capture uncompressed from dv tape?

cordvision

cordvision

Well-known member
I had a little argument with a friend about capturing with a capture card (blackmagic or AJA etc.) from a miniDV tape (SD) to uncompressed using Component or SDI. He claims that this gives him better results than if he transfers the video over FW and transcodes it on the computer to uncompressed latter. I don't think so unless you use a bad program for the transcoding that does a worse job than the capture card. In my opinion he doesn't gain anything because the footage on the tape is already heavily compressed, so he could just transfer it over FW to his computer and then transcode it to uncompressed for color correction etc. I don't think that I am missing something or do I really?
(If he captures live, while shooting, it is different of course. I am talking about capturing from a miniDV tape to uncompressed).
 
Capturing via Component or SDI as uncompressed can result in a better looking image than transcoding. Transcoding might include a lot of factors that make the image looks worse. BUT, you cannot improve the image you have already...correct. The damage has been done when you shot DV. HOWEVER, color manipulation and green screen compositing results in a much cleaner look if the footage is uncompressed as opposed to DV.
 
What do you mean by "it can result in a better looking image than transcoding"? If the transcoding is done with a cheap program it probably does but if it is done with a decent program there shouldn't be any difference, it might be even better. A blackmagic card does a pretty good job in transcoding the image but a computer can do it as good as a blackmagic card even though probably not in real time. Do you agree? My point is that it is a waste of money to use a blackmagic card (unless you don't have the time to transcode your footage with your computer).
 
Graeme Nattress has a very good, very in depth, article on chroma sampling at KenStone.net that shows the benefits of capturing DV via SDI as opposed to FW. He is working w/FCP and Apple's DV codec so I don't know how this applies, or doesn't apply, to other NLEs and their codecs. Basically, capturing via SDI results in less blocky chroma channels so a better looking image (especially if you are doing any keying or color correction). How much better in terms of 'real world' usage depends on how critical you are of your footage and how much you plan to manipulate it. For a student film probably not a huge deal. For a paying gig where you need every last pixel it can be the difference between pulling a nice looking key and a key that is 'meh.'

Long story short, capturing via SDI is best, using one of Graeme's chroma smoothing filters is next best, and capturing via FW is low man on the totem pole.


-A
 
thanks for the article. That was exactly what I was talking about. Quote: "Because we have reference image, it is impossible for us to tell which is better, SDI or Firewire, but we can tell that the differences between them are subtle indeed. There seems to be a subtle, random difference between the two Y' components, which, due to it's very low level, is hard to determine the origin of. The chroma components are certainly different, especially on the visible edges. Again, the differences are at such a low level that they are practically inconsequential."

The images are slightly different but not better or worse. This was to be expected because the up-sampling is done slightly different. I would guess it varies from deck to deck as it also does from NLE to NLE. Therefore, capturing trough SDI doesn't help, the footage doesn't get better..
 
cordvision said:
thanks for the article. That was exactly what I was talking about. Quote: "Because we have reference image, it is impossible for us to tell which is better, SDI or Firewire, but we can tell that the differences between them are subtle indeed. There seems to be a subtle, random difference between the two Y' components, which, due to it's very low level, is hard to determine the origin of. The chroma components are certainly different, especially on the visible edges. Again, the differences are at such a low level that they are practically inconsequential."
Click to expand...
Careful about pulling quotes out of context. That quote was talking about a specific example, out of many used, comparing SDI capture with FW capture that had a smoothing filter applied. From the same section of the article, "This time I did not look at the unsmoothed Firewire chroma as we know it's going to look bad. Instead I concentrated on the difference between the smoothed Firewire Chroma and the SDI chroma."

Like I said in my previous post, there *is* a difference but whether or not it's a difference that makes a difference depends on how the footage is being used. If you have more time than money then applying a chroma smoothing filter to FW captured footage would probably be the better way to go. If your time is money then capturing via SDI would probably be the way to go. Either way, untreated footage captured via FW looks poor in comparison.


-A
 
Our discussion was never about untreated footage, we talked only about the final outcome. If you use an other NLE than FCP, you wouldn't have to smooth out the chroma anyways, that is just specific to the FCP QuicktimeDV codec. There is a difference, but that doesn't mean that one is better than the other. Depending on what NLE and SDI combination you use, the outcome might be in favor of the FW-NLE method or the SDI-Deck solution. Looking at it from a technical standpoint, there is no reason why a stand alone computer can't transcode the footage in the same way as a SDI deck...
I absolutely don't dispute that the SDI solution is faster and more convenient. It does the transcoding in real-time which is hard to beat...
 
cordvision said:
What do you mean by "it can result in a better looking image than transcoding"? If the transcoding is done with a cheap program it probably does but if it is done with a decent program there shouldn't be any difference, it might be even better.
Click to expand...

Transcoding = recompressing.

Recompressed media will never be better than the original, because compression always throws away data.
 
You must log in or register to reply here.
Back
Top