PDA

View Full Version : DV Codec Compression


wottodo
10-14-2005, 11:48 PM
Hi,

Does anyone know what kind of quality difference there is between different DV codecs? I've seen that Canopus has a 'broadcast quality' DV capture unit, costing about $3200... At $3200, I would hope to see quite a bit of quality improvement over the standard Microsoft DV codec. However I have had no experience in this field, as I have never had access to that kind of equipment. So my question(s) is(are), are there significant quality differences between different DV codecs? Is it possible that my XL2 is recording nice footage, but it's getting ruined when I capture it using a standard Microsoft codec? If so, can I get the better quality with a software codec, or would I need to get a hardware capture card (such as a canopus or pinnacle)?

Integrity
10-15-2005, 03:31 AM
As far as I know there is no difference in the "image quality" of the different DV codecs. They all adhere to the DV format (well some), which is always compressed and stored the same way at a rate of 25 Mbps. The reason it's $3200 is because your paying for the hardware that decompresses/compresses the DV and other things on the fly instead of having your CPU decompress/compress it as with software codecs. Most consider the software ones better now since CPU speed's are very high compared to back then when getting a hardware version was faster.

That Microsoft codec, if your wondering why the quality is bad when you play it in Media Player, is because it renders it out at half the resolution on playback for performance reasons, when you mess around with it in an NLE it'll look fine.

Your camera takes the uncompressed signal of the image and compresses it into the DV format for transport on MiniDV. Whether your using a hardware or software variant, all they do is decode that format. The difference is that some are geared more towards performance, and others to quality. Some add extra headers and other data, some store strictly raw DV.

wottodo
10-15-2005, 06:23 AM
Your camera takes the uncompressed signal of the image and compresses it into the DV format for transport on MiniDV. Whether your using a hardware or software variant, all they do is decode that format. The difference is that some are geared more towards performance, and others to quality. Some add extra headers and other data, some store strictly raw DV.

Does that mean that there are some software codecs (irrelevant of how they are implemented) that give you a better picture at the cost of performance? That seems to contradict your statement that they all have the same image quality. I know I'm probably just misunderstanding your statement, so if you could just clarify that for me that would be great.

Integrity
10-15-2005, 10:03 PM
Your right, I contradicted myself. Mainly because I don't know it down to detail...the actual coding. When I read your first post I was reminded that at one time I had wondered about the same thing. So to finally answer my own questions and yours I went and searched on Google and I got most of the information from here...

http://www.well.com/user/richardl/Features.html

...the rest I knew just because I knew it beforehand and through common sense. At one time I used Canopus's DVRaptor card, because I had thought that (from their advertising) they had one of the best codecs; my goal was to find one that made the file size the smallest it could be without quality loss. Then I found out that they all comply with the DV format, so no matter which one I chose it would be the same. So now I use one of the software codecs with no problems with a standard IEEE-1394 port that came with my computer.

From that article, to answer you last question read the section called "Smooth Playback on Computer Desktop" that's in bold. What I meant to say instead of performance versus quality in the last paragraph was performance versus accuracy. Which means yes, down to the actual rendering, they should have all the same quality, because they are all using the DV format. But I don't know for sure, some might be more accurate, and others might be trashing some information to keep performance. Just keep in mind that when you capture DV from your camera, all your doing is copying the data from the tape to your hard drive, the DV codec just decompresses it for editing/playback, and compresses it back when your done.

CGTalk Moderation
10-15-2005, 10:03 PM
This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.