caught one situation where it happened reliably, and I fixed that long ago.

cheers,

Robin

[KenHarpster]

Robin,
Thank you for the information! I am running Windows 10 so updates are rare, but I will check the graphics driver for updates. I did some testing yesterday (albeit not exhaustive) and overall the CPU usage was peaking at about 35% while imaging, which is of course relative to the machine. I did not have any disconnects to RDP, mount, or focuser.

As an FYI my machine is a Mele Quieter 3C with 16gb memory, 256gb eMMC, 1tb SSD.

Edit: I used Windows Device Manager to check for updates and lo and behold it installed an update for the Intel UHD Display Adapter. Hopefully that will resolve the issue.

[ChrisR Oz]

Hi Dave,

I tried various ways to drive my AVX mount and EdgeHD8 and found that wifi was very unreliable, and would always drop out at some stage. The trick I discovered on Cloudy Nights was to use the Celestron Focuser as a hub. Connect USB to the Focuser and an Aux cable from there to the mount. This has worked without drop outs for years now.

Cheers, Chris.

[admin]

Hi,

I would certainly go along with Chris, and in fact go further... If you are connecting wirelessly (wifi, bluetooth, etc) between your PC and the mount then you are asking for trouble - use a cable for a big improvement in reliability. The nature of wireless connecitons is that they sometimes drop for no obvious reason, and the ASCOM drivers tend not to have enough patience retrying a failed connection.

cheers,

Robin

[Beepa]

Hi Robin & Chris, Thank you for your insights!! With the fix you provided in Sharpcap, my cpu no longer shows excessive usage, but unfortunately the mount disconnects have continued. I may have to (reluctantly) resort to the cabled approach. It's good to know there is a good option. Nonetheless, I've encountered the curious effect with wifi that mount disconnects occur with my newer faster HP, but do not seem to occur with my old slow Dell running windows 10. I need to confirm this with more measurements if the clouds ever cooperate.
Cheers,
Dave

[admin]

Hi Dave,

well, good to hear that the CPU issue is fixed, shame about the disconnects still being there. I wonder if Windows logs anything useful about WIFI state changes (disconnect/reconnect) in the 'Event Viewer' logs? If it does then maybe you could see if there is any correlation between disconnects and relevant log messages. Alternatively you could leave a command prompt open with 'Ping' running - continually checking connectivity to some known system (you can ask ping to just keep going with the -t option) - again, if you get missing ping responses when the mount glitches then you have something interesting.

cheers,

Robin

[Beepa]

I finally gave up on the wifi link between my remote laptop & Celestron mount at the suggestion of Robin and others. I had tried everything from battery heaters, dedicated router, circuit board coating to avoid humidity issues, elimination of other interference, different power supplies, etc but mount would still usually disconnect after a while. I bought a minicomputer for $130 & taped to telescope. I connect from mount to Celestron focuser & then to minicomputer via short usb cable. Two cameras and rotator also connect directly to minicomputer with short usb cables. Then I use wifi to connect between my minicomputer and my remote laptop. What a phenomenal improvement!! No failures whatsoever, even when temperature dropped to 5 deg F. And I no longer have to run cables from camera, rotator, or mount to my remote laptop. SharpCap runs great on minicomputer, no cables to remote laptop, great performance, big smile!!

[Beepa]

I should clarify that Robin didn't suggest giving up on wifi, but rather he felt I would likely get a more consistent connection if I hard wired the mount to the computer.