yeh, i've been watching the Avid site regarding MC 3.0 also.

i'm doubtful encoding mpeg from Avid would look much (if any) better than what we already achieve. i think, to kick it up to the next level, a standalone transcoder would be needed.

-rob

It is not the best thing to let your NLE or or worse yet let the DVD Authoring software do your transcoding. If you want the best use a compression specific software i.e. Squeeze, expensive yes but the "best" comes with a price.

That being said Vegas does a decent job encoding for DVD. Just select the Mpeg 2, DVD Architect NTSC or PAL setting. It is the best you are going to get without spending hundreds or even thousands of dollars on compression specific software.

I haven't read this thread fully, but here are some things that might help:

1- You might check that the color space conversions are correct. If you are using the default Vegas codecs (main concept mpeg2, vegas DV) in a 8-bit project then the levels will be correct by default (*certain extremely illegal colors may get clipped). In a 32-bit project you will need to manually intervene.

2- If you encode your DVD at a constant bitrate of 7mbps (or higher; I can't remember what bitrate you shouldn't exceed) then the quality of the encoder shouldn't be an issue.

The PAL DV standard (except PAL DVCPRO) calls for different chroma siting than what MPEG-2 calls for. I've read neither standard myself so this might not be accurate. But I believe that's what it is. Poynton has a diagram here:
http://www.poynton.com/PDFs/Chroma_s...g_notation.pdf
In PAL DV, the Cb is shifted one line compared to Cr... I believe it's a DSP shortcut to lower cost.

In practice, it's likely that PAL cameras, the codec in NLEs (and intermediate codecs like Cineform), and MPEG-2 encoders all differ in what chroma siting they follow... so the chroma siting can get screwed up and shifted a line. I think this is subtle enough that most people don't notice. But interlaced 4:2:0 has more objectionable artifacts to begin with, so I wouldn't worry too much about the chroma siting getting screwed up.

see this site:
http://www.hometheaterhifi.com/volum...ug-4-2001.html
(the artifacts are easier to see on a LCD)

Hi,

Glen, thank's a zillion for the links to the very interesting and informative articles. I like to read such material, due to my background in electronics. Additionally, it helps to understand the compromizes and shortcomings of all the different codecs available. It never hurts to know to much... ;)

It was good to learn that there are subcategories of 4:2:0 coding in MPEG-2.

Knowing now the most important Vegas lurks (what comes to color space conversion), I'm fully aware of and keen to get always these correctly set up. I encode, if possible, always at 8Mbit/s rates, assuming the video fits on the final DVD. Having done all this you suggest but still - I'm not satisfied with the overall sharpness (or lack of the same) in the final DVDs.

I would eagerly like to try another codec, but wich one? Can anyone share his/her experiences here? Frankly speaking, I'm tired of experimenting, just let me know what codec will produce better results, and I get it!!!

Christian

Several were mentioned above, but again.... they cost money.

Christian

Mpeg 2 is the only codec that works on DVD. If you plan on staying within Vegas you got the best it is going to get. No way around it.

Another possibility is that differences in setup might be making one look better than the other.
I believe only the >$10k DVCAM VTRs output 7.5 IRE setup correctly... the lower end models will output 0 IRE (which is wrong).
Your DVD player probably puts out 7.5 IRE (but some players might be switchable or just output 0 IRE).

One source might be 0 IRE and the other 7.5 so the colors look different. To quickly check things, you can go into a broadcast monitor's settings and toggle 0/7.5 IRE for the analog input... if this makes the image look right, then that might be the problem right there.

2- I'm assuming you live in a NTSC country other than Japan, where the standard black level is 7.5 IRE. It's 0 IRE elsewhere.

Seems to me from what I've read in this thread, the cost effective place to place dollars and effort for maximum quality in the final project would appear to be into better cameras, not better software. It seems that while tweaking settings or buying a $2k on a better encoder could help, ultimately something is going to be lost in the encoding process. And while it is, of course, useful to experiment and to try to sqeeze the best quality we can from the footage we have, one would be best served by focusing most heavily on obtaining quality footage with the best/largest chipped cams we can afford.

Hi,

Am I wrong in the following assumption:

My SR12 AVCHD cam is (under good lighting conditions) producting high quality video, that is comparable (after downsampling to SD) in sharpness with a "professional" standard definition cam - assuming that this downsampling is done from AVCHD format to AVI.

In other words, shouldn't I be able to produce sharp professional looking material (using a prosumer AVCHD cam) and downsampling it to PAL resolution AVI, and then rendering out to Mpg2?

At least my material looks stunning on my PAL monitor in a PAL project during timeline previews. Rendering to DVD Mpeg2 ruins this sharpness... What bogs me is that I have seen comercially produced DVD's that are so much sharper. Ok, I know that they have better encoding, and that's what I'm looking for. My source material looks sharp enough as to be from a professional SD cam...

I'm prepared to spend something like 500$ for a better quality Mpeg2 codec. Just need to know wich one to get. Again, feedback of real experiences from you is worth a lot... Why should I spend $$$ on a pro-grade CAM if the Mpge2 encoding is to blame?

Christian

Christian, I guess $500 for marked improvement would be worth it, IF it is REALLY significantly better. If a better codec is all that was needed to produce professional looking productions, there would be no need for professional cameras.

It seems to be no matter what codec we use, however, we are going to lose something, unfortunately. I'm saying that in my uneducated opinion the bigger improvement in your end quality will come from better cam.

Example, $500 plus the price of your cam and you'd be getting dangerously close to the cost of a used 3-chip camera.

Under good lighting conditions your cam will get great footage, and it may rival a better cam under perfect conditions, but your digital footage will never consist of as high quality information as a camera with bigger, better chips.

I have had a single 1/3" chip consumer camera that in perfect conditions shot great video, comparable to my 3-chip camera. But ultimately it was a toy. I used the small cam when I started out as a backup cam because I couldn't afford a third pro camera. Ultimately I sold off the small cam and got a better cam because the small camera just couldn't cut it. I tried in post to do everything I could to "fix it", but what can you do? There is very little you can do.

There is simply no making footage from a consumer camera appear better than it is. You will alway expeirience a loss in quality in the conversion process, so you must start with the best camera possible. That is just my my opinion. In my opinion I think you are going in the wrong direction in trying to improve your video, but I think it is a useful discsussion and am glad you brought it up.

Ben, I suggest you post a short clip demonstrating the problem. Simply render a few seconds to DV-AVI as well as MPEG-2 for DVD.