What type of TBC will fix this type of wiggling in the video?

[ChibiBoi]

I'm a bit confused about the types of jittering and stabilization abilities that TBCs can do. I've uploaded a sample of a second generation VHS tape with no TBC, and as you can see there are tiny little waves that crawl up and down the footage. It's not as severe as the ripples of the horizontal jittering I've seen on my other tapes.

The question is, will this type of footage be fixed with a full-frame TBC like the Datavideo TBC-1000? Or will a S-VHS VCR Line TBC work fine? I tried playing the video with my Panasonic AG-1970 but it didn't play the tape properly at all, which I found strange considering that VCR has actually corrected problematic tapes with the ripples of jittering I mentioned above. Maybe I need a JVC S-VHS VCR?

This footage was captured with a Panasonic PV-VS4820 and deinterlaced with AnimeIVTC(mode=2) since you can't really see the little crawling movements too well while the footage is interlaced.

I actually just placed a bid on a Datavideo TBC-1000 on eBay by the way, so I'm hoping it will fix the footage.

[hech54]

Maybe I need a JVC S-VHS VCR?

Uh Oh.....you are going to catch hell for that one.
Wait for it.

[jagabo]

A line TBC. Like those built into a S-VHS deck or some DVD recorders. Note that S-VHS decks generally don't play tapes recorded at the slower speeds.

https://forum.videohelp.com/threads/319420-Who-uses-a-DVD-recorder-as-a-line-TBC-and-what-do-you-use
https://forum.videohelp.com/threads/319420-Who-uses-a-DVD-recorder-as-a-line-TBC-and-wh...=1#post1983288

[sanlyn]

I don't think ChibiBoi will be catching hell for his request. But cautions are in order. The glitches with any high-end consumer or prosumer VCR you might find today is that they haven't been made for years, official parts support ended long ago, and amateurs and pros alike have run them into the ground. Many are being sold "for parts" at auction sites, meaning the units no longer function or won't even power up but can often be sources of parts for those that can possibly be rebuilt.

You could search the forum for posts by member orsetto, who is pretty much a reident expert here on VCR's and has apparently bought and used every VCR ever made since 1990 (more or less) and has even rebuilt a few. I feel his evaluations are objective, and from personal experience. My own experience with JVC's, and that of many other owners, was OK while they lasted. None of the three that I owned would play a damaged tape, and EP playback was dreadful. They worked best with tapes recorded on another similar JVC. As of now, at any rate, cautions about machines in poor condition apply to every brand, not just to JVC's. Remember that the TBC and early digital DNR circuitry in high-end JVC and Panasonic "AG" series was relatively primitive, although the AG's seemed to perform with fewer digital artifacts and more accurate color. So when it comes to the high-end stuff you would do well to see what owners such as Orsetto have to say.

A line tbc is called for here, but the question is: do all your tapes play this way, or just damaged tapes? a line tbc would calm things down, but the heads on the VCR you mention are in sad shape as well. The machine appears poorly aligned, too. The Panasonic "PV-V" series dates from 2000 or so. They represent the trend at that time: cut cost to the bone, use tin and plastic, use cheaper parts all around, phase out VCR production altogether and divert investment to DVD. The best of the Panasonics and JVC's were made before 1998. Some of the 1995-96 models tracked so well that they often didn't even seem to require a tbc except for damaged or copy protected tapes. From orsetto I learned that the last of the decent SVHS Pannies was the PV-S46xx series. Earlier Pannies didn't have the second "V" in the model number. By and large, the PV-V's were junk. They achieved popularity by juicing up contrast and insane oversharpening. The last of the decent Pansonics was the PV-8000 series, but were still a clear step step down from earlier Pannies.

As jagabo suggested, many today achieve line tbc status by using a DVD recorder in pass-thru mode. Pass-thru means that you don't record noisy VHS to the recorder, you jhust insert the recorder between the VCR and your capture device. The DVD recorder's Built-in tbc circuitry cleans up line timing and does some frame-sync work as well. No one pretends that a pass-thru will compete with a $5000 shop-grade TBC (adda few more thousands for associated equipment required by shop gear). SOme of them will even defeat Macrovision in pass-thru mode. Not all DVD recorders can be work as pass-thru. The most popular are the Panasonic ES-15 and the Toshiba "RD" series made up to about 2005. Other Pannies could be used as pass-thru, but after the ES-15 the circuitry was cheaper and less talented. There are also a few high-end camcorders around with effective tbc pass-thru functions.

[vhelp]

the sony trv22 camcorder is a good example. it has a very good tbc, knocks out all mv i've thrown at it.

the only problem i have with respect to mv is with the type. there are several types or levels of mv. the one i'm most concirned with is the one that dramatically effects AGC. this is when you see 'flicker' in the image. this type of mv is found on select commercial tapes, like Jurassic Park, and Galaxy Quest, are some examples. the only way i found to get rid of this type of mv was to feed it into two tbc's, (a full frame (stand-alone) tbc) and my dvd recorder's line tbc, but that has another problem. whenever you mix two TBC's together, they sort of cancel out and you have flagging or other tbc errors. i have not determined which of the type/level mv those two tapes have so that i can determine which mv remover (if one exists) that can remove this type of mv for those tapes.

[ChibiBoi]

Looks like I will have to get an ES-15 as a passthrough since it was mentioned in my thread about macrovision as well (though I'll have to use it with TBC mode off for it to get rid of macrovision right?)

This seems to be the right DVD recorder right? http://losangeles.craigslist.org/wst/ele/3725281561.html
I can't tell if it's an ad for the DVD-S42 or the ES15

[sanlyn]

The ES-15 doesn't have a TBC "off". It's on all the time, thru Line 1 input only. The picture is of a DMR-ES15K (The "K" = silver finish). I don't know what the DVD-S42 blurb is about. The DVD S42 is for UK, AFAIK. You will see some "processing effects" thru the Es-15, according to owners. A bit more powerful tbc than later Pannies and Toshiba's, but Toshiba looks cleaner. Black levels thru all Pannies are a bit pumped, so watch that during capture to prevent crushed darks. All DVD's do that to some extent.

[ChibiBoi]

The ES-15 doesn't have a TBC "off". It's on all the time, thru Line 1 input only. The picture is of a DMR-ES15K (The "K" = silver finish). I don't know what the DVD-S42 blurb is about. The DVD S42 is for UK, AFAIK. You will see some "processing effects" thru the Es-15, according to owners. A bit more powerful tbc than later Pannies and Toshiba's, but Toshiba looks cleaner. Black levels thru all Pannies are a bit pumped, so watch that during capture to prevent crushed darks. All DVD's do that to some extent. Toshiba seems to have less aggressive enhancement than others.

Does the ES-10 have a TBS off feature? Because I read that in order to get rid of macrovision, I can use a DVD recorder pass-through with TBC off. Also, that part about the black levels scares me, since I already have a Panasonic S-VHS VCR that darkens the image (which I actually like) and sharpens the image, but if I pair that with another Panasonic, will that make it twice as dark?? If this is the case, what other TBCs or DVD Recorder pass-throughs would work for me?

[jagabo]

Minor changes in levels isn't a problem, just adjust the proc amp on your capture device. You can get much better filtering and sharpening in software than with your S-VHS deck.

[ChibiBoi]

Minor changes in levels isn't a problem, just adjust the proc amp on your capture device. You can get much better filtering and sharpening in software than with your S-VHS deck.

Would you recommend a Toshiba RD-XS over Panasonic ES10/15?

[vhelp]

if you need to get this done now.. pick up the toshiba dr430, it performs the same as the panasonic es10, but is only a dvd-r/rw but tbc works in pass-through. is $130 at bestbuy or $99 or less online..your choice.

[ChibiBoi]

if you need to get this done now.. pick up the toshiba dr430, it performs the same as the panasonic es10, but is only a dvd-r/rw but tbc works in pass-through. is $130 at bestbuy or $99 or less online..your choice.

I don't absolutely need it now, but some time within this week and the next week would be good. I'm just trying to get a better understanding of the options available so I know what's best to correct the waviness, so far the ES-15 or Toshiba RD-XS34 seem to be the best options. I saw something about the Toshiba DR series as I was reading but didn't pay too much attention, how does it compare?

[vhelp]

my money would go with the Toshiba DR-4, because that was the one i saw in a post here (long ago) that seemed to have gotten rid of all the chroma crosstalk (you see this in all vhs captures) that was demo'ed at the time. it was a commercial vhs of the movie, the titanic. the op said it was the vcr, but i didn't believe it then and still don't.

but the probem is finding this model. and i'm not willing to spend $300, $600 or whatever its going for (when it is selling) for any vhs. if you want to see that demo, search the old threads from 2004/2005 i think. the original pics are in there. if you find any recently pics of it after that time period, don't bother, since the posters reconverted to jpeg. i saw a recent one..that's why i mention it. if you find one that says DR-4, be careful, because there are units floating around that say DR-4UK or something like that. they are not the original. if you find one, ask questions, and as for pic of the units front and rear.

otherwise, i'd just go with the dr430.

[Brad]

There is no crosstalk on the Titanic VHS when played out through S-Video...

[sanlyn]

if you need to get this done now.. pick up the toshiba dr430, it performs the same as the panasonic es10, but is only a dvd-r/rw but tbc works in pass-through. is $130 at bestbuy or $99 or less online..your choice.

I'd avoid new DVD machines. None are as competent or made as well as pre-2006 products. The DR430 isn't made by Toshiba. It's a Funai, also sold under the Magnavox label.

The ES10/15 have tbc's, and they do work OK, but the ES15 or the early Toshiba RD series are better than the ES10/15/20 in tbc effect. Your pass-thru unit doesn't even have to have a working optical drive (that's the weak point in almost every DVD recorder). All you want is the tbc.

The difference between the RD and the RD-XS series is that the latter had hard drives. The old Dr-4 and DR-5 were the last of the good non-HDD Toshibas.

[sanlyn]

[QUOTE=ChibiBoi;2236456]

Does the ES-10 have a TBS off feature? Because I read that in order to get rid of macrovision, I can use a DVD recorder pass-through with TBC off. Also, that part about the black levels scares me, since I already have a Panasonic S-VHS VCR that darkens the image (which I actually like) and sharpens the image, but if I pair that with another Panasonic, will that make it twice as dark?? If this is the case, what other TBCs or DVD Recorder pass-throughs would work for me?

None of the machines mentioned can have their tbc disabled. I've never heard that story about the Macrovsion business. My Toshibas and Panasonic have defeated Macrovision on every tape I threw at it except one. I think that retail tape was permanently damaged anyway (it was a repackaged rental), so I don't think the frame hopping was due to Macrovision. It was more likely $12 wasted on a bad piece of tape that nothing could fix, and my outboard full frame tbc had no effect on that problem.

[jagabo]

I already have a Panasonic S-VHS VCR that darkens the image (which I actually like) and sharpens the image

The sample in your first post was way oversharpened. It has a very pronounced halos. That's not true sharpness. You should disable that feature.

[sanlyn]

And it makes the broken, ragged, wiggling lines look worse and more difficult to repair later.

[ChibiBoi]

This was my AVISynth script I used for my sample:

setmtmode(2)
AVISource("C:\Users\Public\Recorded TV\smrmovie -506ms.avi", audio=false).AssumeFPS(30000,1001)
assumetff()
converttoyv12(interlaced=true)
animeivtc(mode=2)
chromashift(l=-4,c=-4)
crop(4, 4, -4, -4)
Lanczosresize(640,480)
fastdegrain()

I didn't use a dehalo filter because they tend to make the subtitles look pretty bad I just checked my Proc Amp on my capture settings and they seem to be set at 2, which was the default. So I set it to 0, and then set the contrast from 32 down to 30 and captured a new clip, and these are the results after adding the exact same chain of filters. I've also attached untouched screencaps from the original capture files.

My concern is this: would pairing two Panasonic devices together exacerbate the sharpening and contrast problems so much that my capture device's Proc Amp would be useless at fixing it? For example, would it turn Sailor Moon's skin the same shade of white as her costume? And then if that were to happen, and I used the Proc Amp to decrease the contrast, would that make her skin the same shade of gray as her costume? (maybe not in this particular scene but I do have some tapes where if I left the proc amp contrast at the default of 32, the captured video had her skin super white)

[sanlyn]

The coding gurus will have to look this one over. The one thing I noticed was...

This was my AVISynth script I used for my sample:

....
....
crop(4, 4, -4, -4)
Lanczosresize(640,480)
.....

The lines in bold made my eyes smart a bit. Then I saw the that crop/resize was followed by denoising. rather than preceded by it. Having no idea what the unprocessed source looks like, I must withhold further comment on the processing. That's some rather messy chroma smear; It almost looks like tape played via composite. But maybe that's just the way the tape looks. It sure is rattling around inside the player, though (You'll be happier with a line tbc).

Using a proc amp for anything more detailed than very basic levels and (perhaps) saturation correction can make life difficult because of the way VHS relentlessly changes color balance from minute to minute, and sometimes second by second. I guess it depends on the workflow.

My concern is this: would pairing two Panasonic devices together exacerbate the sharpening and contrast problems so much that my capture device's Proc Amp would be useless at fixing it?

No.

For example, would it turn Sailor Moon's skin the same shade of white as her costume? And then if that were to happen, and I used the Proc Amp to decrease the contrast, would that make her skin the same shade of gray as her costume? (maybe not in this particular scene but I do have some tapes where if I left the proc amp contrast at the default of 32, the captured video had her skin super white)

Play it on your tv without the proc amp. The same color changes will occur. That's because VHS has not been called video garbage for nothing. You'll go crazy if you try to make a shortcut out of VHS color correction by trying to do it all during a single capture run. It's possible, but only if you start and stop every few minutes and change your settings. By that time, you could have corrected most of the color in software. As it is, your sample changed levels and color balance quite visibly several times within only a few seconds. That's typical for VHS. Find a decent middle point of correction for capture that provides for the worst scenarios you're likely to encounter, and tweak later. You'll have to tweak anyway, regardless.

[ChibiBoi]

Having no idea what the unprocessed source looks like, I must withhold further comment on the processing. That's some rather messy chroma smear; It almost looks like tape played via composite. But maybe that's just the way the tape looks. It sure is rattling around inside the player, though (You'll be happier with a line tbc). The sample mkv looks cleaner than the first one, but the bitrate seems low for noisy vhs and fast action.

I've uploadde the capture file which was done with Lagarith Lossless Codec. The mp4 file was encoded with CRF23, though I may consider bumping it up to 21 or to a lower number. http://www.mediafire.com/?kavaakpathjaoe1

And as for color correction, my main focus is to neutralize whatever it is that the Panasonic VCR is doing to the video footage (from what I understand, it's just the contrast?) I've attached a snapshot of something I encoded without any type of color correction from another tape, captured at default settings. This is the type of thing I don't like seeing. Unfortunately, the thing is I only have 2 VCRs, my most reliable one that I use the most is the PV-VS4820 which generally yields good results but I'd like to undo the contrast bump I've been told it applies to the footage. The other one as I mentioned before is the AG-1970 which has its own set of problems with colors.

[Brad]

That's some rather messy chroma smear; It almost looks like tape played via composite. But maybe that's just the way the tape looks.

I believe it's an old fansub, which by their nature are multi-gen.

[jagabo]

Try adjusting your capture card's proc amp. You're getting crushed blacks somewhere around Y=24:



You need to take care of that first. There's no way to fix it later.

[sanlyn]

I believe it's an old fansub, which by their nature are multi-gen.

I'm not familiar with fansub (had to look up that one. You learn something new every day!). You're must be right, I was thinking something like that.

I have old and supposedly "digitally remastered" (quote/unquote) retail tapes that look similar, and worse. Likely a lot of the noise and other problems are in the maker's processing, and it looks as if the tape is fading. These typical variations and glitches don't bother many viewers. But for others they're annoying. A headache to fix. How much trouble to take depends on how much one values a source. I just let some oldies go, as they don't seem worth that much effort.

This is another reason for capturing old or poorly produced VHS to lossless media. Then, piece by piece, it's repaired. Some people (including myself) end up correcting scene by scene; I've even got down to frame by frame for what I consider important. But you can't do with that with everything. Trying to correct everything during capture is even more frustrating, IMO. The mkv has some glitches, but it's still easier to watch than the earlier post.

ED: my internet is really slow this morning. Downloading the AVI now. Thanks, ChibiBoi, for some of the original capture.

[sanlyn]

Re: the sample smrmovie test.mkv in post #19. I've seen animeivtc metioned but never used it, so I just now downloaded it (something else to learn!!). ANyway, the image below is frame 555 from the mkv. I see frame blending here. Would one have to follow animeivtc with another step to clean this up? I see several blend effects in this sequence.


[Attachment 17539 - Click to enlarge]

[ChibiBoi]

Re: the sample smrmovie test.mkv in post #19. I've seen animeivtc metioned but never used it, so I just now downloaded it (something else to learn!!). ANyway, the image below is frame 555 from the mkv. I see frame blending here. Would one have to follow animeivtc with another step to clean this up? I see several blend effects in this sequence.


[Attachment 17539 - Click to enlarge]

AnimeIVTC at mode=2 actually gets rid of the blended frames that were encoded onto the video source. These tapes come from anime laser discs from the 90s and most of what I've seen, if not all, of these laserdiscs have field-blended frames during the production's 23.976 -> 29.970 conversion. That being said, the particular portion of the video you're talking about is a stylistic choice on the animators to show slow motion when the girls get hit. So no, there's no need to follow AnimeIVTC up with another filter because it actually already cleaned up the frames which were erroneously field-blended.

Try adjusting your capture card's proc amp. You're getting crushed blacks somewhere around Y=24:


[Attachment 17536 - Click to enlarge]

You need to take care of that first. There's no way to fix it later.

I've seen you post histograms in other threads while I was searching the forums, can you explain how to read them? I couldn't find anything useful on how to read them when i searched the net lol And also, to adjust, you're saying I should change the contrast on the procamp some more right? I've already changed from the default 32 down to 30, so I should lower the contrast a bit more to get rid of the crushed blacks? This is one of the problems that I read about from Panasonic VCRs. And like I said before, I'm not too keen on color adjustments except to undo what the Panasonic has done to the original tape.

[Brad]

What is your capture device?

[ChibiBoi]

What is your capture device?

I have a StarTech SVID2USB2 and I capture on my laptop. I've had it since last November and it's worked pretty reliably so far.

[jagabo]

I've seen you post histograms in other threads while I was searching the forums, can you explain how to read them?

The histogram I posted was with AviSynth's Histogram() filter. This is really a waveform graph, not a histogram:

https://forum.videohelp.com/threads/340804-colorspace-conversation-elaboration?p=212156...=1#post2121568

In short, it's a graph of the intensities of the image. The height of the line represents the intensity. The brightness of the graph indicates how many pixels had that intensity.

When working with YUV you want to keep the waveform out of the yellow areas. You want blacks to be just above the lower yellow band (Y=16), and the brightest brights just below the upper yellow band (Y=235).

AviSynth's Histogram() draws the graph at the right of the frame and works on columns instead of rows. To get a more traditional horiztonal waveform graph I turn the frame 90 degrees clockwise, call Histogram(), then turn the result 90 degrees counter-clockwise. Or you can use the VideoScope() filter. It draws a horizontal graph below the frame (actually it can draw at the bottom or the right of the frame and can show chroma graphs too) and can include tick marks within the graph. But it doesn't clearly demark the area below Y=16 or above Y=235.

[ChibiBoi]

Thanks jagabo, I understand how to read it now, but how can you use it to tell if there are crushed blacks? I'm assuming the indicator on the graph isn't the positioning near the yellow bands, since you said the darkest darks should be just above the yellow band. So is it because of the sheer intensity of the blacks at that position? And from what I described above, setting the capture ProcAmp to a lower contrast should solve the problem, correct? Though from what I saw by setting contrast 32->30, it seems to have dulled the color of the video as a whole