What's the benefit of exporting to Compressor directly from FCP??? at DVinfo.net

Rob Moreno · March 16th, 2006, 12:27 AM

Can anyone tell me why it takes soooo much longer to compress video from FCP (i.e., using Export to Compressor) than it does to export the timeline as a QuickTime movie and then use Compressor? For example, exporting a 14 min. sequence from the FCP timeline using Export to Compressor (MPEG-2 best setting) gave me an estimated render time of over 6 hours! Exporting the same timeline as a standalone QuickTime movie and then importing that file into Compressor with the same settings gave me an estimated render time of 1 and a half hours. What gives? Is there any difference in quality? Does exporting to Compressor directly from FCP give better rendering quality by, for example, placing I-frames at the beginning and end of each cut and transition?

Nate Schmidt · March 16th, 2006, 12:39 PM

I think it does have something to do with the I-frames. If you export directly to compressor it places one at the beginning of the transitions, and if you export a quicktime movie and then go to compressor it places an I-frame at the end of the transition, (or the other way around I'm not 100% sure) other than that I don't think there is any difference. I would reccomend exporting a quicktime then dropping that into Compressor. If you need an I-fram in a specific spot just place a compression marker in the timeline where you need it.

Jack D. Hubbard · March 16th, 2006, 02:00 PM

I use Sorenson 4.3 ... works pretty much the same. By taking the project down to a QT movie makes a lot of sense. Project (in HDV) might have as many as four audio and four video streams with effects, etc. FCP just hates that, so I make a QT movie; now you only have one audio and two video streams and the time to make is much less. Then send the QT movie (self-contained) to the compressor and the process ought to be a lot quicker.

Rob Moreno · March 16th, 2006, 06:47 PM

I thought the placement of I-frames would be about the only difference as far as quality goes. But still, a 4.5 hour difference in rendering time just for I-frames??? In the case of the example I had mentioned previously, the sequence only had 2 layers of video and 1 mono audio track. A chromakey filter was applied to a few clips, but other than that there was no complex editing. There must be some kind of bottleneck issue when exporting to Compressor from FCP.

Ben De Rydt · March 17th, 2006, 06:25 AM

FCP will render each frame to uncompressed (twice*) when doing Export to Compressor. It's functionally equivalent to exporting the movie to Quicktime Uncompressed and importing that in Compressor, minus the disk space needed. That means that you need to account for time FCP spends rendering on top of the time for MPEG 2 compression.

(twice*: according to Graeme Nattress. I can't verify that.)

Rob Moreno · March 17th, 2006, 06:53 AM

Wow. Well that certainly explains the huge difference in rendering time. Of course, the example I mentioned at the beginning of this thread was 10-bit uncompressed, so I wonder what uncompressed format FCP was converting this "uncompressed" footage to that would make it take so long? Anyway, I would hope it results in significantly higher image quality. I can't say I've ever noticed though...

March 16th, 2006, 12:27 AM	#1
Rob Moreno Regular Crew Join Date: Jul 2002 Location: Tokyo Posts: 102	What's the benefit of exporting to Compressor directly from FCP??? Can anyone tell me why it takes soooo much longer to compress video from FCP (i.e., using Export to Compressor) than it does to export the timeline as a QuickTime movie and then use Compressor? For example, exporting a 14 min. sequence from the FCP timeline using Export to Compressor (MPEG-2 best setting) gave me an estimated render time of over 6 hours! Exporting the same timeline as a standalone QuickTime movie and then importing that file into Compressor with the same settings gave me an estimated render time of 1 and a half hours. What gives? Is there any difference in quality? Does exporting to Compressor directly from FCP give better rendering quality by, for example, placing I-frames at the beginning and end of each cut and transition? __________________ Media Kobo Japan

March 16th, 2006, 12:39 PM	#2
Nate Schmidt Major Player Join Date: Jul 2005 Location: Welland ON Posts: 515	I think it does have something to do with the I-frames. If you export directly to compressor it places one at the beginning of the transitions, and if you export a quicktime movie and then go to compressor it places an I-frame at the end of the transition, (or the other way around I'm not 100% sure) other than that I don't think there is any difference. I would reccomend exporting a quicktime then dropping that into Compressor. If you need an I-fram in a specific spot just place a compression marker in the timeline where you need it. __________________ "We've done the impossible, and that makes us mighty."

March 16th, 2006, 02:00 PM	#3
Jack D. Hubbard Major Player Join Date: Jun 2005 Location: Menlo Park, CA Posts: 342	Compression I use Sorenson 4.3 ... works pretty much the same. By taking the project down to a QT movie makes a lot of sense. Project (in HDV) might have as many as four audio and four video streams with effects, etc. FCP just hates that, so I make a QT movie; now you only have one audio and two video streams and the time to make is much less. Then send the QT movie (self-contained) to the compressor and the process ought to be a lot quicker.

March 16th, 2006, 06:47 PM	#4
Rob Moreno Regular Crew Join Date: Jul 2002 Location: Tokyo Posts: 102	I thought the placement of I-frames would be about the only difference as far as quality goes. But still, a 4.5 hour difference in rendering time just for I-frames??? In the case of the example I had mentioned previously, the sequence only had 2 layers of video and 1 mono audio track. A chromakey filter was applied to a few clips, but other than that there was no complex editing. There must be some kind of bottleneck issue when exporting to Compressor from FCP. __________________ Media Kobo Japan

March 17th, 2006, 06:25 AM	#5
Ben De Rydt Regular Crew Join Date: Jan 2005 Location: Belgium Posts: 139	FCP will render each frame to uncompressed (twice) when doing Export to Compressor. It's functionally equivalent to exporting the movie to Quicktime Uncompressed and importing that in Compressor, minus the disk space needed. That means that you need to account for time FCP spends rendering on top of the time for MPEG 2 compression. (twice: according to Graeme Nattress. I can't verify that.) __________________ Please keep in mind that English is not my native language.

March 17th, 2006, 06:53 AM	#6
Rob Moreno Regular Crew Join Date: Jul 2002 Location: Tokyo Posts: 102	Wow. Well that certainly explains the huge difference in rendering time. Of course, the example I mentioned at the beginning of this thread was 10-bit uncompressed, so I wonder what uncompressed format FCP was converting this "uncompressed" footage to that would make it take so long? Anyway, I would hope it results in significantly higher image quality. I can't say I've ever noticed though... __________________ Media Kobo Japan