JVC Video Stabilization can solve this problem?
5 Attachment(s)
My project is to digitize a bunch of old tapes that were recorded on a cheap VCR, likely many using some kind of long play mode. Here's my workflow:
JVC HR-J692U consumer VCR --> Datavideo TBC-1000 --> ATI TV Wonder 600 USB --> Win7 PC w/VirtualDub 1.9.11 & ATI drivers from this forum. I have a tape that where the pictures seems to be scrolling vertically very fast, like the VCR's not able to find the where the frame actually begins, and just starts at some random place in the middle. However, the picture appears fine in fast forward/reverse. When I turned on the "video stabilization" feature of the VCR, the problem seemed to go away. See the attached images. The "badframe" images show the problem (with a few adjacent frames), "video_stablizer_on.png" show the video stabilizer fixing it, and " HT-J692U-menu.png" show the exact settings. All the images were with the TBC-1000 in the chain. I'm under the vague impression that the video stabilization features of consumer VCRs like mine are junk, but it that always the case? Is there a better way to fix this problem? I think the fixed capture has some "flagging," but honestly that's acceptable to me at that level for these tapes. -------------------- In a related note. I found this forum post by someone who seemed pretty down on TBCs and said something interesting: https://forum.videohelp.com/threads/...BC#post2473579 Quote:
|
It looks like the top of the video has the typical "flagging" issue which is best fixed by panasonic dvd-recorders. The cypress and datavideo TBCs don't usually manage to correct for that. The video stabilizer creates a new vertical sync signal based on the head switch point (it works the same in all JVCs I think, they shared a lot between the hi-fi and SVHS decks from the same lineup. It does sort of fix the issue, but you are still left with the bent image with jitter on the top.
If you have access to those dvd-recorders I would give them a try. The toshiba is one of the late-model funai dvd/vcr recorder combos I think, those have panasonic chips that can stabilize a fair bit, though they are known to have issues with flicker. The EZ47V I'm not sure about as it's from what I can see one of the few Panasonic models that used a different system chipset (LSI, later bought by Magnum) rather than Panasonics ICs like in e.g the ES10 but it doesn't hurt to give it a try before looking for an ES10 or similar. |
What is the correct term for the issue shown in the "badframe" images? I think I'm going to just capture this with my current setup, but make some notes so I can come back and potentially re-do the capture later.
|
I'm trying the full capture of this tape and I'm getting a very large number of inserted frames. After 55 minutes, I have 14,700 inserted out of a total of 86,000, which is 17%. Is this caused by the combination of the JVC image stabilization and the external Datavideo TBC? I think I read some comment somewhere that they were "incompatible" but it didn't explain why.
|
JVC's S-VHS VCR stabilizer is not incompatible with a TBC-1000. So that's not your problem here.
|
Quote:
What would cause inserted frames with an external TBC? Is it a bad tape, bad tape/vcr combo, or do I need more processing in the chain (e.g. a Panasonic ES-10 type passthrough device)? |
The function is the same on the non-SVHS ones, they use the same hardware for those parts.
If you are getting a lot of reported inserted frames with a tbc in the chain it's probably some issue on the computer side (unless there is something faulty with the tbc). The TBC can decide to drop frames on the input side (though it's very rare unless you are feeding an NTSC signal to a PAL TBC or something), but it should always be outputting a stable 59.94 (or 50 for PAL) fields per second which shouldn't cause any issues for the capture card, that's it's purpose after all. If the TBC has issues handling something what you get is baked in errors instead. If you are not using a SVHS deck with TBC, using an ES10 or similar will normally look nicer though, especially a tape with tearing/flagging/whateveryoucallit on the top like this one (other than on tapes with copy protection), as they stabilize horizontal jitter much more than the datavideo is capable of. |
Quote:
After rebooting, I ran three different tests using a different tape (a 30 min commercial one from the late 80s), and I saw it start with very low numbers of inserted frames then increase to very high numbers at different points in the capture (once about 15 min in, another time about 25 minutes in). I tried again using the OS disk (a mSATA SSD), and I didn't get any inserted frames (though it's quite small, only just large enough to hold a 30 min capture of this tape). I haven't had a single dropped frame reported. I'm new at this, but my assumption is that inserted frames would be a problem getting data from the capture card, not writing to disk. Is that correct? Could the HD activity over USB eventually start interfering with the capture card activity, increasing latency or something and forcing VirtualDub to insert frames? This PC has a SATA port, but no space to actually mount anything that uses it, so I'm thinking of jankily connecting an internal HD externally. Is it so bad to capture to the OS disk if it's an SSD? It would be a lot nicer if I could just get a bigger mSATA SSD. |
That artifact is capture card /Driver related, You wouldn't get that by connecting directly to the TV. To rule out the TBC1000 just take it out of the workflow and capture the same scene again and see if the artifact is still there, pretty simple.
It's a common mistake that people troublshoot VCR and TBC problems with capture cards, First you would hookup to a TV first to have a general idea on how the picture looks like then procede to capturing. That's why proffesional workflows have monitors connected to their VCR's analog output and the other output goes to the workflow. |
Quote:
With modern Hard Disks and OS, the times when I had to format the second (capture) HDD on my Windows XP Desktop before each capture (even for small bitrate dvb-s) and tweak tenth of settings to improve data transfer are long gone ;-) |
Quote:
Are you thinking my TBC isn't working because it should have corrected this? I can try without it but I think it's necessary because in my family's home movie capture project 5+ years ago, we had terrible audio sync problems without it, and many (most?) of these tapes are worse than those. I do have a TV in this workflow, but it's connected to the TBC because I only have one set of outputs on the VCR (and the TBC has four). I did see the badfame artifacts on the TV, but I don't actually see the flagging after I turned on video stabilization (due to the overscan I think). |
I didn't say capture your tapes without TBC, I said try. Trying is a part of the troublshooting procedure if you have never done one. Try all combination and report back to get a better help, Speculation is not going to get you anywhere.
|
I think the issue was mainly with using the USB storage. Once I started capturing to a SATA SSD, I've had much lower inserted frame counts. They haven't always been zero (they range anywhere from 0 for a 6 hour capture to 120 during a two hour capture), but it seems like leaving USB drives disconnected, rebooting, then first thing starting the capture can almost totally eliminates the inserted frames.
The write speed of the (new) USB drive I was using has fallen off a cliff. Must be one of those SMR drives. |
Quote:
|
Quote:
I'm guessing it's the capacitors, so maybe I'll try to replace them once I've made at least one pass through my tapes. |
Try replacing the ac adapter (power supply) before doing surgery on the caps. My 1000 was acting flaky too and I was reading another post on here and LS recommended the ac adapter from an external hard drive. Apparently these TBCs eat them up.
|
I did replace the adapter with one from TGrant Photo, but it didn't seem to fix any issues (though I didn't expect it to, I was mainly hoping to keep things from getting worse).
|
Quote:
Quote:
Quote:
Quote:
Quote:
Quote:
Quote:
Quote:
Quote:
Quote:
Quote:
Quote:
For the record, a good APD (Asian Power Devices) 12V 2A is often what WD/Seagate included with HDDs. There's several brand name PSUs that were used, all 12V 2A center-positive. Phihong, Ktec, few others, but APD is my first choice. When I sell TBC-1000, if the PSU is dodgy whatsoever, I recycled it, replace with new APD. A $2k item isn't worth risking on a $10 PSU. Quote:
|
3 Attachment(s)
Here are some examples of the glitches I'm seeing.
P3-06a three stooges-glitches.mkv (darkening, flicker, and glitches: pay particular attention at about 9s and 30s) J1-11a TRANSFORMERS Pilot-glitches.mkv (a little flicker and some glitches: pay particular attention at about 4s and 14s) J2-05a Brazil-glitches2.mkv (a little flicker and many glitches) |
looks like Macrovision to me. Also a line TBC is needed here, you got flagging on your samples. I think you will be better off with an ES10/15.
|
Site design, images and content © 2002-2024 The Digital FAQ, www.digitalFAQ.com
Forum Software by vBulletin · Copyright © 2024 Jelsoft Enterprises Ltd.