Comments

FrigidNDEditing wrote on 7/24/2005, 3:24 PM
2-3-3-2 = less work for your cpu - when you're in editing
2-3 = what you want to use when you render it for output to DVD

(maybe it's the other way around, but I don't think so - and someone will tell you for sure)

But that's the 25 cent answer.

Dave
IanJames wrote on 7/24/2005, 3:34 PM
thanks -
is there a second 25 cent answer as to the difference in appearance between these two? (we're up to 50 cents now...)
FrigidNDEditing wrote on 7/24/2005, 3:48 PM
None - to the best of my knowledge, big spender.

only real difference is that the 2-3-3-2 makes it more playable in the T/L, but the 2-3 works better for DVD (I THINK* - because it takes up less space maybe? - It's been a while since I've checked into it. If you check the help files in Sony Vegas - it will explain it.)

Dave
Trichome wrote on 7/24/2005, 4:35 PM
http://download.sonymediasoftware.com/whitepapers/24p.pdf

This is the link to the Sony 24P Whitepaper. I'm sure you've seen it but perhaps a second read will help you answer these questions.

Cheers.
TimTyler wrote on 7/24/2005, 7:59 PM
AFAIK -

2-3-3-2 is the format of video shot with a DVX100 in 24PA mode. All frames are progressive; there is no visible interlacing. Kind of like working with real frames of film. Obviously since it's still NTSC video there are 29.97 frames, however the extra frame data (29.97 minus 24) is ignore by Vegas in 2-3-3-2 mode.

2-3 pulldown is the format that you'd have if you transfered film to tape in a telecine. A couple of the video frames contain one exposure in one video field and another exposure in the other video field of the same frame. The rest of the video frames are progressive; one exposure = one frame = two fields. Vegas can recognize and undo the pulldown so that you can work in 24P frame mode.
BarryGreen wrote on 7/25/2005, 12:34 AM
To understand the difference you first have to clarify what the pulldown sequences are accomplishing. Each pulldown scheme turns 24 progressive frames into 30 interlaced frames, adding five frames per second. As in, creating five frames. The way this breaks down is, for every four source frames, the system will output five destination frames. It does this six times per second (4 source x 6 = 24, 5 dest x 6 = 30).

2-3-3-2 means lossless rendering on unchanged frames. The 24p frames are written out as raw pure frames, whereas the added fifth frame is created by combining fields from adjacent frames. If you haven't modified the footage (i.e., through color correction or adding a title or something) then the original DV frames will be written directly to the output file with no recompression and no additional loss.

2-3 means that the 4 source frames will be "spread out" across 10 video fields (or five video "frames"). So frames are split in half into fields and then fairly evenly distributed.

What this means is, your footage will be uncompressed, field-split, combined, and recompressed when you render out to 2-3. That means an added layer of DV compression. Whereas with 2-3-3-2, the frames will be written out as-is, and a new discardable fifth frame will be added just to round out the sequence.

Okay, babbling and gobbledygook over, here's what you want to know: always, always, always use 2-3-3-2 whenever possible. The only times you ever want to entertain using 2-3 are:
1) when you're making your final render for distribution on NTSC tape (VHS, BetaSP for broadcast, etc)

or

2) If you're editing on a 60i timeline instead of a 24p timeline.

Unless you're doing one of those two things, always stick with 2-3-3-2. When shooting, when editing, when making temporary renders, when rendering preview files, etc -- always stick with 2-3-3-2.