Yes, I have head cleaning tape. What I noticed, that often at the beginning of the tape jitter can be bigger than at the end - so it looks like it not related to the head pollution.

johnmeyer has lots of experience with this. Interested to hear what he says.
Certainly some motion tracking could be a big help.

I don't see the jitter in the camcorder viewfinder, I am ordered 220V transformer to run my old PAL TV and PAL S-VHS VCR. I read that directing signal through VCR may help. I will take a look does the picture on TV have the same issue. By the way, algorithm to fix the jitter is very simple - each line should be scale to 720 horizontal pixels to eliminate black "wave" on the left of each frame. I believe that exists software that can do it.

I'm slightly confused, you say in your first post that the problem persists even if you use a TBC but where's the TBC? Is the TBC working?
Certainly a TBC should fix the problem but if you really are using one and that's not fixing it I doubt running the signal through a S-VHS VCR will address the problem.

Bob.

Since my name got mentioned ...

Can you post 5-15 seconds of video that shows the problem? Everyone is assuming that this is a problem caused either by time base issues or dirty heads, but there are other possibilities as well. If I can see the problem, I might be able to come up with a way to adjust the equipment for a better transfer or, lacking that, a way to reduce it using software.

Bob's point about the TBC is a good one. Most Hi8 equipment supposedly has a TBC built in, but I don't know if that is always the case. Also, since the only 8mm deck I have is standard 8, I don't have direct experience for what interface Sony provided -- if any -- for enabling/disabling the TBC circuitry. I can say that with VHS tapes, there are sometimes reasons why it is useful to disable the TBC circuit, and my two main VHS/S-VHS decks do have that ability.

From our experience the best piece of kit for digitising Video8 and Hi8 tapes is the old GV D800 VCR. Our one's transport controls are getting a bit dodgy from the amount of use however any NLE will give you transport control via firewire so that's not an issue.
1000's of hours of old 8mm footage has been captured using that VCR by various clients using a variety of NLEs. I've yet to have anyone mention having a jitter problem. Serious head clog problems from the old tapes shedding oxide is a constant issue but that doesn't cause jitter.

Bob.

First of all, Hi8 is an analog format, Digital8 (or D8) is digital and has essentially the same video encoding as a DV camcorder. Both Hi8 and D8 use the same tape cartridges. I only know of one Hi8 camcorder that had a built-in TBC and that was the Sony CCD-V5000 Hi8 Professional model.

Secondly, with Hi8, you're dealing with mechanical tolerances so precise, it borders on the realm of spiritual. In almost all instances, you will have the best luck playing tapes back on the machine on which they were originally recorded.

There were a few early D8 camcorders which would play back analog Hi8 tape and encode them to DV format and send that digital signal out the Firewire port. However, D8 camcorders do NOT correct time base errors of analog recorded 8mm/Hi8 tape when these tapes are played back directly in the D8 camcorder. This has been a common misconception for years, even among members of this forum.

The best way to digitize 8mm, Hi8, VHS, S-VHS, BetaCam or 3/4" U-Matic tapes is to have a properly adjusted analog playback machine and play the tapes back (preferably using the Y/C S-Video ports, if available) and running that signal through a real TBC. The lowest cost dual-field, full-frame TBC of which I am aware is the AV Tools AVT-8710 which can be found for under $200.

There are a few high-end consumer (Sony EVS-3000) and professional 8mm/Hi8 machines with built-in TBCs. Some VHS/S-VHS machines have built-in TBCs but they are typically single-line TBCs and not full-frame.

However, D8 camcorders do NOT correct time base errors of analog recorded 8mm/Hi8 tape when these tapes are played back directly in the D8 camcorder. This has been a common misconception for years, even among members of this forum.That's news to me. I've read that so often, both here and in other forums, that I always thought it was true. In fact, many people used to buy old semi-broken camcorders on eBay just so they could use them for pass-through transfers and take advantage of the "built-in TBC." I guess that was all for nought.

Since I never owned an 8mm camcorder, only a deck, I never could confirm this often-made statement (about 8mm camcorders having built-in time base correction).

I learn something new every day.

Still waiting for a clip from the OP.

I am inclined to look at John Cline's comment about built-in TBC somewhat circumspect. I have top of the line cameras built by Sony for Digital-8 and Hi8 (vintage late 1990's). The D-8 has an internal switch to engage built-in TBC function on/off. The functionality of this switch AFAIK has never been fully explained by Sony in user documentation that I have ever seen. I have had the opportunity many times to play both Hi-8 recorded tapes and D-8 authored tapes in this gear and attempt to switch TBC on/off in tests. In all practicality this has never made any difference in playing of tapes.

In fact because I maintain the equipment and keep it all clean with good maintenance of machine and tape, I have never had trouble delivering digital firewire or Y/C style capture of either analog or digital tapes. Because D-8 servos control the playback speed of either tape format - that is the most problematic part of the equipment compatibility.

A great service of the D-8 is being able to digitize signal on the fly of the pass thru of analog inputs.

" I have had the opportunity many times to play both Hi-8 recorded tapes and D-8 authored tapes in this gear and attempt to switch TBC on/off in tests"

The TBC would play no part in D-8 transfers since it is purely digital-only. No processing is done except to clean up the noisy digital magnetic signal.

The only problem I have had with the HI8/Digital is the tapes had some rolled over edges (caused by guide alignment) so minute you could hardly see them on a magnifying glass. What I am saying is like .005" rollover.
JJK

JM,
I'm as mystified by John Cline's comment as well.
My understanding is that the sole purpose of the TBC in D8 gear is because they can play out 8 and Hi8 tapes over firewire and to do that they must have a TBC.
Like many I've also used analogue to digital pass through on my D8 camera to capture VHS. Proof of the pudding that there's a TBC in that chain is doing that could also strip Macrovision.

What I'm not clear on is there are two different types of TBC, one is line and the other frame based. Off the top of my head I can't remember which but the ADVC 300 has one type and D8 gear the other.

Certainly Sony claim the GV D800 has a TBC and specifically as already mentioned it is only used for Hi8 and Standard 8 tapes. Page 51 of the manual covers this as well as the DNR.

Bob.

[edit]
I just found this video on YT:

It compares a video 8 capture via S-Video through a ADV-8710 TBC and into an ADVC-110 with using the TBC in the GV-D800 and firewire.

Bob,
The tapes are 17 years old, 8 years ago I did some experiments in digitizind them. At that time the result with TBC was better then without. Now without TBC digital to analog converter can't convert video at all. I have mentioned above AVT-8710, my camcorder is Sony-TR880E and Pyro AV-Link as digital to analog converter. I will create short video and will post link here. Thanks for everyone for the help.

Here is the video: http://yadi.sk/d/pHKRR0Ac7g-pA - file size is 14 Megabyte. Video is MPEG-2 rendered by Vegas Pro 12. Press blue button. I can upload file to other place too.

Years ago (mid-1980s) my dad had an 8mm Sony camcorder and it malfunctioned like this. He sent it back under warranty and it was somewhat fixed, but the problem later returned.

I don't have much knowledge of how the 8mm system works so I can't give any sort of definitive suggestion. The most obvious thing is to try another camcorder or deck. This would help determine if the problem is stretched tape, or whether it is the deck.

If it is a time base problem then it is something specific to this technology. Time base errors in other types of analog video usually show up as either flagging at the top of the screen, or herring bone "teeth" that appears on vertical detail like flag poles. This video obviously has issues with scan lines not starting at the same instant in time, but the artifact is not what I normally see.

So, try to find another deck that you can borrow, and capture a few minutes of a problem tape on that machine and see what happens.

That's pretty bad.
I still have a Pyro AV Link, I never got it to work properly. I know many did but they were all in NTSC land. Mind you that capture looks better than anything I could get out of it but you have got a TBC in front of it and that was a significant part of the problem with the Pyro.

Bob.

John,
I also have results which are much better then this one. The problem in the particular tape I think. The question is it possible to improve the result by some software.

Something's definitely screwy.
I have Regular 8 analog tapes going back to the nineties that play and capture to digital far better than that.

I also have results which are much better then this one. The problem in the particular tape I think.That's good information to have. It doesn't entirely eliminate a possible problem with the deck because, at least with VHS decks, when they get out of alignment, they can still play some tapes but not others, depending on how the azimuth was adjusted on the deck originally used to record the tape.

The question is it possible to improve the result by some software. I did try the obvious solution, which was motion stabilization. Mercalli made the problem much, much worse, both with and without rolling shutter compensation (the wobble has a vague resemblance to rolling shutter artifacts). I also tried Deshaker, with and without rolling shutter compensation but it too made it worse.

Several months ago, over in the doom9.org forum, there was someone who attempted to write a "software time base corrector." He tried to use the variations in the black area around the edge to create offsets for each scan line. It sounded good in theory, and he claimed it worked, but no one else could get it to do anything.

A few months ago I was faced with a distantly related problem, namely remove floating horizontal noise bars from an old amateur Kinescope. Here is the before/after:



I looked at a magnified view (using pan/crop) of the upper left corner of your video (the sky) where it meets the black edge, but did so after using the "trick" so that Vegas displays one field at a time instead of both fields simultaneously. The edge was actually quite crisp and clear. Having seen that, it might be possible to pick an offset from the edge where there is always some video, and then measure the distance from that point to where the video is pure black, and then create an offset for that scan line equal to that difference, minus a constant. This is pretty similar to the failed software TBC attempt I just described, so perhaps it won't work.

If you want me to pursue this, and if you don't have a problem with my posting your clip on Mediafire (a hosting site that the people in that forum prefer) I can see if anyone has some specific ideas to offer over at the doom9.org forum.

John, please post my clip to the forum you mentioned.

John, please post my clip to the forum you mentioned. Done.

8mm Videotape Problem - Possible TBC Issue

We'll see if anyone in that forum has some ideas.

I have a good and bad news for the community. I will start from the bad. I capture already 10 tapes from 33 that I have and all should be repeated. The good news are that I fixed the issue - even from the worse tape video now is perfect. The trick was that I need to select manual TV system detection mode on the TBC (select PAL manually). Thank you all for the interesting discussion. Special thanks to John. John you may post the solution to that forum.