-
Video Compression vs RAW video data
This is a general question about digital video, posting here because I use Vegas, but am interested in hearing from some hacks/pros about this topic. i have been dipping my toes in the water of video editing and have some hands on experience however, I don’t feel like I have a real handle on understanding the standards of digital video.
From what I understand, please correct me if I’m wrong, virtually all video is compressed to some degree. Using completely uncompressed video is impractical because most computers can not handle that amount of information very well and it seemed to be unnecessary given our current displays/monitors.
However, having the best(least amount of) compression gives you the most flexibility in exporting a final product. (Also Vegas does not seem to be able to edit some types of compression formats, such as divx, some avi, etc.) \
So, getting to my question, it has to do with navigating or getting past the marketing propaganda that tends to promote confusion rather than understanding.
Are there a set of variables that you can hang your hat on(in other words, that all codecs can be described in terms of), such as Frame dimensions, frame rate, and bitrate, to ACCURATELY & OBJECTIVELY describe any type of video/format?
Is there a general ideal amount of compression when digitizing/editing from an analog source(let’s say with DVD being the final product.)? I know that depending on the type of motion & image complexity can vary, but is there a general standard that we can say will definitely will work for all types of video content? I’m guessing everyone with have their own experiential rule o’ thumb.
Are there any specific things to watch for, where information can be misleading? I know this is a HUGE topic, but I am hoping to get some clarification on a topic I don’t think think I’ve fully been able to wrap my head around.
Thanks in advance!
-
10 Replies
-
Your question cannot be specifically answered, because it does not take codecs into account.
Various codecs offer various levels of compression ranging from 2:1 to 40:1, with full or partial color sampling. It also doesn’t take into account whether the compression format is spatial, temporal, or both.
AVI isn’t a compression format, it’s a wrapper that contains a compression format. DivX on the other hand, is a compression format/codec that can live within an avi wrapper.
think of a wrapper as a paper sack that can contain many different products/items. An avi wrapper can contain DV, uncompressed, mpeg, etc.
Perhaps if you explained what you’re looking to better understand at a specific point, you’d get a better answer.Douglas Spotted Eagle
VASST
Aerial Camera/Instructor
Certified Sony Vegas Trainer -
Think of a (avi) wrapper as a paper sack that can contain many different products/items.
That’s the kind of analogy my students (and most faculty!!) can easily understand.
Thanks Spot 🙂 -
Well, DSE, in proving that my question was unanswerable, you actually gave insight into my question.
However, I did ask some more specific questions, such as the analog-digital conversion problem.
And I did give the disclaimer that this is a broad question. Maybe too broad for this forum. I thought if many people offered what they knew, then it would paint a larger picture and be more helpful to me and whoever else happened along this thread.
One of my projects includes converting analog video. So, I’m on the market for something like the Canopus ADVC110 or 300. But it is difficult to shop with any sense of understanding of what’s really going on. They say they use a high-quality codec, so I guess you either have to just take their word for it, or rely on people in places like this to backup their assertion.
I am also considering a camera, but don’t know what the factors I need to consider. Well, I have a basic understanding of digital video, but when it comes to numbers, how is one to know what is acceptable or not. I will not rely on sales people to answer this question for me.
Thanks for your reply,
Naiche Starhawk Lujan -
More information about your application provides the answer you need.
A converter like the ADVC series is DV. End of story. It’s a DV codec. And it’s about as good as it gets. Sony has a slightly better codec, but it’s a discussion of mosquito hair. It’s meaningless at that point.
DV is as good as this gets. Coming from a VHS source or similar, the quality of the DV codec is even less meaningful, as you’re upconverting the analog VHS to DV.
There are no “numbers” when it comes to DV codecs. There are curves, in terms of how blacks/whites are managed, but for the most part, it’s a low-grade codec that doesn’t stand for a lot of manipulation.
If you want to go to a converter such as the Convergent Design SD Connect, you could purchase an SD/SDI card for your computer, serious amounts of drive space, and use a lightly compressed file format in a 4:2:2 sample. Depending on the source, you might consider this route if cost is no object. -
Can you explain the difference between DV and the codec you get with C.D. SD Connect? Also, when you talk about quality, how do you measure it except with you eyes?
Did I read correctly that DV is a low-grade codec? Also did you say that there are multiple versions of the DV codec? I thought that was the standard before HD and that it was universal.
What does it mean to use a 4:2:2 sample?
Also, the best quality analog source would probably be Betacam SP. Are you familiar with that? Is it still an upconversion to DV?
-
[Naiche Lujan] “Did I read correctly that DV is a low-grade codec? Also did you say that there are multiple versions of the DV codec? I thought that was the standard before HD and that it was universal.
What does it mean to use a 4:2:2 sample?”
4:2:2 provides more color information than DV – which here is 4:1:1 and in Europe 4:2:0 – both are DV codecs.
DVD’s provide a 4:2:0 color space universally.
-
What does the ratio 4:2:2 represent, mean, stand for?
Is BetacamSP to DV an upconversion, or a downconversion?
-
4:2:2 means that for every four samples of luma (light) there are two samples of chroma/color. Effectively, color resolutions are half that of the luma. The reason color/light are sampled this way is that the bandwidth(size) of the signal is cut by nearly 1/3 with no perceptual difference.
BETA SP is a downconvert from HDV, and DV is a downconvert from BETA SP.
BETA SP is 4:2:2 and DV is 4:1:1. -
Raymond said:
4:2:2 provides more color information than DV – which here is 4:1:1 and in Europe 4:2:0 – both are DV codecs.DVD’s provide a 4:2:0 color space universally.
*****If the 4 is the luma, is the second and/or third digit chroma? I would imagine that DVD is lower quality than DV since the file sizes are different, and the compression algorithm.
So, if I was editing Beta digitally, is it better to use a dv codec to bring it in or dub to DVD? Does what Raymod said mean that all DVD players handle light/color the same?
The best solution is to use the same sampling 4:2:2 which I can achieve with the Convergent Design SD Connect, right?
-
You shouldn’t consider using a DVD at any point in the import process.
This is *really* very simple.
Get a converter. If you have the $$ to afford a Convergent Design and an SDI card, plus have the HDD speed and space, then this is by far the best, highest quality, etc method.
That said, Beta converted to DV hits the airwaves every second of every day everywhere in the world.
The Convergent Design box is just one of several converters available that offer great analog to digital conversion.
Worrying over/discussing color sampling is more theoretical than beneficial. All you need to know is whether you’re dealing with DV or uncompressed.
Reply to this Discussion! Login or Sign Up