-
what’s in s-video output of dv cams?
hi folks,
forgive me if this is not really a decklink question, i also cross post this question in the sony dv forum.
it is also a quite technical one, i think.
anyway. most of nowadays dv-cams have also a s-video output. my question is, at what time point, in the whole process of capturing, processing and storing a video signal, this analog signal is generated?
background:
i do a lot of compositing stuff which involves chroma keying, of course. trying to do the best with what is given to me, that involves keying dv, also. but dv is really not easy to key, simply because it has 4:2:0 color suppression (being in pal-land) and additional intraframe compression. the high color compression contradicts, of course, a good key.
now, if i’d use the s-video output of the camera at shooting time and use a modern capture card (with 4:2:2 or 4:4:4) with s-video input, do i get a “better” image?
if the analog s-video signal is produced after the 4:2:0 color suppression, i may be able to capture in 4:2:2, but that wouldn’t mean i have more color info in my captured image.
hence, the above question.thank you very much.
cheers,
-rayk -
7 Replies
-
It would be unlikely you would get better image quality by capturing with a card with an S-video input. The video is decompressed off of the tape and then converted to analog.
Usually the lower end S-video capture cards are consumer oriented and don’t support uncompressed video capture. Decklink cards do not support capture from S-video.If you want improved quality from DV (especially with PAL, which is 4:2:0), you can capture via SDI from a DVCAM deck that has SDI outputs to uncompressed 4:2:2 with a Decklink card. This will help you with compositing and chroma key.
-
“If you want improved quality from DV (especially with PAL, which is 4:2:0), you can capture via SDI from a DVCAM deck that has SDI outputs to uncompressed 4:2:2 with a Decklink card. This will help you with compositing and chroma key. ”
But, to clarify, you would have to do that while shooting. Once the signal is on DV tape, it’s 4:1:1 (or 4:2:0), and that’s all you have. You don’t get to magically re-create those missing bits.
-jeff
-
Hi Guys..
This was something that I also wondered..and i ran some test for my own amusments. Useing a miranda PICO link “svid-sdi” converter, I shot and some blue and green screen “xl-1s” loaded firewire and also through the pico link into a deck-link card. from there it was off to combustion. What I found was the bkg. was easyer to remove from the sdi version but at the cost of fine detail.. ie hair. while agreeing that there is no magic way to replace missing data. I now always use the miranda for dv transfer.. since most of the footage I use is ntsc D-beta.
also
the other Ray K “original poster” was first to the name, so I guess that makes me the one with the missing bits…
hope this helped
regards
ray -
Forgive me if I sound rude, but you’ve wasted a lot of time and money. Capture the DV over Firewire straight from the camera. Then convert it to Uncompressed or whatever helps you out with your keying. There is absolutely not one single non-artistic reason to convert the signal to analog. And you’re doing this for technical reasons, not artistic.
That Miranda converter, well, I don’t know the price of that thing, but I’m sure it’s a lot. I know their “DVI Ramp” was $10,000 US a year ago (totally outrageous). Anyway, you don’t even need it. It’s totally useless for what you’re trying to accomplish. You don’t need a converter, a card, or any hardware whatsoever. If what you need is to DV footage to some 4:2:2 or 4:4:4 codec (even though you never gain anything back), you can download many free programs for Windows that will do that very same thing, or buy Quicktime Pro for $30 if you use a Mac. You accomplish the same exact thing. The “side-effects” to this are that you don’t massively degrade the quality by converting to analog and back to digital, and you save a bundle.
If this extra step of converting it with software presents a major workflow problem (and based on your QC methods, I’m sure your not in an environment where it would), you want a DV to SDI converter. For professional work, DV and S-Video don’t belong together in the same workflow, or even in the same sentence.
Again, sorry for sounding rude, but your fundemental understanding is rather weak- which is totally fine. We can’t all be nerds. But I do get a bit annoyed when people just blow money on things that they don’t need whatsoever- especially when it’s something that hurts their work, not something that improves it.
-
Hi Neil,
[Sean ONeil] “Again, sorry for sounding rude, but your fundemental understanding is rather weak- which is totally fine. We can’t all be nerds.”
frankly, I feel neither offended by your post nor do I think it was rude.
There are many kinds of nerds out there and I seems to belong just to another tribe 😉
Well, I admit that my understanding of video tech stuff is rather weak, but hey, that’s why I ask questions -before I buy! So no money wasting on my side (yet).
I’d really appreachiate, if you would take the time and straighten the things out for me. Maybe, I didn’t used the right phrasing in my first post. So here’s the point, I don’t seems to cut (forgive me when I talk in laymen terms…):
I have more or less three choices when I want to recocrd video vom my DV-Cam.
a) I record to DV-tape, use a DV-deck later and use firewire to get it into my computer
b) I use the DV-out and record via firewire direct to my computer
c) I use S-video out and use a video A/D card to store the video in some codec in my computercase a) and b) shouldn’t make a difference. In my understanding DV means 4:2:0 color compression (being in PAL land) and 5:1 intraframe compression.
I know, I can convert my DV-codec to any other codec, I like. So I may choose to convert it into an uncompressed 4:2:2 (or even 4:4:4) codec. But that doesn’t bring me back the information which got lost by the DV-codec in first place.
S-video is an analog form of delivering a video. If I need to store it in my computer, I need to digitize it. Decklink offers such cards,afaik. I could capture and digitize that anaolg video stream into an uncompressed 4:2:2 codec for further proceeding.
If the S-video signal was just a conversion of the DV signal in the camera, I could capture 4:2:2 uncompressed as much as I like, I wouldn’t get more information as in the DV. But, if the S-video signal was independed (or created before the DV conversion), I actually had more information compaired to DV.
Or, what’s wrong with this reasoning?Thank you very much!
-rayk
-
[Rayk Hemmerling] “But, if the S-video signal was independed (or created before the DV conversion), I actually had more information compaired to DV. “
I see. Ok, I get what you’re asking. You’d have to check with the manufacturer to be sure, but I’d say that it’s highly doubtful. Think of it this way. When you play back a DV tape, you get S-Video out. That means that the camera contains the ability to decompress the DV stream and convert it to analog. Since it contains that, there’s no reason why the S-Video signal would be created any differently when the camera is recording then when it’s playing back. Make sense?
I wish I could help more but chroma-keying isn’t something I do very much. What I can say is converting DV to Uncompressed digitally is going to look a lot better than S-Video- even if your camera produces the signal independently (which I doubt it does).
BTW, no Blackmagic Design products use S-Video, so this topic doesn’t even belong here. BMD products use SDI, HD-SDI, Composite, and Component.
-
[Sean ONeil] “BTW, no Blackmagic Design products use S-Video, so this topic doesn’t even belong here. BMD products use SDI, HD-SDI, Composite, and Component. “
oops, i had thought there was something.
anyway, thank you for takeing the time!cheers!
-rayk
Reply to this Discussion! Login or Sign Up