ZWO ASI294MM Pro SDK Update

User avatar
admin
Site Admin
Posts: 13122
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: ZWO ASI294MM Pro SDK Update

#11

Post by admin »

Hi,

thanks for the analysis reports – it looks like the camera isn't switching into HGC mode when run at full resolution (there's no clip in the read noise like there is at the old full resolution). The read noise is – give or take – about half the value at low resolution, which would be expected.

On the crash when you are trying to use the focus tools, I think I found the uploaded crash report and it looks like you are using a Celeron J4125 CPU. Unfortunately CPUs of this age don't work with the 64-bit version of SharpCap – SharpCap 64 bit requires a CPU that supports the AVX/AVX2 instruction set. In theory those instruction should only be used if they are available, but with the 64 bit version it tries to use them for some memory housekeeping tasks and try as I might I have been unable to find a way to stop it :(

Unfortunately, that leaves you stuck to some extent, since the 32-bit version will work fine but may hit memory problems with such a high resolution camera. The 64-bit version will work for basic capture, but anything that relies on image processing (like focusing or live stacking) will fail.

Sorry about that, Robin
umasscrew39
Posts: 408
Joined: Fri Apr 28, 2017 1:28 pm
Location: Central Florida
Contact:

Re: ZWO ASI294MM Pro SDK Update

#12

Post by umasscrew39 »

According to ZWO, the new bin 1 is operating in HGC so there is no noticeable drop-off in read noise like with the old bin 1 setting. They recommend using 120 gain and 30 offset in either setting.

I just downloaded the new SDK so hope to try it soon and compare it to the old bin 1 that I have been using.

Bruce
psparkman
Posts: 10
Joined: Wed Nov 25, 2020 9:04 pm

Re: ZWO ASI294MM Pro SDK Update

#13

Post by psparkman »

Thanks for all of the help Robin! I wondered if the error that I get was due to the Celeron processor as the 64bit version does give a warning when I start up. So...yes, I guess that I am stuck until I upgrade that PC. Not really an issue as the main thing that I use SharpCap is to do Polar Alignment. That is a great feature BTW!

I did some testing last night imaging the Heart Nebula with Bin 1, and Bin 2. I used 20 minute exposures with Bin 1 and Gain 100 to get the well depth to near 4000 to match the 12bit output of the camera. For the Bin 2 images, I used my normal 10 minute exposures (these are all narrow band) and gain 120. The reason for the 20 minute exposures on Bin 1 is because the file sizes are so large that I just don't want to deal with that much data. I was only able to get 4 Bin 1, and 8 Bin 2 exposures before the wind picked up. But I integrated those in Pixinsight and compared them. The analysis shows the SN ratio better for the Bin 2 images as expected. However, visually looking at them the Bin 1 images are much better. The noise is much finer and easier to reduce, and the detail is better. I think that will be my settings. Bin 1, Gain 100, offset 20, and 20 minute exposures with narrowband. Will probably use 0 gain for LRGB too keep the number of exposures down.
psparkman
Posts: 10
Joined: Wed Nov 25, 2020 9:04 pm

Re: ZWO ASI294MM Pro SDK Update

#14

Post by psparkman »

Since I need to use the 32 Bit version for Polar Alignment, there is one bug. If I try to do polar alignment after connecting then disconnecting from the ZWO ASCOM driver via Sequence Generator Pro, I can not get a alignment solve. It appears that if you use the ASCOM connection then go to the SharpCap camera driver the image has the wrong resolution on one of the axis, like the image is squashed. If I look at a terrestrial object, you can see that the image looks stretched side to side. If I disconnect the USB and power from the camera after using ASCOM, then SharpCap works correctly.
User avatar
admin
Site Admin
Posts: 13122
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: ZWO ASI294MM Pro SDK Update

#15

Post by admin »

Hi,

Is that with the 294 camera? I just tried this with a ZWO 1600MM and didn't see the problem - of course it may be either down to the camera model or the exact versions of SharpCap/SGP/ZWO Ascom driver.

cheers,

Robin
psparkman
Posts: 10
Joined: Wed Nov 25, 2020 9:04 pm

Re: ZWO ASI294MM Pro SDK Update

#16

Post by psparkman »

This is with the 294MM. The 1600MM always worked fine. I am not the only person having this problem with the 294MM, as others have commented on the ZWO support site. It started after updating the ZWO drivers.
User avatar
admin
Site Admin
Posts: 13122
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: ZWO ASI294MM Pro SDK Update

#17

Post by admin »

Hi,

Ah, okay, thanks for clarifying. If it's just happening with the 294 camera and not with others then it sounds likely to be a bug in the ZWO SDK - hopefully once they have a fix in place then that will automatically fix the problem in SharpCap when I include the new SDK. It's probably not something that I can fix in SharpCap itself – especially not without a camera that shows the problem to test on.

Cheers, Robin
psparkman
Posts: 10
Joined: Wed Nov 25, 2020 9:04 pm

Re: ZWO ASI294MM Pro SDK Update

#18

Post by psparkman »

Actually, I had not downloaded the latest version of the 32Bit application. I installed that, and it works fine. Thanks for all of your support!
User avatar
admin
Site Admin
Posts: 13122
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: ZWO ASI294MM Pro SDK Update

#19

Post by admin »

Hi,

Glad to hear that the update resolved it – thanks for reporting back.

Cheers, Robin
m32guy
Posts: 25
Joined: Sat Sep 21, 2019 9:11 pm

Re: ZWO ASI294MM Pro SDK Update

#20

Post by m32guy »

Quick question here regarding the new binning options. Like umascrew I I prefer running the 294mm in it's original configuration (now 2X bin). I love using the Brain but as was previously mentioned you can't run the sensor analysis at 2X bin. Given that 1X bin behaves differently, is there any workaround to still use the Brain?

Thanks

Steve
Post Reply