SharpCap Forums

Ok,

thanks, I will put that one on the 'to fix' list.

cheers,

Robin

Thanks Robin. It's not a high priority, since 12-bit packed works fine.
Regards Alan

Hi Alan,

I hope it is as simple as byte swapping the data - ie a quick fix. If not I might be tempted not to show the 12 bit (non packed) option if 12 bit (packed) is available - i.e. make it nearly impossible to get to the bug situation.

cheers,

Robin

Good idea. I'm not even sure what "packed" means anyway

Hi,

the 'packed' option means that the 12 bit data comes from the camera to the PC as two pixels packed into 3 bytes. Non-packed means that each pixel takes 2 bytes, so in theory the packed option gets more FPS out of the network bandwidth at the expense of a bit of extra CPU load to do the unpacking.

Anyhow, I found the bug, so Monday's update will have a fix in.

cheers,

Robin

That's great news Robin. Bravo.

Regards Alan

I tried a quick comparison of the BASLER IMX249 with the similar FLIR IMX249 camera for 5 minute exposures of a 50 micron grid in a dark room.
1) The BASLER viewer limits exposures to 10s, but SharpCap allows up to 5 minute quick picks.
2) the FLIR viewer limits exposures to 30s, which is also the SharpCap max. quick pick.
With SharpCap, longer exposures of up to 600s appeared to work, but with interesting differences
1) The BASLER camera did appear to integrate for the full 300s or even 600s exposure (the progress bar worked)
2) The FLIR camera did not integrate, but perhaps averaged multiple 30s exposures (the progress bar did not work)

Surprisingly, noise still seemed acceptable (for my purposes) with 300s exposures, and even amplifier glow was not too bad, especially for BASLER, though that camera seemed to run hotter, with no temperature reading. The Flir camera read 43C, and both cameras had been allowed to warm up overnight and set for 12-bit. Files are on:
https://neutronoptics.com/downloads/Bas ... d-30s.fits
https://neutronoptics.com/downloads/Bas ... -300s.fits
https://neutronoptics.com/downloads/Bas ... -600s.fits
https://neutronoptics.com/downloads/Fli ... 6-30s.fits
https://neutronoptics.com/downloads/Fli ... -300s.fits

The BASLER GigE camera can also bin in firmware 2x2 and 4x4. The FLIR GigE camera can bin 2x2.
A camera using the IMX432 (9 micron pixels on a 1.1" chip) would be interesting, but costs twice as much from FLIR and is not available (?) from BASLER.

Hi,

interesting stuff. SharpCap does some fancy footwork with Basler cameras to use their trigger functionality to bypass the normal exposure length limits. There's nothing clever like that in the FLIR code - it just uses the FLIR SDK, so no surprise that it mirrors the limits of the other software.

BTW, I put out a SharpCap update today that should fix the 12 bit unpacked issue.

cheers,

Robin

Thanks Robin. It's great that SharpCap allows long exposures for Basler cameras. I'm still trying to work out if I can use such cameras for long exposures, instead of cooled CCDs. I had assumed that CMOS was much inferior, but am not so sure now about the "much", thanks largely to SharpCap. Firmware binning is also important when the commercial pressure is for small pixels, as is isolated hot pixel filtering in ImageJ. It would be nice if the cooled cameras weren't so big. I fondly remember a cooled version of the old Watec WAT-120N camera that was still only about an inch cubed

Cheers, Alan

BTW I summarised my Basler and Flir software experience on
http://neutronoptics.com/basler.html
http://neutronoptics.com/flir.html