DVC Pro vs Mini DV

Can anyone tell me if they see a noticeable difference when using dvc pro tapes to capture. Does it really make a difference in resolution when the camera is the same?

""" What you do with what you have, is more important than what you could do, with what you don't have."

> > > Knowledge + Action = Wisdom - J. Corbett 1992 """"

[www.adamwilt.com]



For instant answers to more than one hundred common FCP questions, check out the LAFCPUG FAQ Wiki here : [www.lafcpug.org]

thanx
i see i makes no real difference.

""" What you do with what you have, is more important than what you could do, with what you don't have."

> > > Knowledge + Action = Wisdom - J. Corbett 1992 """"

In PAL land, they are different. Dvcpro25 is 4:1:1, DV is 4:2:0. Slight difference.

Quote
strypes
In PAL land, they are different. Dvcpro25 is 4:1:1, DV is 4:2:0. Slight difference

So never convert from one of them to the other in PAL land. You'll always end up with 4:1:0

Andreas

Interesting, Andreas. How do you calculate that?

Pretty simple simple math (and finally viewable math)
If you got a 4:2:0 sampling which is digitally transfered to 4:1:1 the source "2" will be sampled to the target "1" and the source "0" (non existent sampling) will be sampled to a target "1" sampling which results in 0 again as there was nothing within the source.
Same other way round.

Sure there are ways to re-sample sources to get 4:2:2 sampling from either of those sampling rates, but as with any re-sampling this is always a best guess.
As far as I know the 4:1:1 sampling was used to be more compatible within mixed environments with analog and SD digital during life cuts - like with sports, news etc. This is meanwhile not needed any more - I think.

Andreas

4:1:1 is a better choice for interlaced material than 4:2:0, but PAL has a lower vertical chroma resolution because of the PAL method of maintaining colour fidelity over broadcast, making 4:2:0 a more perceptually relevant method. For progressive video, 4:2:0 is superior to 4:1:1, but 4:1:1 is much less prone to bother on interlace material.

Of course, these days, there's no need for 4:x:x at all, and compression should be just used more strongly on a full resolution chroma, rather than decimating and attempting reconstruction afterwards. I guess what I'm saying is that for chroma, stronger compression on full resolution data looks better than less compression on lower resolution data.

Graeme

It's always good when the pros chime in

Andreas

Graeme

So in relationship to the dv and dvcpro is it better to capture from on to the other in terms of editing or output quality?

I do understand that my export choices effect it a lot. However i have found a great receipt to create near perfect quality sd dvd.

""" What you do with what you have, is more important than what you could do, with what you don't have."

> > > Knowledge + Action = Wisdom - J. Corbett 1992 """"

> I guess what I'm saying is that for chroma, stronger compression on full resolution data looks
>better than less compression on lower resolution data.

Hmm... So in this case, stronger compression (4:2:0) on higher resolution data (1920 x 1080) vs lesser compression (4:2:2) on lower resolution data (960 x 720).

Speaking about DVDs... Is it possible to output a progressive 25fps SD DVD? I haven't been able to find the specs for that, especially when there are progressive players out there.

i have never seen or used it, but you can create any thing with compressor. out put it as a m2v file and i am sure that dvdsp can burn it.

""" What you do with what you have, is more important than what you could do, with what you don't have."

> > > Knowledge + Action = Wisdom - J. Corbett 1992 """"

DVCPRO is a Panasonic format. Unless you shot DVCPRO specifically, you need to capture as DV/NTSC. You cannot improve the quality of the footage you shot on DV..that is set in stone once you shot to DV tape. You can improve the color space in terms of rendering filters and color correction (less DV artifacts) if you drop the footage into a DV50 (DVCPRO 50) or uncompressed 8 bit timeline and rendered.


www.shanerosseditor.com

Listen to THE EDIT BAY Podcast on iTunes
[itunes.apple.com]

4:2:0 is not compression per se, but just reduction of resolution. I'm saying that:

say we have a 1920x1080 image, in Y'CbCr, you should get a better image by compressing the chroma 4 times as much with your lossy compression than reducing it's resolution by a factor of 4 by going to 4:2:0. Both should produce the same data rate, but I'd think the first option would look better. However, it might be slower to encode / decode.

Graeme

> You can improve the color space in terms of rendering filters and color correction (less DV
>artifacts) if you drop the footage into a DV50 (DVCPRO 50) or uncompressed 8 bit timeline and
>rendered.

Color space? Rendering DV footage in a DV timeline results in a generational loss, as the footage is decompressed and recompressed back in the render process. Rendering in an Uncompressed sequence results in lesser generational loss, since the footage is not recompressed back to DV.

Btw, what's the equivalent of this workflow in HD? Assuming the footage was acquired from Dvcpro HD. Do I render out my timeline in an "HDTV" preset, with the pixel ratio set to "HD" or is it set to "square"?

I render DVCPRO HD as DVCPRO HD...in a DVCPRO HD timeline. You CAN opt to do this in a ProRes timeline if you want, but I haven't seen any recompression in a DVCPRO HD timeline. It might be there, but you'd have to look close and SQUINT.

DV is another matter...


www.shanerosseditor.com

Listen to THE EDIT BAY Podcast on iTunes
[itunes.apple.com]

Nice tip shane since i adopted your upres techniques i have seen a considerable improvement in my final output.

This tip shall be implemented. So i will edit in dvntsc TL the open a dvcpro hd TL and tweak it.

""" What you do with what you have, is more important than what you could do, with what you don't have."

> > > Knowledge + Action = Wisdom - J. Corbett 1992 """"

> DV is another matter...

DV is infinitely more obvious. An added note is that in PAL world, rendering in an uncompressed timeline requires a shift in fields. Also, that does not go well with speed ramps (they need to be upconverted to uncompressed first before applying a speed ramp)

Reason I'm asking about dvcpro HD, is because it's an 8 bit format. Since I'm currently working on shortform projects, I suppose the benefits is that i can afford to render in uncompressed 10 bit. I'm just afraid that there might be pixel aspect distortion... Any word from the pros?