Imaging Source USB camera vendor/product IDs

[JamesF]

I'm idly collecting information on the IS cameras so I can support them with my Linux capture application, but information on how to get the cameras to work at all on Linux appears to be sketchy at best.  Allegedly they're not-quite-compliant UVC cameras that may already have a driver "tweak" in the Linux UVC driver to cope, but that only works for cameras with a USB VID/PID of 199e:8102 and I'm far from convinced that's the only VID/PID used for the IS CCD cameras.

If you have one of these cameras and are able to get Windows to cough up the USB VID/PID I'd appreciated it if you could post the information (together with the camera model here.

Thanks,

James

[allcart]

And just how do we do that James. I'd like to help if only I knew how.

[JamesF]

And just how do we do that James. I'd like to help if only I knew how.

If I were a Windows person, I wouldn't be writing a Linux capture application

I'll see if I can work it out.  It's probably somewhere in the Device Mangler.

James

[JamesF]

Aha!  In Windows 7...

I bring up the control panel and click on the "Hardware & Sound" header, then select "Device Manager" from the "Devices and Printers" section.  In the Device Manager, find the device and double-click on it to get the properties window.  Select the "Details" tab and in the "Property" drop-down, choose "Hardware Ids".

In the "Value" section it should probably then display something like

USB\VID_199E&PID_8102...

which is the information I'm after.

James

[JamesF]

Oh, I've also started looking into what is required to support the TIS cameras.  Looks like it could be something of an uphill battle as the company doesn't really seem that interested in anything other than Windows, but there you go.  It's got to be desirable to at least try to support some of the most popular cameras out there...

James

[allcart]

Ok James, this is the DMK31 AU03.AS

hardware ids gives 2 lines as follows.

USB\VID_199E&PID_8101&REV_0100&MI_00

USB\VID_199E&PID_8101&MI_00
 Hope that is of some help.  If you want the DMK41 as well, I'll give you it tomorrow.
 

[JamesF]

Thank you, Allan.  That's consistent with what I have already but not what appears in the source for the Linux UVC driver (which is PID 8102), so knowing the values for the DMK41 would definitely be another useful datapoint.  I guess it's possible that 8101 is for the astro cameras and 8102 are their "machine vision" cameras or something like that.

James

[allcart]

Exactly the same numbers James

[JamesF]

Exactly the same numbers James

Thanks Allan.

James

[JamesF]

If anyone can do the same for one of the DxK21AU04 (ie not the 618) variants that would be helpful.

James

[U Cyg]

Hi James, these showed for both the DBK21AU04 and DMK21AU04:

USB\VID_13D3&PID_5710&REV_1130&MI_00

USB\VID_13D3&PID_5710&MI_00

Angie

[JamesF]

Hi James, these showed for both the DBK21AU04 and DMK21AU04:

USB\VID_13D3&PID_5710&REV_1130&MI_00

USB\VID_13D3&PID_5710&MI_00

Angie

Hmm.  That would be an odd one.  According to the USB vendor ID lists that is indeed a UVC webcam, but made by AzureWave Technologies rather than TIS.

Was this information from a laptop that might also have a built-in webcam?

James

[U Cyg]

Hi James, it is just as well that you know what you are doing!  You are correct in thinking that the information

was from a laptop with webcam, and I foolishly clicked on the wrong camera! 

Here is the correct information, same for both cameras:

USB\VID_199E&PID_8101&REV_0100&MI_00
USB\VID_199E&PID_8101&MI_00     

It looks to be the same as that posted by Allan.

Sorry for any confusion I may have caused, I blame the cold, damp night air and lack of sleep!

Angie 
 

[JamesF]

No problem, Angie  Thank you for providing the information.

It would appear then that most, if not all, of the TIS astronomy cameras are 199e:8101 then, whilst the Linux driver is for devices with the 199e:8102 VID/PID.  TIS have suggested to me that the DMK51 is 199e:8102, but not in a manner I'd describe as particularly convincing.  I'm at a bit of a loss as to what's going on.

James

[JamesF]

Well, I've spent a bit of time firing emails around and it appears that Linux support for the TIS cameras is going to be "hard".

The UVC driver webpage suggests that USB cameras with the 8101 product id (all the "usual" astro cameras by the looks of it) are supported, but that's just not correct.  None of them are, because the driver doesn't recognise their VID/PID at all.

There is a suggestion that some of the cameras (but not the 21.618) will "mostly" provide a UVC interface if it has the firmware changed to a somewhat older version.  But I don't know if that also changes the PID to the one that the UVC driver recognises and a solution that doesn't work with the 618 is less than ideal.  So much for a company that advertises its products as "Technology based on standards".  Hey ho.

I think this one is going to have to go on the backburner for a while

James

[allcart]

Thats a shame James. I don't use Linux and never have, but it was in interesting project to watch. It does seem strange to me that these machine cams can't be used with a simple Linux program. What software would a factory computer operated machine use with these cams.??

[JamesF]

I guess when they're used in commercial installations they're controlled by some sort of embedded Windows environment and use the Windows driver supplied by TIS.  If you want to use it anywhere else then you're out of luck as far as I can see.

It's not an insurmountable problem if I care to put the work in.  Reverse engineering the interface isn't beyond the realms of possibility.  Life would be much easier if it weren't necessary though.

James

[JamesF]

Well, it gets ever more interesting.

I just received an email from someone who basically said "All that stuff TIS support have told you is rubbish.  It won't work out of the box, but some work has been done recently in the kernel to sort things out and what you need to do to get these cameras to work is this..."

And I have to admit that it's looking rather promising...

James