Avisynth editing

[jwfc]

I'm new to Avisynth and am doing some simple editing. I had a couple questions.

(1) When I am doing my editing of joining and cropping videos, I want to resave the output. Do I open the script.avs in Virtualdub and just capture the AVI again? If something is already compressed, is there any default setting to keep the compression the same. I am guessing I will have some loss to my video since I am editing and resaving again, but I was trying to find the best way to save the new output with less loss and keeping the file roughly the same size. Or do I just configure the compression type according to the size I want.

(2) When I capture clips with my PVR, I get the same black bars around the edges of my video. I want to create an AVIsynth script to crop out the black bars and resize my video. I also use the same AS@L5 two-pass Xvid compression and was wondering if there was any way to automate that so I don't have to always save some file as the first pass, and then save another file as the second pass to get my output. Just trying to reduce the redundancy of my clips . Thanks!

(3) What is the difference between LancosResize and Lancos4Resize?

[jagabo]

(1) When I am doing my editing of joining and cropping videos, I want to resave the output. Do I open the script.avs in Virtualdub and just capture the AVI again?

Capture isn't the right word but yes, you open the AVS script in VirtualDub and AviSynth passes frames one by one to VirtualDub.

If something is already compressed, is there any default setting to keep the compression the same.

No. AviSynth decompresses each frame, does its processing, and then passes uncompressed frames to VirtualDub. You have to select what compression codec you want for your output within VirtualDub.

I am guessing I will have some loss to my video since I am editing and resaving again,

Yes, if you are using a lossy compression codec in your output video.

but I was trying to find the best way to save the new output with less loss and keeping the file roughly the same size. Or do I just configure the compression type according to the size I want.

Quality vs file size varies depending on the codec used. For full D1 video (720x480, 29.97 fps) the size can range from roughly 75 to 110 GB/hr for uncompressed video, 20 to 50 GB/hr for losslessly compressed video, 2 to 4 GB/hr for MPEG2, to 1 to 2 GB/hr for MPEG4 part 2 (Divx/Xvid). The MPEG2 and MPEG4 codecs usually have contant quality modes where you select the quality you want and the file comes out to whatever size is necessary to deliver that quality. This usually works pretty well when recompressing. Otherwise you can specify a desired bitrate (2-pass encoding works best) but it's difficult to predict exactly what quality you will get for a particular bitrate (and hence file size). If the output will be using the same codec as the source, the same bitrate (or slightly higher) will deliver about the same quality (depending on what filtering you are doing).

(2) When I capture clips with my PVR, I get the same black bars around the edges of my video. I want to create an AVIsynth script to crop out the black bars and resize my video. I also use the same AS@L5 two-pass Xvid compression and was wondering if there was any way to automate that so I don't have to always save some file as the first pass, and then save another file as the second pass to get my output. Just trying to reduce the redundancy of my clips :). Thanks!

Try using AVSP to create your AVS script. You can easily edit the script and see what the result will be right within the program. For cropping in particular, I often open the source file in VirtualDub and use it's cropping feature to determine the right values to use in AviSynth.

With multipass VBR encoding you have to perform at least 2 passes. During the first pass the encoder is only looking at the video and saving information about each frame. During the second pass it uses that information to allocates bitrate to each frame to get the best quality for the requested average bitrate.

(3) What is the difference between LancosResize and Lancos4Resize?

In short, Lanczos4Resize is a little sharper. Sharper isn't always better though.