Sony EVI D30, D31 + VISCA over RS232 - anyone got a predone scene?
-
I've relinquished an old Polycom video conferencing box of it's internal EVI D30 camera & it's currently hooked up to my Macbook Pro via a Prolific USB to Serial converter running as COM3 using Repleo drivers, with it's SVHS feed coming in via BlackMagic Intensity Shuttle. All is working perfectly, and I'm using the "Send Raw Serial Data" actor to send hex codes to the camera and it's all behaving as it should. Am just wondering if anyone's already got a scene set up with all the commands in etc to save me having to build from scratch?
If not, I shall carry on and just build the whole thing...! *shrug*NB: it's MUCH easier getting it to work via Isadora than trying to get libVISCA to behave on OS X - original plan was to control via python-libVISCA from a web interface. Think now I'm just going to use OSC from my phone via Isadora instead. This appears to be the only satisfactory way to control an analogue PTZ camera from OS X, which I found vaguely surprising.
See further down the thread for final patch files. -
I had a patch that did this previously but it is on a drive that I can't access right now.... it was based on a patch I found on the old forum.
@Skulpture is the old forum still online, I can't find it? -
@marci , i think it was this patch but can't verify it without a visca cam … 215694-canon_vc-c4_vc-c50i.izz.zip
-
-
Ace - thanks muchly folks!!
-
Finally had chance to sit down today and look at both patches mentioned above... neither was a perfect fit, but @Michel - I loved your methodology for triggering the stops and allowing momentary buttons, so stole it. Anyways, managed to put together the below covering the majority of the D30 / D31 feature set - I've been picking these up dirt cheap off eBay recently, and via a cheap SVHS to USB stick with VideoGlide and a Black Magic Intensity and a few analog and digital matrix switchers I can now switch between 4 HDMI inputs and 4x SVHS Inputs, still with another 2 remaining for use within Isadora!
Fully tested with a combination of Sony EVI D30 (NTSC) and D31P cameras, 4 total, wired via VISCA Daisy-Chain to RS232 via Repleo cheap USB>RS232 adapter on OS X Yosemite with latest Isadora.All pan / tilt / zoom / focus / white balance / speed options are handled at the moment.In the future I _may_ add relative drive positioning for pan / tilt / zoom via a USB joystick x / y / z axis (or some variant thereof - have got a few Leo Bodnar boards kicking about and some spare sliders & pots).I **_WILL_ **be adding OSC to this and a Touch-OSC template in the very near future. (done)NB: Everything assumes serial port is 1\. Sorry. Will add a controller to the panel for this also. (done)New files added to post further down the page... -
-
As per the above, updated with control to choose Serial Port, OSC streams are now set up, and matching Touch-OSC template (for iPhone 5) 741814-evi-d30-d31-visca-over-rs232-control.izz aafebf-visca.touchosc.zip
-
I've been looking at this patch, and pondering...
Now that it's simple to get body part OSC values from the Kinect V2 into Isadora.... in theory, could I use your patch and take the Kinect OSC values rather than the TouchOSC ones for Pan and Tilt to use the PTZ camera to follow a performer around a stage? I'm sure some calibration would be needed, but it ought to be possible, don't you think? -
I like the way you think. What you're saying makes perfect sense to me; seems like you'd just lock it to the OSC values for the torso and then calibrate. Maybe try an easy small-scale test first, substituting in something for the Kinect component, like using Chroma Key and Eyes to track an colored, animated shape just moving back and forth to get the feel for it. After that maybe move to using the Kinect and the camera but still not full scale, as reducing the distance between your target and the camera will make your life easier. The only reason I say this is because personally, I tend to get too excited about trying new things, and try to dive into full-scale systems a bit too soon. Thus, I've adopted the practice of starting incredibly simple, breaking the task into components, getting each component working and making sure I understand them fully, before finally combining them for a full-scale test. It makes my trouble-shooting a lot easier, because it usually allows me to rule out various components, which I've independently verified as being functional, as the cause any problems I'm having.
Let me know how that pans out. I'd be very interested to see,
Woland
-
Sage advice, Lucas. And nice pun too. I'll start experimentations tomorrow, and keep you in the loop!