or ... http://tinyurl.com/yjb7ytk
Please read some of the reviews! They are quite good!
I have a pile of old video cards, some have rca connections on the back, maybe I already have such a thing.
What am I looking for exactly?
You need a 'tuner' card that can receive rca signals. Most vid cards only produce (output) the signals and not receive even though they have rca jacks on them.
In other words: Vid Card = Out, Tuner Card = In
You can buy a USB device that will do what you need. I have one at home, and used it at work to record some video we have to a DVD without issue (well, except that I had to record time-code to an audio channel, as it didn't embed the timecode in the video stream like I wanted it to... but that just means it filters out certain scan-lines... for your needs, I'm sure it would work fine).
I can't recall the name of it, so I'd have to take a look at it when I get home.
Or, alternatively, I suppose I could record the VHS tapes for you.
No problem... hope it works well for you!
Jan 7 2010 4:03pm from Spell Binder @uncnsrdNTSC VHS tapes run at 29.97 frames per second (actually 30000/1001), with 480
What's the quality of those USB video capture devices? My concern is
whether USB has enough bandwidth to carry a good quality video feed. I
know USB v2.0 compliant devices can do 400 Mbps, but what does that
translate to in terms of video frame size, frame rate, color depth,
lines of 720 pixels each. When captured as digital video, it's normally 24bits
So, some simple math, it's framerate * frameresolution * bitdepth to get the
uncompressed bitrate. For NTSC video, that is (30000/1001) * (720*480) * 24,
which is 237Mbps. That is well under the rate for a USB 2 device, so it should
work. I don't think it'd work for high-definition (I'd use Firewire for that)
but for standard definition and lower it should
This is all theoretical of course, a USB video digitizer could be terrible
quality or it could be prestine. The digitizer may stream to the computer raw
video to be compressed by the computer, or it could compress it using DV or some
flavor of MPEG, reducing quality.
Is there anything that hasn't been turned into a small cheap usb
device that you can get for $2.00 from china?
Not quite sure. How about I jump in my $2 USB car and drive to your house, you can start up the $2 USB coffee pot, and we'll chat about it?
Don't feel so bad -- until a couple of weeks ago I still thought webcams were expensive.
The USB video capture device I used captured the video at a decent enough quality that we used the results for a tradeshow. As long as the source video is of decent quality, it can capture the video at very good quality.
That's not fantastic, but it's a lot better than I though it was, considering how crappy video is on a television.
My perception comes from using opera on the Wii. You get maybe half the width of the page and have to scroll around a lot, and often have to zoom in to be able to read anything.
Unless the pixels are not square ... but if the target device is a computer, the pixels *are* square.
Unless the pixels are not square ... but if the target device is aYeah, NTSC pixels aren't square. They're slightly elongated
computer, the pixels *are* square.
vertically. On a computer they are displayed at 640x480 to give it a
4:3 ratio, that and they have to be converted from YUV colorspace to
RGM. Analog TVs just handle it naturally.
It's so much easier to think about aspect ratios with high-def
formats. 720p is 16:9 with square pixels, so 1280x720, no
shenanigans. Same with 1080.
One of the most interesting resolutions I've seen is the NTSC Super
Video CD: it's 480x480 then stretched to a 720x480 NTSC signal by the
player, and then the TV displays that at a 4:3 ratio. That's just odd
/me wants a 1080 television
NTSC, technically speaking, doesn't have discrete pixels in the horizontal dimension. There is just a continuous analog signal that is scanned across the line. Of course it has discrete scanlines in the vertical dimension. Due to the subcarrier frequencies involved, the Nyquist-Shannon Sampling Theorem tells us that the horizontal dimension contains approximately 704 non-square pixels worth of information (or 720 if you count the blanking interval.)
At least, if you're talking standard definition.
High definition is a different story.