VHS archiving project, nothing but problems?
So I've taken on this project of archiving a ton of old VHS tapes for my brother in-law. Since the late-'80s, he's recorded between hundreds of VHS tapes, all filled with old and new wrestling shows.
I originally tried to copy the tapes directly from VCR to computer via a USB capture card. I've had problems with this method. On the computer, the scene would flicker every couple of seconds and flash a still frame from a few seconds prior. This re-occurs every few seconds, depending on how deteriorated the tape is. In most cases, this can happen over 100 times every half hour. So I was puzzled at first. The tapes played fine on the TV, but not through the adapter. What does the TV have that my capture card lacks? :question: So I pondered for awhile and did some research and that led me to discover (1) old VHS tapes sometimes require a Time Base Corrector to keep the image intact and (2) that TVs by-and-large have TBCs built-in to handle these issues. Since my TV doesn't have a video output, I can't simply route the connection. For me, it just needs to look the same as how it looks when played back on TV. I've tried several VCRs and up to this point, the one I've had the best luck with is the JVC-XVC29. This was the one my brother-in-law used for some of his recordings and it has a “Video Stabilization” feature that helps remove most of the issues. But then came the next hurdle. How do I get a TBC? I can't buy one standalone because they're outrageously expensive and hard to come by when you live in Canada. After some time had passed, I stumbled upon a forum (probably from here) and learned that DVD recorders typically have a TBC built-in. I bought one second-hand right away. The one I have is a Toshiba D-RW2. It's okay. It doesn't prevent every little flicker, but it cuts it down to about 15-16 flickers in a 2.5 hour span of footage which is much easier to tolerate and to fix in post-processing. Now you would think that it's smooth sailing from here, right? Not really. I have a new problem. The old tapes come out looking very dim. Not only that, but it doesn't affect the whole tape for some reason. One tape for example, it only affects the first 2 hours and 12/13 minutes. When the next fight begins at that time, the brightness returns back to normal. The link below is a sample video I took demonstrating the issue. I tried a 2nd. VCR to plug the main one which plugs into the TV. That didn't help. Instead, I was gifted with what I think was Macrovision. The image looked like it was going through trouble with tracking every couple seconds. Off, and then on. Off, and then on. I had tried another DVD recorder. This one was a DVD/DVR from Panasonic and it also exhibited the same issue. https://streamable.com/2u5fc. In this video, you'll briefly see how the prior footage was dim and then the last fight stripped the color entirely. After that, I fast-forward the tape to around 2 hours and 12/13 minutes in. At this time, the next fight begins and the brightness comes back to normal. Just to reiterate my point from earlier. The picture accuracy is on point when the connection comes straight from the VCR. Doesn't matter what output device I use (TV or computer). If I use the DVD recorder as a pass-through device, it'll also look dim on the TV. I'm not sure what to do about this. I would like some feedback from you guys to see if there's anything I could do resolve this. P.s: This issue does not affect all of his tapes. His recently-recorded tapes (those from his PVR) are fine. It's the old ones that give me problems. -- merged -- Here is a list of things I have tried to do to resolve this from the beginning to now: • Uninstalling and re-installing USB capture card software • Tried different VCRs, cables, computers, USB capture cards • Played with VCR settings • Fast-forwarding the tape and then rewinding to the beginning before recording • Tried different capture software (OBS, VLC, NCH Debut, icuVCR, MovAVI Video Editor 15) • Plugging in USB capture card in different ports and unplugging all other USB devices • Installed Windows 7 alongside 10 to see if it was a Windows issue • Messed around with Windows codecs and installed new codecs, figuring it was a software issue • Capturing footage using Linux (based on Ubuntu) • Tried two DVD recorders because they may have an internal TBC and/or frame synchronizer • Tried a Videonics MX-1 Digital Video Mixer. Didn't work. Output displays odd color scan lines on two different brand-name VCRs. |
Some tapes are in such poor condition or are recorded in such a way that you can't get a clean frame-level signal. The result is often false Macrovision effects. While it's true that some DVD recorders can be used as a tbc-pass-thru, their tbc's are obliged to honor Macrovision or be taken to court. My remedy was to use a frame-level tbc to clean up the signal for a clean capture.
Why it takes exactly the time span you mention on a tape to reach a clean area is a complete mystery. True, Macrovision errors don't show up on TV, but it's not because tv's have tbc's -- it's because they just don't respond to Macrovision errors when tapes are played as you're playing them. There's a lot of specific info you haven't mentioned. "Different codecs?" Which codecs? "Different capture cards"? Which capture cards? There are devices and methods that we recommend, and there are devices and methods that we recommend should be avoided. We would not recommend the JVC player you're using. The Toshiba pass-thru you mention is barely adequate with a very weak tbc (I used to have that model and tried using it as a line tbc). I realize that some of the gear needed to properly undertake such a project is likely to be expensive and/or difficult to find. This forum is often a frustrating experience for me, because I work soley along lines recommended in our guides and never had problems, so when "different" hardware or software causes issues I'd like to have the answer but don't have one. Meanwhile I, too, have difficulty keeping up certain equipment, especially high-end VCRs, but that's what is required and so that's what I use. Anyway, much of your problem appears to be false copy protection response. It wouldn't hurt to have a better player, although in my case I was getting copy protection errors from a couple of tapes that were recorded from a corrupt cable tv signal -- and I used high-end players to play them. The only fix was an external frame-level tbc. They are often on sale here in our marketplace forum. |
On the dim part of the video, you can see some heavily flashing colour on the right side of the image. It could be that this is related to the dimming and color loss somehow. I know both VCRs and A/D chips will typically have a "color killer" circuit that will mute the colour signal if it thinks it can't decode the color properly which could be causing the color to disappear if the color is too bad. I suppose whatever is causing the colour flashing may also be causing some issue with the gain control of the DVD recorder, or making it think there is a macrovision signal. Some DVD recorders will recreate a macrovision signal if they think they detect it, while others will simply mess up the image. They will also refuse to record so maybe you can see what they say when trying that. A/D chips vary in how they deal with bad signals, so I suppose the one in the TV is handling it better in this case.
I suppose this may be one case where you really need a powerhouse TBC to get a proper image. I do spot some white strips at the very top of the image, which suggests there is something in the blanking area which isn't normally visible. I believe the video mixer you mentioned is capable of stripping much of the macrovision signal and stabilize the signal somewhat, don't know if the one you got is faulty somehow, or if it too is reacting badly to the tapes. Does it cause the mentioned issue on all tapes? Also, what USB capture card have you got? |
Quote:
In general TBC's will strip out most of the Macrovision signal but may leave some artifacts behind. Macrovision works in part by messing with (fooling) the AGC in VCRs to make the image alternately too bright and too dim, and that can result in messing with AGC built into some capture devices. Some cable TV set top boxes were designed to put Macrovision on their analog video outputs if the broadcasts were so encoded. Early pay-for channels/programming on some cable TV used similar protection schemes. How was the MX-1 connected in your setup?The result viewed? Not sure I understand your description of its output. |
Yes, you're seeing timing/timebase errors. TBC needed.
But not just any TBC. It's a loose term, can mean many things. Line will tackle the image issues, but it needs to be a line known to resolve those problems. Not just any line, or something claiming to be line. TVs don't have TBCs. pre-HDTVs just ignore certain errors, as well as help the signal generate what's needed for good display. Not just VHS, but any signal. JVC-XVC29 is just a consumer VHS VCR combo, no line TBC. The Toshiba DVD recorders don't have line TBC. At least nothing powerful, and TBC(ish) at best. But as you see, it doesn't do much. The Panasonic ES10/15 is a minimal TBC(ish), will probably remove most of the errors you see. Read up past post on it, so you know what it is, what it does. Brightness up/down is either anti-copy or false anti-copy. For that, you need framesync TBC, in additiona to line TBC. Attach sample clips to forums. Which capture card? Some are excellent, many are junk. Only use VirtualDub. Never VLC, NCH, MovAVI. OBS and iuVCR only if VirtualDub refuses. What codec? Use Huffyuv. I know some have "good" (unqualified) results with the MX-1, but others have not, and I've never been impressed. It's not a TBC, it's a video mixer that claims to have TBC (and again, claims can be BS or fudged). |
Quote:
The MX-1 was designed to meet Hi-8/S-VHS signal bandwidth and the newer MX-Pro to meet the DV signals bandwidth. It could be controlled by the Videonics AB-1 Edit Controller in a linear editing setup. It is worth trying if you have one, or access to one, but may not be equal to the recommended dedicated TBCs. If memory serves, the MX-1 needed clear, unobstructed air flow around to avoid overheating effects during long sessions. |
Quote:
Quote:
Quote:
Quote:
Quote:
• DigitNow Master AVcap BR117 (horrible one, don't recommend. Also horrible customer service) • Diamond Capture VC500 Quote:
I'll try to keep an eye out if one becomes available for cheap. Quote:
Quote:
Quote:
Quote:
I will have to look into the marketplace side of the forum to learn more about obtaining an external frame-level TBC. Quote:
Quote:
Quote:
Quote:
• Headphone Volume • Input/Output Frame Rate Lock Disable • Noise Filter • Black Level • Composite Chroma AGC The two that interest me the most are Noise Filter and Composite Chroma AGC. The noise filter setting can be adjusted from 0 to 3. I've tried playing around with the VCR, turning off Video Stabilization, playing around with different picture modes and different levels of the noise filter. None of my attempts so far have fully removed those weird-looking scan lines. The other problem I have is that different portions of the tape respond better than others. There's one particular fight in the tape that has virtually no scan lines. The image is stellar. But carry on to the next recorded footage (recorded at a later date at the end of the last fight) and the scan lines come back. The beginning of the tape is particularly bad. Moving on to Chroma AGC, it can be only set from 0 or 1 (off/on accordingly). The AGC controller doesn't need to be on because the color accuracy is already pretty good, but I leave it on anyways because it appears to enhance the color better and brightens it up a little. Quote:
Quote:
Quote:
Quote:
Quote:
Quote:
Quote:
Quote:
Quote:
Quote:
Quote:
Quote:
Quote:
Quote:
Quote:
Quote:
• DigitNow Master AVcap BR117 (horrible one, don't recommend. Also horrible customer service) • Diamond Capture VC500 Quote:
Quote:
Quote:
|
Quote:
It seems most usb dongles have been rather cheaply put together, and with a minimum effort put into driver development as well. It also seems most the higher-end chips have only ended up in TVs, DVD-recorders etc. There seem to be some differences in how they handle lack of video signal lock, TVs, DVDrs, etc seem to often simply blurt out something no matter what, while the USB dongles seem to prefer to freak out and not output anything instead. That said it seems to vary a bit between TVs too, what model of TV do you have? Quote:
The VC500 has a good quality image as well, but it's not very good at recording directly from a VCR. Quote:
Quote:
Quote:
As for the DVD-recorder, some have adjustable brightness and AGC settings, don't know about the models you used though. Quote:
|
Quote:
The sample MX-1 output appears to have consistent 4% delay in the start of random individual scan lines. Something is fooling the MX-1 as to individual line sync. Varying white in the first 21 or so lines above the image (vertical blanking interval) could be vertical interval time code and/or closed captioning. Macrovision may appear as super white and black on some scan lines as well. |
Quote:
Quote:
Quote:
Quote:
Quote:
Quote:
Quote:
Quote:
Quote:
Quote:
Quote:
|
1 Attachment(s)
For some nice examples of the Macrovision protection signal:
https://www.youtube.com/watch?v=-VqsU1VK3mU This graphic shows the vertical interval time code signal (when it is present, not usually on VHS recordings) in the blanking interval in the overscan area above the color bars. Typically around scan lines 10-20. Closed captioning would be similar but on different scan lines (typically 20-21) and changing as the caption changes. |
The success with the TV sounds like its holding you back from trying some things.
The first is it appears you have basic tracking errors, which might be the condition of the tape, or the brand of VCR may be different from the brand of the VCR that made the tape. Even if a single tape was used in multiple VCRs, try to do some detective work and find out what brand or model was in use.. or not in use. -- the reason tracking errors appear apparent is that JVC stabilization is mostly automatic tracking correction based on a few samples at the start and periodically as the tape plays.. even if you turn On Screen Display off (OSD) if Stablization is left on it will periodically reset the tracking and that may explain the AGC varying after a period of time. Picking a "great" VCR is almost never as good as playback on the original model, or a like built model. -- after you nail down the original model, then look at adding in a frame synchronizer (external TBC).. you won't be able to really find an external line TBC.. well you sort of can, maybe in the ES10.. but the correction for line errors is best done by the VCR itself. So after trying a different similar brand/model VCR, turn off the Autotracking/Stablization whatever its called.. its fixing things.. but its also resetting the AGC in your capture device.. the external TBC should handle the flickering.. so you shouldn't need the autotracking (but if you do, you can use manual tracking to "steer" it into at least a consistent playback Gain mode.. so dimming variability should go away) Moving from (same VCR as recorded on) > (external ES10 "pseudo line TBC") > (external frame TBC "frame synchronizer") > capture device I have the "Startech SVID2USB23" its an EMPIA based system, chipset spec is on the startech website. They always document the chipset which is one of the nice things about Startech. The driver set is also very complete.. but AGC is not "defeatable" you cannot turn it off. So its nice when you don't have any chance of false or positive Macrovision.. but its obviously the wrong device for this job.. since you've proven you do have AGC problems. Early capture devices like the ATI All-In-Wonders gave you control over AGC, or did not have the feature. It was added later to many capture chipsets, or set permanently [On] so you could not turn it off.. partially I think because Macrovision depends on it to screw up the video. If you have AGC feature, and have removed User control.. you are Macrovision complaint and "defacto Enforcing Macrovision detection" in your product. so uncontrolled AGC control was preferred by manufactureres of chipsets and is the problem for end users. For perfect signal without macrovision protection, the end user is legal to copy, and the signal should be perfectly copyable.. but the world is not perfect.. and AGC tends to run crazy and overreacting to highlights and lowlights even in normal signals.. so its not a feature.. its a plague. There are not a lot of sub $100 capture devices that give you back control of AGC, if you have AGC. There are a few.. and they skirt the issue of Macrovision compliance usually by putting in their documentation "Copying Copyrighted material is a Crime.." and move on. A few early DVR recorders were famous or infamous for either lack of AGC or deliberately switching it off.. iLO, LiteOn 5045, Polaroid 2001g, Matrox O2.. and of course ATI Theater 200 chipset based devices. For somewhat obvious reasons these are still highly sought after and trade very little on the used market. ATI after the 200 chipsets came out with the 550 chipset with embedded undefeatable AGC.. they were 100 percent Macrovision compliant with and without the ATI software.. and are terrible choices for capturing and digitizing dodgy VHS tape. Literally for any poor VHS tapes that had signal problems with Gain -- AGC in the capture device is going to defeat your efforts.. you have to look backwards at either a Macrovision "reconditioner, stripper or defeater" or find a capture device where you have control over the AGC ($$$) or it never had an AGC ($ but semi hard to find). In theory an external TBC will "strip" the blanking interval where the Macrovision, or false macrovision is tricking the AGC.. but dodgy tapes will "leak" lines back and forth across the edge of the viewable field.. which can still get caught and detected as a Macrovision signal.. so again.. finding a method that takes AGC out of the equation is the best most stable bet. Ideally.. I'd say probably an ATI 9200 or 9600 All in Wonder capture card.. with the complete cable set (very important) and on an AGP motherboard running XP SP2 is your absolute best shot. Sometimes "complete capture systems" like this turn up on the Marketplace on this forum. And from credible members too.. which tends to be a better shot than assembling random parts from eBay. -- Its an investment that you should think about like a Power Drill.. or Tool.. not for general computing use.. its only for capturing, its not something to be justified as another "home computer".. but likely to be flipped and turned over to another user, after the job is done. You might also consider taking (one) Tape and sending it off to be "diagnosed" by a professional who could recommend what you need to treat a larger set of tapes.. I'm not sure there are a lot of places you can still get this done.. but its a thought. Regarding the ATI 600 USB, an excellent Texas Instruments chipset, with tons of documentation, and a different take on AGC.. but its still AGC beholding. It still does a great job with fairly good signals. Regarding the VC500, there are fans, and there are detractors.. it has one of the most up to date drivers sets.. meaning you can use it on Windows 7 or 10.. but don't let the OS be your driving force when selecting a tool for the job.. its a member of (we have surrendered to 'mighty' will of the all powerful AGC).. its yet another variant however.. and you can try it. Regarding USB. ATI made (one) and only (one) USB Theater 200 based capture device.. the "ATI USB 2.0 N" its not that hard to find.. one goes for sale about once a month. (But) it only works under Windows XP SP2 and it had a special installation sequence.. done wrong it will not be totally functional. For real experts its an awesome capture device for a steal in price.. but no one has really written a good guide to properly installing it that I know of.. after its installed.. it does an amazing job of capture (for video "only) you have to capture audio through a seperate sound card or sound input.. or you risk unreported dropped frames. Virtual Dub supports it superbly and its one of the most complete USB capture solutions I know.. but it runs very "Hot" and you have to keep it cool. Otherwise it does need a good signal, you still do need line and frame TBCs.. but it is immune to most AGC problems.. but it also does have software Macrovision detection.. its a complex bag of tricks. |
Quote:
Quote:
Quote:
|
I'm stepping out in a bit. But when I get back, I'm going to try something else. I can't remember if I tried this already. I'm thinking of running the JVC VHS to the Videonics via composite (it's the only way). Then run the output of the MX-1 to the DVD recorder via S-VIDEO and then S-VIDEO out from the recorder to the computer.
My theory is that the video mixer will resolve most of the issues and then the DVD recorder can clean up the rest of the image 🤞 |
Alright, I'm back. So yesterday, I tried using two middle-man devices instead of only one. So here's what I did. I outputted the JVC VHS to the MX-1 using composite (because it was the only way). Then I ran an output from the MX-1 to the DVD recorder via S-VIDEO. From the DVD recorder, I ran an S-VIDEO output to the USB capture card.
My theory was that these two devices would each play a role in cleaning up the image and since the image is mostly clean by the time it gets to the recorder, I figured it would not have a hard time to fix it up. I also ran an output from the VCR to the DVD recorder via coaxial cable. I decided to record a short video demonstration so that you guys could see the differences in real time between the output of the DVD recorder and when the MX-1 is involved. The first demonstration is with the VCR running directly to the DVD recorder. The second demonstration is with the VCR running through the MX-1 and then going to the recorder. I made sure Video Stabilization was disabled, as I've heard that the feature can mess up the MX-1's effort to clean the image — which I can vouch for because I've seen it happen. I also made sure Picture Mode was set to Edit so that the VCR's AGC wasn't trying to mess with the picture in any way. Unfortunately, my plan didn't work out. It still shows the scan lines in the DVD recorder output both via composite and S-VIDEO. Even when I applied the noise filter in the MX-1 settings, that didn't fix it. I may try fiddling around with it again to see if there's a sort of Goldilocks set of parameters that makes the image cleaner. I'm even willing to compromise if there is some scan lines here & there. |
I general, any line time base error becomes locked in when you run the signal though a frame TBC. That is because the TBC generates new sync for each line it detects. The MX-1 assumes the start of the line based on what it detects as the scan line sync pulse. If it is poorly formed, or false pulse material is present it can cause the tearing you observe. In general the line TBC should come first, preferably in the VCR.
|
1 Attachment(s)
Quote:
For some reason, when I click on the “Color Format/Compression” menu, I see duplicate “YUY2” entries. Each one seems to offer different amounts of resolutions. Some allow me to go up to 640x480, some as high as 720x480. Also, for some reason, when I used iuVCR, the footage was a little bit cleaner. At the beginning, I wasn't seeing as much scan lines as before. I don't know what to make of that, but I'm going to look into it more tonight. Quote:
I just thought of something though. I'm thinking of running two inputs to the MX-1. One input straight from the VCR and the other from the DVD recorder (pass-thru). My theory is that I can blend the two sources and perhaps the image will be somewhat better. |
In some cases video AGC effect is be based on the sync pulse height, not the image content brightness. If the video signal has proper sync, as the output of a TBC should be, then AGC would have no little to no effect in image brightness.
|
Quote:
|
In summary, if the portions of the signal that are defined by standards; e.g., sync pulse width, height, color burst, pedestal, max brightness, etc. are correct how can a machine tell whether or not the garbage in the image is not intended. Common image display flaws such as snow, picture roll, dark image, etc. are occasionally used as special effects.
|
Site design, images and content © 2002-2024 The Digital FAQ, www.digitalFAQ.com
Forum Software by vBulletin · Copyright © 2024 Jelsoft Enterprises Ltd.