|
|||||||||
|
Thread Tools | Search this Thread |
April 3rd, 2006, 03:56 PM | #1 |
Major Player
Join Date: Sep 2004
Location: New York, NY
Posts: 775
|
24F pulldown options through SD-SDI
Hey,
I'm thinking of shooting something soon using the SDI port and connecting to a DVCPro50 deck. When shooting using the SDI port, does the 2:3 pulldown apply or do you have the ability to apply 2:3:3:2 pulldown to tape (just by selecting it from the menu). Thanks in advance ... Mike |
April 3rd, 2006, 06:38 PM | #2 |
Trustee
Join Date: Jan 2006
Location: New York, NY
Posts: 1,267
|
Michael,
The camera applies the pulldown. DV has advanced but HDV doesn't because of the way it is recorded. See Page 56 and 57 Manual For Outputs. It is in PDF form on the Canon website It is too hard to cover all the scenarios easily. You should check out the Manual on Page 100, Page 101 One thing I did notice is the SDI Output looks like it is 640x480 while the tape is 720x480. This may be a problem with the Manual. Or it may be the way SDI fills a 4x3 screen. HDV is 1920x1080 |
April 3rd, 2006, 06:57 PM | #3 |
Major Player
Join Date: Sep 2004
Location: New York, NY
Posts: 775
|
Thanks for the response. I'll check out the manuals when I'm not at work.
I know that the camera applies pulldown. I'm just not certain whether the menu option of 2:3:3:2 in SD mode applies to the SDI out put (I know in HD mode, it automatically applies a 2:3 pulldown). I'll look at the manual and see if it answers the question. |
April 4th, 2006, 09:01 AM | #4 |
Trustee
Join Date: Jan 2006
Location: New York, NY
Posts: 1,267
|
Michael,
The pulldown in the SDI follows what is set in the camera. In DV 24F if it is advanced then it is 3:2:2:3 and regular is 3:2. HDV doesn't use advanced |
April 4th, 2006, 09:32 AM | #5 |
Major Player
Join Date: Mar 2005
Location: UT
Posts: 945
|
Michael,
Why even rent the DVCpro50 deck? If you shoot HDV and downconvert via MPEG Streamclip or Compressor, you will end up with a 4.2.2. DV50 file anyway. Whether you go SD-SDI out or HDV downconvert, you will have to extract pulldown. The SD results will be similar but if you shoot HDV you will have HD material to return to if ever desired. |
April 4th, 2006, 04:12 PM | #6 |
Major Player
Join Date: Sep 2004
Location: New York, NY
Posts: 775
|
Thanks Barlow ...
Your method sounds like a great workflow (which I use to convert 24F .m2t's to DVCProHD currently). But I'm curious ... How is it that the converting HDV to DVCPro50 with MPEG Streamclip turn the 4:2:0 to 4:2:2? |
April 4th, 2006, 05:03 PM | #7 |
Major Player
Join Date: Mar 2005
Location: UT
Posts: 945
|
There's something like 720x540 chroma sampling in 1080i HDV. Thats' actually enough to have 4.4.4. SD via downconvert to Animation codec or PhotoJPEG@100% quality.(my recommended format) Of course, that's in theory, but I do know that you will definitely get good 4.2.2 DV50 from HDV.
|
April 4th, 2006, 10:01 PM | #8 |
Major Player
Join Date: Sep 2004
Location: New York, NY
Posts: 775
|
So in your experience, which yields better quality ... downconverting to DVCPro50 or PhotoJPEG for SD?
|
April 4th, 2006, 10:14 PM | #9 |
Major Player
Join Date: Mar 2005
Location: UT
Posts: 945
|
PhotoJPEG. I like it because it's a square pixel format which makes it especially easy to integrate photoshop graphics without all that pixel aspect ratio preview crapola. Renders faster than DV50 too. Also, it's just plain cleaner and kinder to the image than DV50...but it's very close. You have to make sure to give it a good 75% quality conversion or higher. I would actually go 100% with SD downconversion. And make the frame size 854x480 so that you have a full raster to convert to widescreen DVD and whatnot.
|
April 5th, 2006, 07:55 AM | #10 |
Major Player
Join Date: Sep 2004
Location: New York, NY
Posts: 775
|
very interesting. how does PhotoJpeg render faster? I did a test yesterday where I brought in the same clip, one encoded to DV50, the other PhotoJPEG at 100%, into FCP timelines with their respective properties. Then I applied a set of 3 processsor intensive filters on the footage and rendered. THe PhotoJPEG took twice as long to render than the DV50. Could it be because DV50 is a native setting in FCP and the PhotoJPEG, I have to make a custom setting (including the 854x480 aspect ratio)?
Oh, I have a G5 Quad by the way ... |
| ||||||
|
|