I am the Senior AV Broadcast Technology Analyst at Loyalist College where I primarily design, install and maintain campus AV, Digital Signage and Broadcast systems and equipment. I have also do some teaching for Camera, VFX and Audio/Video editing as well; I love my job. My background includes Broadcast Television, Video Production and Technical work. My passions include traditional and digital drawing, fictional creative writing, VFX, and cinematic video projects. I have also recently rekindled my interest in robotics and have tinkered with Arduino and Raspberry Pi and a few programable robots.
I find I most comfortably find myself in the chaotic limbo between the artistically and technically creative, believing one can indeed excel at both. Which is probably why I am loving VUO! :)
The Logitech C920 only provides 30fps at 1080p with applications that supports H.264 directly, and can pull in the H.264 stream directly from the camera. The C920 does on-board H.264 compression, but most applications don't support pulling the compressed stream straight from the camera; instead, they have to decompress then re-compress the stream, dropping the framerate.
Does that sound right? Does VUO support H.264 camera streams?
Partially occlude datamatrix (or sometimes just leave it as it is)
Output from datamatrix node sometimes jumps back and forth between numbers and proper reading of text.
We had a few instances of the Barcode node incorrectly reading a Datamatrix and ended up displaying numbers momentarily rather than the text that it should have read. Sometimes it was briefly, sometimes it flipped back and forth between numbers and a proper reading and other times numbers were visible for a few seconds before it was read correctly and displayed text. We also had a couple of instances where it displayed numbers when it decided a wood grain was a barcode (The card holder we have the camera mounted to was made of wood). This got me thinking that because the barcode reader can read different types of barcodes perhaps before it read the datamatrix as a datamatrix that it may have incorrectly identified it as a different type of barcode and spit out the results before correctly detecting it as a datamatrix. I think a simple fix might be to add a drop down list of the type of barcode we would like the node to look for. This could prevent incorrect identifications.
I managed to snap a few pictures of two incorrect readings of the same datamatrix (one with a partially occluded datamatrix) along with the correct reading. Because it contains student information I can't post it here but could privately send them to you.
That looks like what I am talking about. Thanks Bodysoulspirit I will look into Jaymie's work around, right now I'm will to try anything! After the developments team's amazing fix for the AMD issue the released today this is one of the last ducks I need to line up. Thanks for finding this. :)