Matt Devino
Forum Replies Created
-
There is no need to conform in After Effects at all, it’s a complete waste of time. Any i/o card (i.e. Kona 3, Blackmagic, etc.) will be able to output your FCP sequence to HDCAM as 1080i, whether your work in a ProRes sequence, a DVCPRO HD sequence, an XDCAM sequence, or an Uncompressed sequence. The only important thing is you pick a frame rate to work in and stick with it.
Everything can be converted to 1080i very nicely, it’s going from 29.97 to 23.98 that causes problems.
Because you have multiple frame rates going on here, and one of them happens to be 29.97, and you are finishing at 29.97, then you should work in 29.97. Your 23.98 material will convert to 29.97 by adding pulldown. HDCAM @ 1080i is just Uncompressed 8-bit video running at 29.97 interlaced. So work at 29.97. Also 30P timelines output to 1080i tape perfectly fine, trust me I do it all the time.If you want to conform all of your footage to the same format you can do that, but I personally wouldn’t. FCP does a perfectly fine job of scaling 720P and DVCPRO HD 1080P material to 1920×1080 in the timeline. Also the timelines in FCP are open format, so if you were to work in a Pro Res HQ timeline at 29.97, and you turn unlimited real time on, you can edit in real time (you did say you had 4 brand new mac pros so the hardware is fast enough for this). Then you just do your edit and render when you are finished, and output that timeline to tape. Another reason to work with the original clips is so you don’t have 4 copies of every clip laying around (original clip, converted clip, and backups of both).
If you do decide to conform to one codec, I think you should go to ProResHQ 29.97. Progressive vs interlaced is your choice. Here’s why to chose either:
Progressive:
The only reason I said to set your field dominance to “none” was because all of your footage is progressive. If you work in 1080i the 59.94 material would look very similar to the interlaced video. This will cause the “look” of your show to change from shot to shot if you are viewing it on an interlaced monitor. If you simply work in a 29.97 progressive timeline, all your formats will look like they were shot 30P to begin with, and the 59.94 footage will look like it was shot 30P.Something to figure out is what the intention of the of the 59.94 material was. If it’s for creating slomo clips from it then you will need to conform that footage to 29.97 using compressor or cinema tools. This will make it proper 30P and be a smooth slomo clip.
Interlaced:
You have a lot of 23.98 footage, and if drop the 23.98 clips into a 29.97 timeline it fills the gaps with duplicate frames (I don’t think FCP adds proper 3:2 pulldown to 23.98 clips just by dropping them in the timeline, unless that was added in an update I don’t know about.). This looks perfectly fine to most eyes, but it can look like the footage is strobing to some. To get around this you will want to make sure you add proper 3:2 pulldown to your 23.98 footage which you can do in compressor. In this case you would probably want to convert the 59.94 material to 29.97 before you start working with it or you will get that mixed looks problem I was talking about.My biggest reason for working in a ProRes HQ timeline vs another kind of timeline is HDCAM tape is uncompressed 8-bit 4:2:2, ProRes HQ is lightly compressed 10-bit 4:2:2. If you chose ProRes HQ your renders will look better and take up less disc space than working in uncompressed 8-bit, and you can edit in real time off of a firewire drive if you wanted to instead of needing a RAID for the uncompressed video playback. You also have a ton of ProRes video already because of the RED footage, so that material will be native in the timeline. ProResHQ is also a higher quality codec than the XDCAM and DVCPRO footage you have so you won’t be losing any quality by using ProResHQ. in the end if you are worried you can change your timeline to uncompressed after your edit is done and re-render it before you output, but it would be wasteful to do all your cutting in uncompressed from the beginning.
-
Work in a ProResHQ 1920×1080 29.97 timeline w/field dominance set to None. This will output to your HDCAM tape nicely.
-
I agree with these guys, ProRes (not HQ) is the way to go. For the 60i stuff just capture it as 60i, but work in a 30P timeline and it should look nearly identical to the 30P footage. Use the easy setups FCP provides for HDV to ProRes conversion over firewire, they will keep your footage at the original 1440×1080 aspect ratio which is a good thing as it will save drive space. You can convert to 1920×1080 square pixels when you are finished editing and ready to master.
-
Sounds like a lot of the jobs I end up working on. Like everyone else has said so far, once you have an editor talk with them about it. Second thing is making sure your timeline is setup right. If there is ANY 29.97 footage I would work in that frame rate, not 23.98. Third, since you have so many different formats you are going to have to choose a codec to work in, maybe go with ProRes HQ? Fourth thing is the RED footage. When I first started working with RED in a mixed media project I used to do what you mentioned, using clip finder then rendering out selects to ProRes. This can take FOREVER. It’s a lot easier to just bring import the full size proxies and work with those in FCP, if you have a fast enough machine you can edit them in real time when you set your RT to unlimited. Just render when your cut is done. Good luck.
-
Here’s another question for you, any way to create a second partition on my boot drive and clone the 1st partition to the new partition, and install Snow Leopard on only one of the partitions? Is this just crazy talk?
-
ProRes and 29.97, it’s straight 30P out of the camera.
-
If you HAVE to work in 24P, use compressor. Set the frame rate to 23.976 in the video settings, you will have to type this in manually. Then turn the frame controls on. Set rate conversion to best and make sure fields are set to progressive. DO NOT click the button at bottom that says “so source frames play at 23.976”. This is your best bet for converting 30 to 24, it actually looks pretty decent, it does an even better job if you are starting with 1080i footage but the 5D is progressive. Really though, if you aren’t finishing on film, work in 30P, 99% of humans won’t know the difference.
-
What kind of sequence are you working in?
-
What kind of footage are you working with? What are your sequence settings? Are your timeline render options set to 8-bit, 10-bit YUV, or RGB? Are you exporting a quicktime using “quicktime conversion” or did you render an uncompressed sequence then just export that using the normal “export quicktime movie?”
-
If you are shooting correctly with the v1u you won’t need to reverse telecine with cinema tools, the easy setup you are using will remove the pulldown on capture. You have to be shooting in HDV 1080P 24A mode on the camera. In 24A mode the camera records a constant A-frame, in normal 24 mode it doesn’t do that so that A-frame shifts every time you hit record. If you shoot 24A then just use the HDV 1080P 24 to ProRes easy setup and capture normally, and you will see the capture window display something that says a percentage of how far it is behind the camera, this lag occurs because FCP is processing the pulldown removal on the fly, but it can’t keep up with the camera so it needs a buffer. If you didn’t shoot 24A, then you need to capture using the HDV 1080i 29.97 easy setup, and remove the pulldown and convert to prores using compressor, not cinema tools, because cinema tools can’t remove pulldown from HDV or any other mpeg-2 material. Hope this helps.