×

INDI Library v2.0.6 is Released (02 Feb 2024)

Bi-monthly release with minor bug fixes and improvements

Driver for the ZWO Optics ASI Cameras

  • Posts: 712
  • Thank you received: 174

I tried to build it on MacOS X but libraries (both 32 and 64 bit) are not compatible with the system :-(

ld: warning: ignoring file SDK/i386/libASICamera.a, file was built for archive which is not the architecture being linked (x86_64): SDK/i386/libASICamera.a
ld: warning: ignoring file SDK/amd64/libASICamera.a, file was built for archive which is not the architecture being linked (x86_64): SDK/amd64/libASICamera.a
10 years 2 weeks ago #786

Please Log in or Create an account to join the conversation.

  • Posts: 18
  • Thank you received: 3
Exactly, this is the reason for the trick of having multiple instances that identify them self by their executable name. A new API is worked on.
10 years 2 weeks ago #787

Please Log in or Create an account to join the conversation.

  • Posts: 18
  • Thank you received: 3
Hmmm I'll try to learn something about MacOSX ELF format.
10 years 2 weeks ago #788

Please Log in or Create an account to join the conversation.

I added a switch to select which camera to connect to. Please check if that works.
10 years 2 weeks ago #789

Please Log in or Create an account to join the conversation.

  • Posts: 18
  • Thank you received: 3
Fixed 2 bugs (see commit diff), updated doco and tested with both openphd and ekos running with 2 camera connected. I guess now it's all ready for release :-)
The following user(s) said Thank You: Jasem Mutlaq, Thomas Seckler
10 years 1 week ago #790

Please Log in or Create an account to join the conversation.

  • Posts: 54
  • Thank you received: 3
I tested the driver yesterday and it works quite ok. I only see problems (broken frames) with exposures of less than 1 second. Is there any way to work-around this?. In the past, I remember seeing in the SDK an option to play with the USB traffic. I don't know if it's still there. chripell, do you think this can improve the situation?. Or maybe discarding the first frame with such fast exposures?.
The following user(s) said Thank You: Christian Pellegrin
10 years 1 day ago #884

Please Log in or Create an account to join the conversation.

  • Posts: 18
  • Thank you received: 3
Thank you for testing! The parameter is USB bandwidth and now it's not exported, but it's easy to do. Now it's not touched so it's set to "auto". You see just the first frame broken, don't you? It doesn't happen to me but it really depends on the USB bus load. It's pretty easy to discard it anyway, the driver is already doing it when changing exposure because some frames with wrong one linger in a circular buffer.

BTW I'm working on a clean room reimplementation of the driver: github.com/chripell/yaaca/blob/master/asill.c I haven't forgot our MacOSX friends:-) It's in a very early stage, but can already captures frames without any closed-source code on the Linux machine. I studied a bit the chip and I managed to reduce the pixel clock to 8 or 2 Mhz, that means really low bus overhead on the USB bus (but you can have just longer exposure, which is OK for deep sky but not for planetary imaging). Anyone who wants to enjoy the hacking is welcome. :-)
10 years 1 day ago #887

Please Log in or Create an account to join the conversation.

  • Posts: 54
  • Thank you received: 3
Do you mean that it won't depend anymore on the binary blob?. That are really GREAT news !!!. The reduced framerate doesn't bother me very much. If I want to use INDI with this camera is because I would like to build an all-sky camera to make timelapses, capture meteors and so. Exposures of more than 1 second (30s-1min sounds reasonable) are mandatory I think.

Will your new driver support the AR0130CS sensor of the ASI120MC?. I've had a look in your code and I see mention to the MT9M034, but in some parts you make references to 'channels' (R, B, G1, G2) that confuse me :) .

Just one question more, if you let me. Can a sensor like this (with electronic shutter) make images with the 'shutter' closed (such as creating dark frames). If so, it would be really great for long exposures, wouldn't it? :) .

I don't have a wide enough hardware knowledge to help you with the coding, but I can test your driver with my camera (ASI120MC) whenever you want.
10 years 1 day ago #888

Please Log in or Create an account to join the conversation.

  • Posts: 18
  • Thank you received: 3
Yes, my goal is to not have any binary blobs anymore. I want to have finer control on what the sensor is doing too. For example now you have always a black level correction enabled which is cool for nice looking pictures but it makes the sensor useless for photometry because you cannot calibrate it. I want to make that optional. Another thing that I don't like is that you control just digital gain with the knob exposed by the SDK (just check the histogram of a daily picture, you will notice holes in the codespace which are the smoking gun of a digital multiplication going on): i want to tinker with analog gain and have digital gain set to unity.

Don't worry, from the data sheet the AR0130 an the MT9M034 are quite the same from the registers POV. I guess the first one just has the microfilters in front of the sensors. The possibility to change the gains of the single channels, the RGB mode for binning, etc. are in both of the chips.

I guess it could be possible to get a dark frame by programming the coarse and fine shutter registers in the right way but I haven't tried yet.

I will keep you informed about the progress!
10 years 1 day ago #889

Please Log in or Create an account to join the conversation.

  • Posts: 18
  • Thank you received: 3
I've added (and tested) USB bandwidth control in SVN. It's documented in README.md. Let me know if this helps.
9 years 11 months ago #908

Please Log in or Create an account to join the conversation.

  • Posts: 18
  • Thank you received: 3
I've updated to the latest version of the drivers from ZWO Optical i SNV. There is even a binary blob available for Mac, but I haven't any machine to test it. The open source low-level driver driver is working, if anyone wants to test please have a look here: www.evolware.org/?p=346 I'm considering rewriting the INDI driver so it can use that or the one from ZWO as an user choice.
PS from 24 april to 12 may I will have difficulties with connectivity because I'm on a trip, don't get mad at me if I don't respond to email in that period. :-)
9 years 11 months ago #995

Please Log in or Create an account to join the conversation.

Time to create page: 1.292 seconds