Been lurking here for sometime as I've prepared a project and got heaps of useful info. I'm working on a family project which is using a lot of 30 year old VHS as source footage. I'm capturing from original VHS camera tapes with a DeckLink card via a borrowed Edirol 2-channel video mixer for TBC (after a fashion). Results have generally been great but on some tapes I'm noticing these two moving helical bands moving in different directions horizontally across the screen (see attached animated GIF). It doesn't look like ground-loop interference but is it related to the VCR heads? The machine and all tapes were cleaned several weeks before started digitising.
Big question is, is this an artefact that can be mitigated either pre-/post-capture?
Thanks in advance.
+ Reply to Thread
Results 1 to 16 of 16
I've never seen that one before. Sort of a windshield wiper effect.
If other tapes transfer just fine, that more or less means it is not the heads.
However, do you know if these tapes are recorded at a different speed than the other tapes which playback without any artifacts? If so, it could be something to do with alignment that only shows up at certain speeds. The intermediate "LP" (4-hour) speed was dropped from many decks in the 1990s. They could still play back, but could not record. I suspect that the playback on such machines might be marginal.
Finally, you might want to look really closely at the captures that you thought were good, just to see if there is a hint of this artifact. I have often found that when I get problems like this, that ALL of the captures are affected to some degree.
Others can correct me if that isn't quite right.
They’re al SP tapes - recorded out of a higher-end Panasonic VHS camera circa 1991. I hear you about the effect possibly not being visible but present elsewhere (certainly possible). I might try routing through a DVR I also have on hand and see if I get a different feed. It’s perplexing for sure!
Aside from the "windshield wipers", it looks like there are two distinct thick horizontal bands at the top 1/5th that differ in brightness from the bottom 4/5ths of the image.
Personally, I would suspect the Edirol mixer before anything else, unless it's a newer variety. I see a Videomaker review of the Edirol V-4 from 2004, which was during the Capacitor Plague. Have you confirmed that the same tapes direct from VCR to TV show the same thing?
Last edited by vaporeon800; 12th Feb 2020 at 01:43.
That is another video image, not in sync with the main one. Either bad cables is causing cross-talk, something in the mixer is causing cross-talk or if it is an off-air recording there was another TV station breaking through. It would be very difficult to remove!
I don't think it's related, but I'm curious: how?
That is another video image, not in sync with the main one. Either bad cables is causing cross-talk, something in the mixer is causing cross-talk
I reviewed the tapes again and it definitely only appears on two of the six so perhaps it's something environmental affecting the mixer or a running time issue for it? In the worse cases, a strong ghosting artefact appears and moves with the bands (couldn't find the worst example but see highlighted area of attached).
[Attachment 51990 - Click to enlarge]
I didn't know analog had artifacts. Aren't artifacts digital?- My sister Ann's brother
Stanco, F. & Allegra, Dario & Milotta, Filippo. (2014). Tracking error in digitized analog video: automatic detection and correction. Multimedia Tools and Applications. 75. 10.1007/s11042-014-2068-1.Why? Odds are tapes will be ruined from this, not cleaned.
So, re-arranging cables and recapturing seems to be going ok. I did a test capture of a problematic area and this was free of the problem. Now recapturing the full 3 hours of footage again so will be interesting to see if its something that kicks in over time. Might be a case of resting the gear between captures.
[Attachment 51993 - Click to enlarge]
Another example of many. 1978 patent, Yves Faroudja
Another object of the present invention is a more effective elimination of dot patterns and distortion artifacts from a color television picture through comb filtering of chroma and adaptive comb filtering of luminance.
It will also be appreciated by those skilled in the art that the circuitry implementing the comb filter system 10, while described herein as employing analog techniques, may easily be implemented with digital techniques and hardware.
Obviously what you're seeing is ghosting of the horizontal and vertical blanking intervals. I seem to recall seeing that when a new recording was made over an older one but the erase head wasn't working. So you get a slight ghost of the original recording.
I don't know about the "windshield wiper" artifacts, but the horizontal problems might be fixable using some of the techniques I developed here:
Bad 1950s Kinescope - Hopeless?
This was the result:
I was rather proud of that one: it turned out far better than I ever expected, and had it been a paid gig, I could have actually done even better, fixing some of the residual artifacts at the top and bottom of the screens.
I don't think it will work with the diagonal wiper blade problem because that will be tougher to track, a requirement for developing the moving exposure mask.
Well I’m pleased to report that switching the cable between the vcr and the mixer and plugging into a different input on the mixer seems to have eradicated the issue. I discovered even a slight knock to the cable near the mixer input would disrupt the signal. With the new arrangement I can twist the plug in the input while capturing and still get a rock solid image.
So thankfully NOT baked in! Onwards and sideways... Thanks to all for suggestions.
Shielding and ground loops can be a bitch to troubleshoot and correct. I'm glad you have it figured out and working.
Be glad you didn't ruin a tape (though since you're not done, that may yet to be seen).
I got walked through the technique by a former colleague who is a professional audiovisual archivist who works with a lot of tape formats.