Adion
Halfway the entire screen would always be grey no matter the source, so that probably makes it a bit less interesting (and more strobe-like)
There's probably some things that could be done in different color representations, like shifting hue in HSV for example, but it would probably not feel as beat-related (probably similar to colorize now, but with more colors showing at the same time)
Inverting brightness could be an option, but the hue in dark regions is often noisy/unknown, so that probably ends up with a lot of colored noise.
I suppose either way shadertoy can be a good start to experiment with these.
There's probably some things that could be done in different color representations, like shifting hue in HSV for example, but it would probably not feel as beat-related (probably similar to colorize now, but with more colors showing at the same time)
Inverting brightness could be an option, but the hue in dark regions is often noisy/unknown, so that probably ends up with a lot of colored noise.
I suppose either way shadertoy can be a good start to experiment with these.
Posted Sun 09 Jun 24 @ 11:11 am
locoDog
I suppose it will, do shaders now interact with the image drawn behind them?
I don't know the right terms but can they take a live feed as an input, I don't think they did before.
oh and I have a thing that uses colorize in a beat kind of way, on wait off shift colour wait on
I don't know the right terms but can they take a live feed as an input, I don't think they did before.
oh and I have a thing that uses colorize in a beat kind of way, on wait off shift colour wait on
Posted Sun 09 Jun 24 @ 11:58 am
mwestling
I would love to see either one of these options listed below.
Option A: Add a 'Window Capture' option in the single source video option. I would like to use a 3rd party visualizer like Luminant vs. the shaders that are available when a song being played is audio only.
-OR-
Option B: The ability to send track info like Title, artist, etc. but ALSO include file type to a websocket server. Using Streamer.bot I have the ability to setup a websocket server and can change the source in OBS based on a script that checks the file type (mp3, mp4) or a boolean such as (has video = true) and I could switch the video source in OBS from the video output window of VDJ to the output window of Luminant or any other visualizer.
Option A: Add a 'Window Capture' option in the single source video option. I would like to use a 3rd party visualizer like Luminant vs. the shaders that are available when a song being played is audio only.
-OR-
Option B: The ability to send track info like Title, artist, etc. but ALSO include file type to a websocket server. Using Streamer.bot I have the ability to setup a websocket server and can change the source in OBS based on a script that checks the file type (mp3, mp4) or a boolean such as (has video = true) and I could switch the video source in OBS from the video output window of VDJ to the output window of Luminant or any other visualizer.
Posted Mon 15 Jul 24 @ 7:26 am
freppa
I could make a long list of what I miss on the Videoside of 64 version, Outline, VideoBars, Negative are a few that I used alot before :(
Posted Tue 16 Jul 24 @ 12:59 pm
freppa
And then it turns to silence, haha :)
Posted Sun 28 Jul 24 @ 4:44 pm
andytaylor125
This is a random idea I just had, if you are broadcasting or recording video with a webcam…
Currently you can switch cameras on a time basis. I’m not sure if it’s possible, but I imagine it would be simple enough to switch between video output and webcam based on time too, but that’s not the feature request…
What I’m imagining is using AI computer vision to auto switch cameras or perhaps activate a particular camera, if, and only if, I am looking directly at the camera and holding a microphone near my mouth.
Then, when I look away and move the mic away from my mouth, the camera deactivates or switches back to another view…A bit of a gimmick, but it would be kind of cool.
Maybe there could be physical gestures for other controls? When I put my hands up in the air, trigger the air horn sample, perhaps?!
Currently you can switch cameras on a time basis. I’m not sure if it’s possible, but I imagine it would be simple enough to switch between video output and webcam based on time too, but that’s not the feature request…
What I’m imagining is using AI computer vision to auto switch cameras or perhaps activate a particular camera, if, and only if, I am looking directly at the camera and holding a microphone near my mouth.
Then, when I look away and move the mic away from my mouth, the camera deactivates or switches back to another view…A bit of a gimmick, but it would be kind of cool.
Maybe there could be physical gestures for other controls? When I put my hands up in the air, trigger the air horn sample, perhaps?!
Posted Mon 29 Jul 24 @ 8:03 pm
andytaylor125
Something slightly more mundane, but probably useful than the last one...
I have started taking a webcam to my gigs so I can record the dance floor reaction to songs, posting clips to social media, etc.
I'd like the ability to split recordings (video in my case) by time. For example, every hour. Recording my 5 hour gig at 1080p using H265 codec, my whole system started to slow down after approx 2.5 hours and I got short gaps in the recorded audio. I stopped recording briefly, changed to H264 and all was fine for the rest of the gig.
For the first 2 hours, the H265 recording was perfect, so I suspect that if the recorded file was seamlessly split into 1-hour chunks, I wouldn't have had an issue.
I have started taking a webcam to my gigs so I can record the dance floor reaction to songs, posting clips to social media, etc.
I'd like the ability to split recordings (video in my case) by time. For example, every hour. Recording my 5 hour gig at 1080p using H265 codec, my whole system started to slow down after approx 2.5 hours and I got short gaps in the recorded audio. I stopped recording briefly, changed to H264 and all was fine for the rest of the gig.
For the first 2 hours, the H265 recording was perfect, so I suspect that if the recorded file was seamlessly split into 1-hour chunks, I wouldn't have had an issue.
Posted Tue 30 Jul 24 @ 1:16 pm