|
|||||||||
|
Thread Tools | Search this Thread |
January 17th, 2006, 06:30 AM | #1 |
Regular Crew
Join Date: Nov 2003
Location: New England
Posts: 195
|
Final Cut Pro Question
Sorry if this is a silly question.
Yesterday I was in the Apple Store. The salesman said that Final Cut Pro can edit HDV in native MPEG-2. Is this true? If so, how? HDV MPEG-2 is a GOP of 15 frames right? So then how can you cut in the the middle of a GOP??? Thanks. |
January 17th, 2006, 08:48 AM | #2 | |
Regular Crew
Join Date: Jan 2005
Location: Belgium
Posts: 139
|
Quote:
It's no problem for FCP to display any frame from anywhere in the GOP. The only trouble is that your camera doesn't work that way. So FCP will recompress the full timeline to HDV when you write the timeline to tape with Print to Video.
__________________
Please keep in mind that English is not my native language. |
|
January 17th, 2006, 09:48 AM | #3 |
Major Player
Join Date: Nov 2004
Location: Portsmouth, UK
Posts: 611
|
FCP will place a new I-frame directly at the point of the edit, splitting the GOP. When you what to go back to tape, it will conform the timeline, and all split GOPs will be re-rendered as proper 12/15 frame GOPS.
|
January 17th, 2006, 10:22 AM | #4 |
Major Player
Join Date: Sep 2005
Location: NE of London, England
Posts: 788
|
To add to the question, is there ANY way to do realtime HD monitoring whilst editing HDV? Even HDV 1 720p? Does it have to be transcoded to an intraframe codec first?
|
January 17th, 2006, 11:49 AM | #5 |
Regular Crew
Join Date: Nov 2003
Location: New England
Posts: 195
|
I understand that. But what the salesman said was that the final video would not have to be re-Mpeged. I said "how is that possible?" and he said that only the frames in between the GOP would be re-encoded, the rest would be the same MPEG-2 data.
|
January 17th, 2006, 02:03 PM | #6 | |
Wrangler
Join Date: Dec 2002
Location: Mays Landing, NJ
Posts: 11,802
|
Quote:
|
|
January 17th, 2006, 02:56 PM | #7 | |
Regular Crew
Join Date: Jan 2005
Location: Belgium
Posts: 139
|
Quote:
The only thing I can think of is that FCP sends an occasional shorter GOP to the camera in order to preserve as much full original GOPs as possible. This needs examination.
__________________
Please keep in mind that English is not my native language. |
|
January 17th, 2006, 03:13 PM | #8 | |
Regular Crew
Join Date: Jan 2005
Location: Belgium
Posts: 139
|
Quote:
Transcoding wouldn't make sense for this purpose. HDV is the only thing your camera will accept.
__________________
Please keep in mind that English is not my native language. |
|
January 17th, 2006, 03:41 PM | #9 | |
Major Player
Join Date: Sep 2005
Location: NE of London, England
Posts: 788
|
Quote:
Sorry, I meant a full quality preview for critical monitoring. From my experience, digital cinema desktop isn't up to that, but is it better on a higher spec system? Is a Decklink or Kona the way to go? I'm hoping FCP6 will output a proper HD signal through a second (or 3rd) DVI. |
|
January 17th, 2006, 04:01 PM | #10 | |
Wrangler
Join Date: Dec 2002
Location: Mays Landing, NJ
Posts: 11,802
|
Quote:
Regarding FCP6, maybe you know something we don't? :-) I've seen the rumors but they didn't look all that believeable. Apple is just releasing the new universal version of FCP5 next month, so I'm not convinced that version 6 will appear too soon. But who knows?... maybe they want to make a big splash at NAB? |
|
January 17th, 2006, 04:23 PM | #11 | |
Wrangler
Join Date: Aug 2005
Location: Toronto, ON, Canada
Posts: 3,637
|
Quote:
__________________
Tim Dashwood |
|
January 17th, 2006, 06:38 PM | #12 |
Regular Crew
Join Date: Jan 2005
Location: Belgium
Posts: 139
|
The problem with Digital Cinema Desktop - and in extension, all full screen computer previews - is twofold or threefold:
a) gamma-curves: TV monitors and computer monitors do not have the same gamma curve. In addition, Macs and PCs use a different gamma curve: 1.8 vs. 2.2. Midtones look completely different on my SD monitor versus Digital Cinema Desktop. b) colour space conversions: all video is stored in some kind of Luma+Chroma system, commonly called YUV or YPbPr. Computer monitors display RGB imagery. There are multiple ways to transform a video in the YUV colour space to the RGB color space. SD video uses ITU-R 601, HD uses ITU-R 709. The difference between both systems is small but it should be visible in the greens. In the case of Digital Cinema Desktop, the conversion from YUV to RGB will be handled by Quicktime. In other systems the conversion may be handled by your graphics card. I haven't seen any documentation on what colour space conversion the various systems, including Digital Cinema Desktop, use. c) interlacing: a computer display is progressive and your footage may be interlaced. Digital Cinema Desktop drops one of both fields if the RT playback quality is less than High. Otherwise is shows both fields with the typical combing artifacts you get when you play interlaced footage on a progressive screen without deinterlacing. To get acurate realtime HD playback with a Mac, you need: a) an interface card from Blackmagic or AJA with HD-SDI and/or HD component out and b) an HD display: either a TV-monitor like JVCs 19 inch HD monitor or Blackmagics HD Link with a 23 inch cinema display. NVidia has some graphics cards with HD component out on the PC side, but I wouldn't really trust it. You don't know what happens with the signal before it gets to the HD output. Is it converted from YUV to RGB and back to YUV again? What happens with interlaced footage? The reason I don't trust these solutions is because of the way they work. AJA and Blackmagic cards interface with Quicktime while graphic cards with HD out interface with the OS, meaning there will be a whole lot more layers between the footage and the display.
__________________
Please keep in mind that English is not my native language. |
January 18th, 2006, 09:35 AM | #13 |
Major Player
Join Date: Sep 2005
Location: NE of London, England
Posts: 788
|
Thanks again guys!
My comment about FCP 6 was just casual. I'm sure it'll be here one day. NAB 2006? Probably by the end of 2007. A decent HD output would be a major selling point IMO. I'm hoping they may even include HDSDI and HD component I/O in the Intel Powermacs. Hoping being the key word (not expecting!). It would save the hassle of additional cards. |
January 18th, 2006, 10:14 AM | #14 | |
Wrangler
Join Date: Dec 2002
Location: Mays Landing, NJ
Posts: 11,802
|
Quote:
But who knows?.... we do know that Steve Jobs likes surprises :-) |
|
January 18th, 2006, 04:21 PM | #15 | |
Major Player
Join Date: Sep 2005
Location: NE of London, England
Posts: 788
|
Quote:
Wouldn't FCP6 be pretty much based on 5's language, so the rewritten 5 could then form the basis for 6... if that makes sense!?! Basically it wouldn't be such a waste of time, as it needs to be done anyway. I don't know if that is how software programming works though. 6 may be "announced" at NAB but Apple tend to announce and release pretty close together. Sorry for the Hi-jack! |
|
| ||||||
|
|