I have bought it in December 2007. I returned it to Imaging Source because the exposures were limited to 30 s. They arranged the camera (firmware?) and since the I was able to take longer exposures, always using IC Capture running on Windows.

OK! I will get the cam in the observatory and check the firmware and available controls. I'll report back.

I was a little bit wrong: It's a DMK41AU02! But the difference is only the size of the sensor. I think the internals are the same.
Output of "v4l2-ctl -l -d /dev/video0"
                brightness 0x00980900 (int)    : min=0 max=63 step=1 default=0 value=0
                gamma 0x00980910 (int)    : min=1 max=500 step=1 default=100 value=100
                gain 0x00980913 (int)    : min=260 max=1023 step=1 default=260 value=260
                exposure_auto 0x009a0901 (menu)   : min=0 max=3 default=3 value=3
                exposure_absolute 0x009a0902 (int)    : min=1 max=36000000 step=1 default=333 value=333 flags=inactive

You are missing the last two options. I'm pretty sure, it's a matter of the firmware, as I stated before. Alas "Imaging Source" is very restrictive in this context, because the update can damage the cam (see solarchatforum.com/viewtopic.php?t=2735). So you have to ask for it explicitly!

You can get more info about the cam with "lsusb -v -d <vendor>:<product>" (e.g. "lsusb -v -d 199e:8101" for my DMK41AU02). The firmware I got from "Imaging Source" still does not comply 100% with v4l2, but at least I can operate now the exposure time.

I remember exchanges about this situation on the Imaging Source forum. That is certainly the right way to solve this problem, thanks for posting the info.

-Eric

Many thanks for your posts. I'm going to contact TheImagingSource to explain the problem, with the hope that a single firmware upgrade will solve the problem. Of course I'll keep you posted.

.

I received a response from S. Geissler at TheImagingSource.
He instructed me to build tiscamera from sources, and then to
    $ sudo cp build/data/udev/80-theimagingsource-cameras.rules /etc/udev/rules.d/
    $ sudo service udev restart
Unluckily, this didn't enable me to take exposures.
Mr Geissler warned me that this woudn't help if my camera was already recognized by the system, and it did, but as a DFx 21BU04 instead of a DMK 31AU03.AS, and, indeed, the suggested procedure didn't change anything.
So, I think this is the end of the story. I'll continue to take videos of the Sun and Moon and, maybe, revive a Windows laptop to take exposures.
Regarding the firmware upgrade, the 41 and the 31 cameras differ by more than just the CCD size: the 41 is "more v4l2-compliant" than the 31.
Thanks a lot to both of you for having tried to help me.

Perhaps we could help a bit further. Your camera firmware was not updated then? You built the tiscamera source package? What was the result, do you have any logs from "dmesg" or "v4l2-ctl"? And what's in the rules files?

-Eric

No firmware update, indeed.
I attach the rules file created during compile.
v4l2-ctl --all gives what seems to be information about the laptop integrated webcam
And the last lines output by dmesg are:
[ 1391.394958] usb 2-1.6: new high-speed USB device number 4 using ehci-pci
[ 1391.505061] usb 2-1.6: New USB device found, idVendor=199e, idProduct=8101, bcdDevice= 1.00
[ 1391.505066] usb 2-1.6: New USB device strings: Mfr=1, Product=2, SerialNumber=3
[ 1391.505070] usb 2-1.6: Product: DMx 31AU03.AS
[ 1391.505073] usb 2-1.6: Manufacturer: The Imaging Source Europe GmbH
[ 1391.505076] usb 2-1.6: SerialNumber: 50700328
[ 1392.639425] uvcvideo: Found UVC 1.10 device DMx 31AU03.AS (199e:8101)
[ 1392.640754] uvcvideo 2-1.6:1.0: Entity type for entity Input 4 was not initialized!
[ 1449.126096] perf: interrupt took too long (4010 > 4000), lowering kernel.perf_event_max_sample_rate to 49750

 

I found this documentation about firmware updates: www.theimagingsource.com/documentation/t...#locate-the-firmware

Do you have a firmware where the page says you might?

Also, you might want to give v4l2-ctl the device name explicitly, check earlier in the thread, there is an example.

-Eric

It's a very interesting page you found.
Unluckily, there is no firmware for my DMK31AU03 (only for DFK and DMK of types 22uc03, 42uc03 and 72uc02).
As Mr Geissler wrote, my 15 years old camera was designed in a time when no one suspected it would one day be used with Linux.

So probably your only solution is to use their SDK to rewrite an indi driver. Well, that's another story. Sorry!

-Eric