I have in different patches at different times experienced that Keyboard (System Global) only work around half of the times I open my patch that has the node in it.
Same behavior when making a new node in a fresh patch it seems that only every other time it is working.
I have made sure that I only have one instance of the keyboard node even though it is no longer a singleton as mentioned in an earlier thread.
I am on win7 pro 64bit. I mostly run several instances of vvvv and I am suspecting that therein might lie some of the reason.
i experienced often some “ALT” or “3” remains in the outputbuffer; so an attached KeyMatch doesn’t work anymore…
yep kalle, hitting ALT again usually clears the hickup…
In my case it is not a question of alt+enter being hung, the output has either a stuck end nothing after that is coming out or nothing at all.
Just working a bit around with making an example patch I have found a consistent behavior:
start a patch with a Keyboard (System Global) node, it works fine. (apart from the previously mentioned ALT 3 kalle mentioned earlier in this thread)
start a blank patch in a new instance of vvvv and the Keyboard node in the first instance stops working. If there is a Keyboard node in the second instance it works fine… until one more instance of vvvv is started.
To me this doesn’t seem like the intended way for the Keyboard node to work.
I made a quick example patch to illustrate.
Keyboard_behavior.v4p (2.4 kB)
can confirm sunes troubles with two instances. same problem with Mouse (System Global). seems for now the only way around is sending the data via udp/sharedmem from the first to the second instance.
yes, that is possible, but the boot sequence is not always predictable and thus starting the keyboard listening patch is not always the last thing to happen.
But I will probably just have make completely sure everything else is started and then start the keyboard listening patch.