Line TBC vs. frame TBC? Still confused!
 

I've been at this for two years, and I still can't explain Line TBC and Frame TBC to friends. Even after reading the official guide:
https://www.digitalfaq.com/forum/vid...time-base.html

My workflow is: JVC SVHS player (with Line TBC function) > Panasonic ES-15 (with weak Frame TBC, or is it Line TBC?) > Datavideo DVK-200 (for it's weak Frame TBC powers) > AIW 9800XT.

1. SVHS's line TBC straightens tearing, right?
2. ES-15 also straightens out tearing? So that makes it a line TBC, right? Not a weak frame TBC? This is what confuses me.
3. Should I turn off the line TBC function in the SVHS player if I used the ES-15 in my workflow?
4. Assuming that one or both of the devices in the flow before help stabilize(?) the image, then the DVK-200 helps keep audio and video in sync? Is that, basically, the reason I am using it?

I know this has been asked a thousand times. I'm still trying to wrap my head around it. :smack:

VCRs have only line TBC (where equipped), DVD recorders have a combination of both line and frame with audio sync, A classic TBC like the TBC-1000 is only a frame TBC and does not sync audio, it is strictly a video TBC with audio bypass with one video frame delay, not noticeable to the average person.

General consensus is to turn off the line TBC on the VHS machine if you're using the ES15 as it does the same thing, or may do it more poorly if it has already been done by the VCR. In such a setup, a fair amount of expense can be saved on the VCR (by not needing a line TBC in it) if you always plan to use the ES15 in your chain. Purists would say that ideally you shouldn't use the ES15 if you aren't seeing issues in the picture with the VCR's line TBC on though.

With your chain as you have it listed, you technically have something like 7 analog to digital and digital to analog conversions going on. So turning the line TBC off in the VCR gets rid of 2 of those and the ES15 takes care of the line TBC part:

1. VCR's line TBC A->D, then D->A at output
2. ES15 - A->D, then D->A output
3. DVK100 - A->D, then D->A output
4. AIW Card A->D




I'd describe TBCs as such:

Frame TBC': "Frame rate stabilizer" - buffers incoming frames/fields which are output slightly irregularly from mechanical media compared to the digital precision of "frame rate arrival" that modern capture devices usually expect. All these will do is prevent dropped frames and potentially prevent vertical jitter (which in itself is caused by a frame arriving slightly early or later than expected) If your capture card gets confused and drops enough frames, audio may also get out of sync. With too few frames present and played back at the normal rate of 29.97, the video will run out before the audio and that will make it seem like there's an audio delay, when in reality, you have missing bits of video and all of the audio. Compressing the audio to the right length won't help either because you don't know where the frame drops actually occurred. it could have been a big chunk in one spot and nowhere else, so the audio might not get out of sync until a certain point in the video (after the dropped frames).

Line TBC: "Horizontal aligner" - left-most edge of each horizontal band of video information (line) is re-aligned to begin at the left-most edge of the display. This prevents horizontal waviness and geometry issues.



However, I'm still not totally clear why you'd need a frame TBC if you aren't getting any "frame drops" or vertical jitter at all in virtualdub without one in your chain. Frame insertions *alone* I think are ok as I believe that means the VCR's clock is slightly slower than the capture card's. No data is actually being lost if there are no drops and audio should not go out of sync if virtual dub is doing frame insertions on the fly as needed to keep the frame rate up to 29.97fps which will be the video playback speed. You wouldn't want to find out after the fact that you have a bunch of dropped frames and vertical jitter and have to repeat a capture, but if you've run 20 tapes and had never seen a single dropped frame, vertical jitter, or had audio sync issues, I'd be pretty confident that the specific hardware chain is typically not going to drop frames and be better to run without one (as it's an extra analog to digital and back to analog conversion that you wouldn't need). Others will disagree, but I'd love to hear why that is.

IMO: A line TBC is essential. But keep in mind that it "bakes in" to the output video line any timing issues for all subsequent devices in the signal path, so only one line TBC should be used. Use the one that works best, which may well be the one in the VCR if you are using one of the recommended VCRs.

A frame TBC should be available to resolve any frame-rate variances but may not be absolutely needed in all cases. Again, the first one in the chain will "bake in" any frame timing issues. There is no advantage to having more than one frame TBC.

TBCs come in varying degrees of effectiveness and strength, they are not all equal and what works with one video may not with another. Also, when in series in the signal path they may interact resulting in a negative effect on the signal. The ES-series has a reputation of being a weak TBC but often is better than nothing.

Keep in mind that this is all old analog gear subject to performance changes due to component aging. Individual pieces of gear my not conform to the original specification. Bottom line is use what works best for you with your specific gear and media.

It's not the case in all vcrs with a TBC function but in in the case of the digipure JVC SVHS VCRs the signal goes through the analog->digital->analog conversion whether the TBC/DNR function is active or not.

The DVK-200 doesn't do a whole lot in this setup, the ES15 already buffers and ensures a stable output signal unlike the in-vcr TBC so the DVK is redundant and is just an extra analog->digital->analog conversion. I would just leave it out unless you are capturing a tape with copy protection and the capture card reacts to it, as the dvd-recorders can re-add macrovision to the output if they detect it, or if you are dealing with a tape that has blank sections recorded on it as that can cause the panasonic dvd-recorders to turn off the outputs because it thinks nothing is connected.

Ah. I was unaware of that. Which S-VHS VCRs on the recommended list do not do that? Or is it only the Digipure ones that you can't skip that A/D step? I just want to make sure I'm aware of what's all happening as I'd like to avoid extra A/D D/A steps whenever possible. Granted, most of the time you'd want it for the line TBC anyway, but this is good stuff to know for eventual testing/comparison purposes.

I tried to explain this to my friend the workflow because he only has a few VHS’s that he was going to send into Legacybox and I explained my gear and why this method is better. I offered to do them for free. Then he told me he was going to take his VHS combo unit to get someone to service it then use it. I thought idiot but I just said okay and won’t bring it up again.

If your friends and family are like mine they are probably telling you that you can just get an easycap and a laptop or a clearclick and just use a regular VCR or something like that. Those clearclicks are awful and they encode directly to h.264.

The compression shortcuts that h.264 uses really throw out a lot of your data. Also they’re hard for your editing software to decode later. They are meant to be a delivery format.

The clearclicks are just using it because it’s popular and they have to output to a small file size like that and lose data because they can’t do a huge lossless file. This guys whole channel is a clearclick and a POS VCR https://youtu.be/ZdwA5n02Gqc?si=sf-reFFq4ESUGTPD Thes have a bunch of dropped frames and you can see awful deinterlacing especially on the circle on the matt on these. Also that circle is supposed to be all yellow. These devices are a shit show. If he was using a lossless codec he couldn’t have frame drops like that because it would stop the capture. A capture card can’t handle that in a lossless capture.

Virtualdub is a lot more stripped down and dedicated to vhs capture than a regular NLE. The Huffy codec they use is less than half the size of lossless but it doesn’t have less video data than lossless. I’ll attach the media info of my lossless AVI in Adobe and my Huffy in Vdub. Huffy is easier to capture and takes up less than half the room to archive.

The frame servers they use to edit in code with handle deinterlacing, upscaling and other things really well also they edit in YUV when they can and handle the YUV to RGB conversion well.

All of the restoration and quality relies on getting that clean losslessley compressed video at the beginning with your gear.

Your friends probably aren’t going to get all of that though. This gear was always meant for people wanting to do this correctly and not your normal person that just doesn’t care.

This might help you understand timebase errors better. https://youtu.be/HEXLlqJQ7Oc?si=k8f8kNjMrLkVmih8.

A bit unrelated to the topic at hand but man that does look bad, compressed, flagging on the top area, jitter, etc. Honestly after a while of using the proper gear to digitize these tapes the right way, and seeing the results of a carefully deinterlaced and restored piece of VHS/video8 footage that had a line and frame TBC used, it's like I can't go back and look at any old 30FPS captures that were done with a lousy Elgato card, those results are very bad and noticeable now that I'm so use to how this footage should properly be handled.

Those clear clicks might "work" but if they do, it won't be good. That's what I believe anyway.

To be fair to that guy they’re not all that bad. They are all really bad though and I think he’s causing those errors. I doubt his tapes are too bad. My tapes are a different story but I’ll do my best with them.

To the OP if you want to know why you need A TBC to correct the analog signal look into the differences between analog and digital signals and you will see why you need a perfect signal when you are going from analog to digital. It will also explain things like why cable length is important with analog and not really as much with digital. It’ll well you why digital is better for storage. It’ll explain why digital is so precise. YouTube isn’t bad for walking you through this topic but it’s also in the guide on this site.

Analog is the world around us and digital is a binary code.

Thank you! From what you are others have written, do I have the following right?

• JVC with TBC / has line TBC / helps reduce tearing effect
• ES-10/ES-15 / has line TBC and (weak) frame TBC / helps reduce both tearing and horizontal jitter
• DVK-200 / has (weak) frame TBC / helps reduce horizontal jitter

If that is accurate, then these should be my only workflows?

Method #1: Use as a first choice
JVC with TBC turned on (line TBC) > Datavideo DVK-200 (frame TBC) > AIW 9800XT

Method #2: If Method #1 yields bad result on difficult tape then use this
JVC with TBC turned off > ES-15 (line+frame TBC) > AIW 9800XT

That's actually not always accurate, as the ES10/15 simple non-TBC frame sync can be overrun fairly easily, and it can output mistimed enough to cause problems with capture cards. It's not a frame TBC. This is source dependent, and almost random. While most problems are nth gen in nature, many tapes that "look fine" can have terrible signal quality.

But again, noting that anti-copy is essentially just an artificial video error, and natural errors can appear the same to the detections. So even tapes without Macrovision or anti-copy can have issues. Because, again, ES10/15 is not a TBC. However the DVK is a weak frame TBC, and can overcome some anti-copy. Not all.

Or this. And a few others various random issues.