hhb
20th May 2006, 17:59
Hello, followed the recent thread here about peer to peer for NSV, which reminded me that some folks here understand computer video technology in great depth, which I do not. Am hoping some one can clue me about what's actually going on in the process described below:
Setup
Live, wireless dv transmission to remote sc server via NSVCap
Live audio transmission from mixer to remote sc server via NSVCap
Video
DVmini camcorder A/V out to 5.8 Ghz transmitter via RCA.
Receiver to converter box via RCA
Converter box to wireless laptop via firewire
Firewire stream to NSVCap via software bridge
VP6.1 encoding at 300 kbs/25 FPS
Audio
Audio stream via mixer to NSVCap
Raw AAC encoding at 32 kbs stereo
5.8 wireless dv transmission seems to work well -- inside (through walls, from 1st to 3rd floor, etc) and outside (okay at varying distances so far, transmits cleanly from outside to inside receiver/laptop).
Problem
When dv from remote cam gets to NSVCap, preview is fine, but online display (webpage player and winamp) shows the bottom two pixel lines incorrectly, sometimes black, sometimes slightly gray, sometimes just movement, which makes me think it's an interference problem.
But both 2.4 -- doesn't perform well otherwise either -- and 5.8 transmitting include the same, bottom 2-pixel line result, maybe suggesting it is an interlace prob.
Adding to my confusion about cause, FFDShow/Dscaler filters correct the display entirely, using either of two methods: TomsMoComp deinterlacer with vertical filter, or dscaler temporal noise filter with default setting -- without noticeable image quality loss.
Question
In the above described scenario, can someone tell me what is actually going on? Is the bottom, two pixel line problem simply interference in the wireless dv transmission, or is it an interlacing problem or what? Read (at http://100fps.com/) that both raw DV and analog (RCA to converter box output) are interlaced, tried both filters mentioned above with success, but mostly got confused about what the root of the problem actually is.
As far as I'm concerned, the fewer filters needed, the better, so am looking for an explanation of exactly what is causing the problem in the first place. Thanks for any advice and for any pointers to relevant info.
hhb
Setup
Live, wireless dv transmission to remote sc server via NSVCap
Live audio transmission from mixer to remote sc server via NSVCap
Video
DVmini camcorder A/V out to 5.8 Ghz transmitter via RCA.
Receiver to converter box via RCA
Converter box to wireless laptop via firewire
Firewire stream to NSVCap via software bridge
VP6.1 encoding at 300 kbs/25 FPS
Audio
Audio stream via mixer to NSVCap
Raw AAC encoding at 32 kbs stereo
5.8 wireless dv transmission seems to work well -- inside (through walls, from 1st to 3rd floor, etc) and outside (okay at varying distances so far, transmits cleanly from outside to inside receiver/laptop).
Problem
When dv from remote cam gets to NSVCap, preview is fine, but online display (webpage player and winamp) shows the bottom two pixel lines incorrectly, sometimes black, sometimes slightly gray, sometimes just movement, which makes me think it's an interference problem.
But both 2.4 -- doesn't perform well otherwise either -- and 5.8 transmitting include the same, bottom 2-pixel line result, maybe suggesting it is an interlace prob.
Adding to my confusion about cause, FFDShow/Dscaler filters correct the display entirely, using either of two methods: TomsMoComp deinterlacer with vertical filter, or dscaler temporal noise filter with default setting -- without noticeable image quality loss.
Question
In the above described scenario, can someone tell me what is actually going on? Is the bottom, two pixel line problem simply interference in the wireless dv transmission, or is it an interlacing problem or what? Read (at http://100fps.com/) that both raw DV and analog (RCA to converter box output) are interlaced, tried both filters mentioned above with success, but mostly got confused about what the root of the problem actually is.
As far as I'm concerned, the fewer filters needed, the better, so am looking for an explanation of exactly what is causing the problem in the first place. Thanks for any advice and for any pointers to relevant info.
hhb