- 参加（入社、統合）: 1 year ago
- 最終ログイン: 7日前
ストリーム - 最近のアクティビティ
Topaz Video Ai v3.1.6, Ai model Gaia, input - HQ, H265 Nvidia, bitrate auto, resolution 2x upscale. I use avidemux before to cut out scenes that i don't like than select as output "Copy" on video and audio to not re-encode and save file. I use ffmpeg to split video in equal parts - like every part to be 3 or 4 or 5 minutes (of course the last part will be shorter than that) because upscaling VR takes a lot of time so I don't want my PC to be stuck for 24H rendering. You can use queue to add several parts if
if you know you will not be using PC for many hours. Also i recommend cutting 10 seconds of video to upscale that first to see if you like the result or what ai model to choose. After upscaling I also use avidemux to join all parts and save - Copy - video without re-encoding. If bitrate is to big than you can re-encode in avidemux, handbrake, or ffmpeg.
Topaz Video Ai v3.1.6, Ai model Gaia, input - HQ, H265 Nvidia, bitrate auto, resolution 2x upscale. I use avidemux before to cut out scenes that i don't like than select as output "Copy" on video and audio to not re-encode and save file. I use ffmpeg to split video in equal parts - like every part to be 3 or 4 or 5 minutes (of course the last part will be shorter than that) because upscaling VR takes a lot of time so I don't want my PC to be stuck for 24H rendering. You can use queue to add several parts if
if you know you will not be using PC for many hours. Also i recommend cutting 10 seconds of video to upscale that first to see if you like the result or what ai model to choose. After upscaling I also use avidemux to join all parts and save - Copy - video without re-encoding. If bitrate is to big than you can re-encode in avidemux, handbrake, or ffmpeg.
Topaz Video Ai v3.1.6, Ai model Gaia, input - HQ, H265 Nvidia, bitrate auto, resolution 2x upscale. I use avidemux before to cut out scenes that i don't like than select as output "Copy" on video and audio to not re-encode and save file. I use ffmpeg to split video in equal parts - like every part to be 3 or 4 or 5 minutes (of course the last part will be shorter than that) because upscaling VR takes a lot of time so I don't want my PC to be stuck for 24H rendering. You can use queue to add several parts if
if you know you will not be using PC for many hours. Also i recommend cutting 10 seconds of video to upscale that first to see if you like the result or what ai model to choose. After upscaling I also use avidemux to join all parts and save - Copy - video without re-encoding. If bitrate is to big than you can re-encode in avidemux, handbrake, or ffmpeg.
