barron
01 Mar 2009, 04:21
My primary use of this software would be for live mixing computer based presentation visuals, HD or SD ENG video of said presenter(s), and perhaps to play back an intro video or two. When capturing any text based apps, being able to record at or near native resolution of the display is vital to maintain the detail and readability. (we framegrab through epiphan vga2usb LR's)
The way I would use this would be to define a 1280X720 program size. I would capture the presentation computer at 1024X768 from my vga2usb. From there I would scale it back to 960X720 and push that video to a side of the screen. Then I could do a PiP in an opposite top corner (from cam) to minimize overlap between images. My final goal would be to do text overlays in the blank space beneath the PiP.
I know that this takes a ton of CPU power, but using "Studio" version 0.42 trial at highest supported resolution, I've been able to use this setup of inputs and encode at 1280x960 (upscaling) into Expression Encoder 2 (advanced VC1 Profile 5000kbps for archive) while hitting an average 70% load across my Q6600 @ 3.2ghz. I know this will greatly increase with higher resolution inside vidblaster, but thats part of the reason why I'm putting together my first I7 rig...
If you follow this link there is a streaming video that around the 3:00 mark demonstrates the whole idea as done in post. Just before that there are some examples of internet browsers etc. You will see they are unreadable at that resolution. If you take the video fullscreen (still fighting the player to maintain aspect ratio), it will become readable as the actual res of that video is 1280X720. Yes that is large, and yes the video clocks in at a "whopping" 10.5MB for 5 minutes of video thanks to low movement and H.264!
http://www.recordwithbarron.com/businesses/presentationrecording/demo
So, to sum up my requests...
1)Resolution up to 1280X720 (includes resolution reported as a virtual device)
2)Ability to resize and position video streams
3)Text overlays with positioning within frame
4)Opacity and position of graphics overlays (sponsor watermarks etc.)
These features would make this a dream product as it would save me hours of transfering, editing, and rendering by working in real time.
Keep up the incredible work!
The way I would use this would be to define a 1280X720 program size. I would capture the presentation computer at 1024X768 from my vga2usb. From there I would scale it back to 960X720 and push that video to a side of the screen. Then I could do a PiP in an opposite top corner (from cam) to minimize overlap between images. My final goal would be to do text overlays in the blank space beneath the PiP.
I know that this takes a ton of CPU power, but using "Studio" version 0.42 trial at highest supported resolution, I've been able to use this setup of inputs and encode at 1280x960 (upscaling) into Expression Encoder 2 (advanced VC1 Profile 5000kbps for archive) while hitting an average 70% load across my Q6600 @ 3.2ghz. I know this will greatly increase with higher resolution inside vidblaster, but thats part of the reason why I'm putting together my first I7 rig...
If you follow this link there is a streaming video that around the 3:00 mark demonstrates the whole idea as done in post. Just before that there are some examples of internet browsers etc. You will see they are unreadable at that resolution. If you take the video fullscreen (still fighting the player to maintain aspect ratio), it will become readable as the actual res of that video is 1280X720. Yes that is large, and yes the video clocks in at a "whopping" 10.5MB for 5 minutes of video thanks to low movement and H.264!
http://www.recordwithbarron.com/businesses/presentationrecording/demo
So, to sum up my requests...
1)Resolution up to 1280X720 (includes resolution reported as a virtual device)
2)Ability to resize and position video streams
3)Text overlays with positioning within frame
4)Opacity and position of graphics overlays (sponsor watermarks etc.)
These features would make this a dream product as it would save me hours of transfering, editing, and rendering by working in real time.
Keep up the incredible work!