Quote:
Originally Posted by lollo2
(Post 87585)
If there is a problem in the tape (i.e. a major timing issues or a drop), that frame will never be captured. If it is marginal a frameTBC may or may not capture it. If it is very marginal a re-capture without a frameTBC can get it.
|
I think this is something I'll just have to live with. With my old VCR, it looked like frame drops mainly happened between different recordings, where they don't matter that much anyway. The audio goes out of sync, but that's relatively easy to deal with compared to if major framedroppage happened in the middle of a clip. I think I encountered only two or three such events in my tapes so far, which means I can just brute-force through them with manual Avisynth magic.
Quote:
Originally Posted by lollo2
(Post 87585)
Although Median plugin has an internal routine trying to align the captures with dropped frames, its goal is to reduce the random noise averaging multiple captures, not to fill the dropped frames, where it may fail.
The operation to rebuild the full frame sequence using multiple captures in a safe way should be done manually frame by frame, sequence by sequence.
|
I'm using the Median plugin (assuming you're talking about Avisynth?) as the part that actually calculates the median, but for finding framedrops and syncing different captures I have written my own helper scripts. It's not fully automatic, but it requires very little manual work. I might have to re-write some of it though as it was quite custom-made for the kind of video my old VCR produced.
Quote:
Originally Posted by lollo2
(Post 87585)
I am not familiar with that card, but component and p60 or p50 are not the native parameters of VHS signal and should be avoided.
|
I meant that since the card can do p50/60 via component, it's not just any cheap garbage you can find for $10. It also (likely?) shares its tuner/ADC between its different inputs (composite, S-video, component), so it's built to handle higher frequencies, so 576i isn't at the limits of what it's designed for. Not sure if this kind of thinking is applicable to all it's doing, but it feels like it shouldn't struggle with ghosting on sharp edges for example.
Quote:
Originally Posted by lollo2
(Post 87585)
There is only one resolution and color format to choose, YUV and 720x480 or 720x576
|
Yeah that's the only format one *should* choose, but this card also allows for various narrower horizontal resolutions, as well as many different YUV formats (UYVY, YUYV, YUY2, YV12 (bad for interlaced, I know) etc.). I have another cheap capture card that doesn't allow anything other than the defaults, so I assume this is coming from the actual card and not just VirtualDub's post processing. And again, the fact that the card actually supports many different formats makes me think it's at least designed better than the cheapest ones.
Quote:
Originally Posted by lollo2
(Post 87585)
Many cards allow that, and for most of them, like yours, it is a basic "digital" processing
|
This I'm actually not sure about, but I was pretty sure that when I tested this on the cheaper card, it made no difference, while on this card it does. Oh well, it's just another feature that I'm likely not going to use. Not sure how I would test whether it's digital post processing or an actual setting for the tuner.
Quote:
Originally Posted by lollo2
(Post 87585)
Not really. With a fully working JVC HR-S9600 you have one of the best player. Be sure to do not use too much the fast forward and fast rewind to do not break the Dynamic Drum mechanical parts.
|
That's reassuring, thanks. So "slow" FF/REW (where you can see the picture go by quickly) is better for the tape? That's good to know!
Quote:
Originally Posted by lollo2
(Post 87585)
If you are satisfied with your capture card ok, otherwise give a try to a IOdata GV-USB2 or a Hauppauge USB-Live 2.
|
How good are USB 2.0 capture cards really? Do they actually stream over 20MB/s of continuous data reliably, or do they resort to compression? Though I'm guessing a good ADC overshadows the losses from high-quality compression.