How to plan post-capture process of my workflow?
 

tl;dr I need some help pointing me in the right direction in my committed journey of VHS Capture. I don’t know how to go about learning the software (besides VirtualDub and Adobe Premiere Pro) or what I should even focus on, settings, restoration, upscaling etc.

I’m stuck on one of the rungs of the Ladder of VHS Capture Understanding and need some help because I don’t know what to plan for next. If this was a video game, I think I’m reached the next level though: Level Two. Scouring the forums is great and always a last resort if anything, but I’m not shy and will ask if I need to for a more specific and immediate result. I appreciate all the forums memebers on here. This place is awesome btw.

Anyways, I don’t have most of my equipment yet, but it’s in the works, so I’ve been trying to plan and prepare for the entire workflow from beginning to end, to including educating myself. As of right now I’ve actually built my XP Capture Machine (WINDOWS XP PROFESSIONAL SP3 X86 - INTEGRAL EDITION) and installed the LS VirtualDub on it. I’m down to the next software portion now, to include settings/configuration and this is my crossroad and I’m not sure where to go from here.

The first part/step one is hardware related. This is the basic gist of my understanding:
UPS → VCR -> TBC -> Capture Card → Windows XP machine → VirtualDub + HuffYUV

What does the Software form of this look like during and after capture? Or how would you describe the process after this? Does it go to a program used to restore or upscale? What programs do you use? I assme this has to be a whole other ball of wax, but I ignored it because I had not reach it yet until now.

I think I have the Physical setup understood for the most part, but feel free to add. Now that I’m essentially here, where do I go? Do I learn the Ins-andOuts of VirtualDub? I know that it comes with a lot of great filters. I also have a lack of some of the understanding between codecs and encoding, mostly terminology that gets thrown about or is used. I came across a post by LS explaining that rendering is actually an incorrect term, which is what I used for years (lol) and that instead I should be using encoding. I also don’t know where to go from here.

I have a Win10 computer with Adobe Premiere Pro that I plan to use, but besides air-gapping over/transferring my Captures to it for editing, how do I learn this "in-between?" What else needs to be done or what other software do I need? And are their any guides/where I should go to learn this for settings for the programs and explainations/guides I guess? So far, I’ve seen recommended are guides from the user “sanlyn.” This is a recommended way/approach?

Getting the video off the tapes, without losing more quality, is the goal of capturing. Post-capture, it's all about what you need/want to do.

- restore?
- edit?
- share streaming? (Youtube/etc)
- share discs?
- compile into something greater, like a documentary

Each option will require various processing, such as deinterlace or upsizing/upres.
Some need a base level regardless, such as processing out chroma noise, masking overscan.

"rendering" = making something from nothing, aka CG animation. The confusion happens when software misuses the terms, or seems to misuse the term. For example, "rendering" a preview. That's sorta-kinda accurate, though not really. "Generating preview" would actually be more accurate.

"encoding" = converting formats

sanlyn went MIA in 2020, older with serious health issues. :(

This is the sort of thing I want to write guides for, when the new site is done. It will tightly integrate things.

How much is VirtualDub part of Post-Capture though? It sounds like it blurs the lines a little and can be sort of both but undoutably just capture/without losing more quality. I think Sanlyn wrote a VirtualDub guide of some kind that I was going to template off of. I'm surprised there are not more guides unless it really is that easy. I don't know where to begin using the dozens of different filters that come from VDub and considered I'd walk that path when I begin actually capturing, but if I know I can plan ahead of time, even if it's familiarizing myself with the program now, that would be great.

I do want to restore them if possible. I feel like that's a "why not" if you can or have the time. Granted, some may be "good enough" and may not need it, perhaps most. Maybe I won't need to restore? My oldest tapes prior to me being born supposedly were created after being taken to some kind of service to be put on VHS and all the audio was replaced with some kind of fill-in music because it lacked original audio. This is the only origin story I've heard from those tapes. These might be restore potential.

Undoubtedly they have to be edited because these tapes are a mix and match of random events all on the same tape, and I want to separate them out and document them in a type of order to be watched from a drive in a file type system; there may be Christmas 1995 and a random wedding among other things on there, so I have to separate this out. I think the end goal would have them be Plex ready to be watched from that front-end because I see that being the most accessible or handing a family member a USB drive with every "clip" organized in a dated file system.

I think you categorize this as "streaming" even though I don't want to put them on a website or anything-- not even Youtube private/unlisted links. When I see streaming I probably somewhat ignorantly only think of things like Twitch or live streaming. I have lots of family members interested in the results of my project, but they don't even own bluray players and only watch video files from mostly phones or computers So no need for disks. I considered disks personally as I am considering them as one of the potential backup sources for all of this, but I thought I'd consider this another time. I am aware of good backup procedures in general. The good news is I think I have enough hard drive space, about 30TB, so I considered myself to be fine even holding master copies, at least for a while. AI has been a huge buzzword, especially with these image generators like Stable Diffusion/Dalle-E, and I know there are hopes that "one day" these videos will be potentially aided by this technology. What are your opinions on this?

Compile into something greater like a documentary? I'm curious? How does this differ from editing? What's the difference for this procedure?

I use it to collect the parameters that my Avisynth script needs, like the trim and crop values, luma curve, etc. The UI in VirtualDub is easier to use than constantly saving and reloading the Avisynth script.

I'll have to keep that in mind. Do you consider Avisynth essential in your post processing workflow or only certain works? What else do you use besides it if anything else?

I started using Avisynth because it provides QTGMC deinterlacing, then I discovered how useful it is to have a text file that contains all the restoration steps I used on a specific video. I can easily compare two Avisynth scripts side by side, and copy lines from one to the other so I can reproduce steps from one video in another. (And as my skills improve, I can go back later and re-restore the video because I archived the original capture file.)

Then I discovered that VirtualDub's "curves" filter clips the chroma channel but Avisynth's doesn't.

My workflow basically looks like this:

1. Capture the video using VirtualDub. Make note of the brightness & contrast capture settings in case I need to re-capture later. These values go into comments in my .avs script.
2. Make a copy of my "template.avs" and put it into the video's capture directory. This has QTGMC/IVTC disabled at first because it's slow and we need source frame numbers anyway.
3. In VirtualDub, find the trim values (frame numbers where the actual video starts and stops) and put them into the .avs script.
4. Use VirtualDub's null filter to temporarily crop out the head switching noise, then put the parameters into the .avs script to mask out the head switching noise. (Mask, don't crop.)
5. In VirtualDub, aggressively crop out all the black borders and use the histogram and curves filter to adjust luma and saturation, then put the parameters into the .avs script.
6. In VirtualDub, check for glitches and fix them in the .avs script (this needs the source frame numbers also).
7. Enable QTGMC/IVTC (if needed) in the .avs script.
8. Export the .avs video to lossless .avi.
9. Export the audio to FLAC, load it into Audacity, check for clipping, normalize, and export to another FLAC. If there's clipping, adjust recording volume and repeat from Step 1.
10. Use Handbrake to encode the video H.264, and ffmpeg to encode the audio to AC3.
11. Use MkvToolNix to create chapter stops and then to multiplex the H.264 + AC3 -> MKV.

Use AvsPmod. :wink2:

x86/32-bit for Avisynth.
64 for 64 Avisynth.
I need to upload those here, there's no "official" type releases, just random versions in random places.

This seems really involved, but definitely what I think I was looking for. Avisynth I get the picture is a hurdle for many to get used to. I'll have to take it in stride, but in the mean time, I should be able to start capturing everything and then work on that when I get to post-capture/editing. I have a question though: Do you use avisynth on your capture system or is this done on a post processing/editing computer after a transfer?

Avisynth is not used during capture, but it's helpful when I start my post-capture process quickly in case I need to re-capture. I always make notes of my capture settings (brightness and contrast) so I can easily reproduce the capture and make adjustments as needed. For example, I just discovered a capture with audio clipping in one channel, so I need to turn down the recording volume a little and re-capture, but I already put the tape away so now I have to find it again...

Also, I've recently added a step to use the vectorscope to check skin tones and adjust hue. This has made a big improvement on some videos.