Player One Uranus-C and Slow Live Stacking

Post Reply
MarMax
Posts: 101
Joined: Sun Sep 19, 2021 11:43 pm

Player One Uranus-C and Slow Live Stacking

#1

Post by MarMax »

I've only used the Uranus-C (IMX585) camera twice and the first night I did not notice any extreme delays between capture and live stacking. Last night was the first time I've experienced delays on the order of 1-2 minutes or more between the capture and adding of images to the live stack. I'm comparing the message box on the right in the bottom banner to the live stacking overview.

Here is a screen grab.
Image

The real impact of the delay between capture and stacking is that the histogram adjustments are basically waiting until the stack updates. During the delay wait the sliders and adjustments do nothing.

I'm wondering if this is a result of the Uranus-C since I've not noticed this amount of delay with either the ASI533 or the QHY183, both of which are larger sensors. I was running gradient reduction which could be bogging things down. I run gradient reduction with the 533 and 183 without any issues. I'm running a 4-port USB 3.0 hub at the mount for the mount, cameras (Uranus-C and ASI290MM guide) and Pegasus power box.

SharpCap V 4.0.9063.0 (64bit)
HP Pavilion 15 Laptop, Intel® Core™ i7-1165G7, 16 GB DDR4-3200 SDRAM
User avatar
admin
Site Admin
Posts: 13173
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: Player One Uranus-C and Slow Live Stacking

#2

Post by admin »

Hi,

I don't think it is likely to be a camera related issue, but I don't have a good explanation for it either. I would be interested in collecting more information to try to work out the reason so that I can fix whatever is going on.

One thing that would be of interest is to find out how long SharpCap was taking to stack each frame - this is shown in the 'Status' tab of the live stacking area, where it will show the stacking time and the render time - these are respectively the time taken to add the latest new frame to the stack and the time taken to redisplay the stack. If either of those are large (near to or more than the 4s exposure length) then that will help home in on the culprit. It's also interesting that the status bar at the bottom is updating but the live stacking area is not. I will try to work out how that might come about, since that might help work out what is going on.

cheers,

Robin
MarMax
Posts: 101
Joined: Sun Sep 19, 2021 11:43 pm

Re: Player One Uranus-C and Slow Live Stacking

#3

Post by MarMax »

OK Robin, I'll get out there tonight with the same rig and see if I get the delay. If so I'll check the Status tab, take a couple screen shots and report back.
User avatar
admin
Site Admin
Posts: 13173
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: Player One Uranus-C and Slow Live Stacking

#4

Post by admin »

Hi,

yes, any more feedback on this welcome. As I say, I don't fully understand what is going on, so I have done a little work to try to improve the efficiency of the code in live stacking to make it use less CPU time in the hope that will help. If you do get to the point where the live stacking is playing up again, it would be interesting to know what parts of the UI are 'live' and what parts are stuck - for instance the status bar at the bottom seems to be updating, but is the image changing and are the menus alive or stuck?

Also, if you get stuck and can get any response from the live stacking UI, it may be worth turning off things that make extra work to see if that helps. These would include

* saving the raw frames
* using the saturation adjustment
* using any of the noise reduction or sharpening enhancements
* having the 'optimize for faint stars' option turned on in star detection

If turning some or all of those off brings things back to life then we have learnt something :)

cheers,

Robin
MarMax
Posts: 101
Joined: Sun Sep 19, 2021 11:43 pm

Re: Player One Uranus-C and Slow Live Stacking

#5

Post by MarMax »

admin wrote: Sun Jun 26, 2022 12:50 pm Hi,

yes, any more feedback on this welcome. As I say, I don't fully understand what is going on, so I have done a little work to try to improve the efficiency of the code in live stacking to make it use less CPU time in the hope that will help. If you do get to the point where the live stacking is playing up again, it would be interesting to know what parts of the UI are 'live' and what parts are stuck - for instance the status bar at the bottom seems to be updating, but is the image changing and are the menus alive or stuck?

Also, if you get stuck and can get any response from the live stacking UI, it may be worth turning off things that make extra work to see if that helps. These would include

* saving the raw frames
* using the saturation adjustment
* using any of the noise reduction or sharpening enhancements
* having the 'optimize for faint stars' option turned on in star detection

If turning some or all of those off brings things back to life then we have learnt something :)

cheers,

Robin
Last night I did not experience any of the extreme delays with the live stacking. A very small amount of delay seems to be expected as there is a lot going on with the CPU. I took a screen shot anyway with the Status tab.

Image

The Stacking Time/Render Time was running pretty much like the above in the 1000-1300ms/180-220ms most of the time. There were some instances where the values were more like 10ms/200ms.

When it was acting up, the image was not changing and the menus were stuck until the backlog of frames was stacked. Regarding the things to turn off, I rarely save raw frames or touch the saturation adjustment, but I do use the noise and sharpening enhancements. With the 533 and 183 cameras I've been able to use Weiner Deconvolution without any problems. For some reason with the Uranus-C the two times I selected WD it crashed SharpCap.

Regarding the "optimize for faint stars" option, is this something I should leave turned on all the time?
User avatar
admin
Site Admin
Posts: 13173
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: Player One Uranus-C and Slow Live Stacking

#6

Post by admin »

Ok, interesting info, thanks.

The 585 has a lower pixel count than the 183 or the 533, so it is strange that it plays up when the others do not. I suppose it could be possible that the code that runs the PlayerOne camera uses more CPU than for other brands (I don't know this... just speculating) which could then impact on responsiveness.

There is a bug in the code for the faint star optimization that doubles the CPU workload for detecting stars - I have fixed that now and the change will be in today's update, but in currently released versions it will increase processing time.

Are you using the PC directly or are you using remote desktop (or some other remote access application) when the problems happen?

cheers,

Robin
MarMax
Posts: 101
Joined: Sun Sep 19, 2021 11:43 pm

Re: Player One Uranus-C and Slow Live Stacking

#7

Post by MarMax »

My laptop is running direct via USB to a hub at the mount.

Got it regarding the faint star optimization and I will also update SC today.

FYI, the last time using the Uranus-C, when it did not act up, I had faint star optimization selected. So who know's what's up with the prior night when I was getting the extreme stacking delays.

Another unusual behavior I experience is with the Brightness filter. Sometimes a new stack will either start below the filter level or drop below the filter level by a significant amount while stacking when conditions have not changed. What I do when this happens is to stop and restart live stacking and it seems to clear it up.
User avatar
admin
Site Admin
Posts: 13173
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: Player One Uranus-C and Slow Live Stacking

#8

Post by admin »

Hi,

Ok, thanks... more data - no pattern yet though :(

On the brightness side of things, the value being measured is basically the sum of the brightnesses of the stars being used for alignment. Looking at the code today (several years after I first wrote it) I wonder if it should really be the average (or even the median?). Anyway, changes in the stars being detected could lead to this sort of change in the brightness filter.

I will have to look into possibilities to stabilize it a bit.

cheers,

Robin
Post Reply