Generally speaking, you control size by choice of codec, frame size, frame rate, and/or bit rate. There is – of course – a trade off between size and quality.
I think you will get the most useful advice if you tell us what you intend to do with this movie. Do you need to upload it to a Web video service like YouTube?
Thanks for writing me back. After I read your respond, I did some tests in Compressor for a few second movie. After I confrimed the setting that I set would give me an optimial quality and file size. I used this setting in FCP to export my project. The problem is transcoding takes forever. I started last night at about 9pm. Transcoding has only completed up to 30%. Is that normal?
Here's what I set for my 1440x1080 29.97i 48min project:
FRAME SIZE: 1280x960
FRAME RATE: 15fps
FRAME CONTROL: Progressive Output
RATE CONVERSION: Motion Compensated
I did output the same projec from Compressor before, it still toke a couple of hours to complete but won't be that long.
Am I doing right?
Am I doing right?
Hard to say; you have to take into account the quality. How does it look? Does it play smoothly?
Without knowing about your system details, there are a couple of things that are probably dragging out the processing time: 1) resizing and 2) de-interlacing.
If I take an h264 file (or a Pro Res 422) and lower the data rate for Web distribution it takes about 2X real time to compress. That is, a 15 minute file takes about 30 minutes without using clusters. (That's on a 3 year old 5i iMac – not exactly state of the art hardware.) So all things equal I would expect your 30 minute to be re-compressed in an hour or less on my system.
But de-interlacing in Compreesor is time-intensive. The good news is it does a good job. The bad new, you already know. If you have the option, shoot progressive.
Tghe resizing issue is a bit complicated. The camera shot an anamorphic (non square) pixel shape. Web services want square. I guess I would re-size to 720P using square pixels.
But do yourself a favor and do a test with a small section of your movie before committing to a setting.
After more than 24hrs of transcoding, my project is done. Like you mentioned, it toke twice longer than normal to convert my project. Before my movie was about 500MB. Now the size is about half of that. That's good. The bad thing is it takes too long to complete the job.
I have a 2.4GHz Dual Core MacBookPro. Hardware would be one factor. The settings (reducing frame rate, compresser quality, resizing and deinterlacing) I set in Compressor to trim down the file size would be another.
I agree. Even though the file size is small, the difference between larger file is really small. I even can't notice.
Glad it's working – in a matter of speaking, that is.
You're right that the hardware is a llimiting factor. The CPU will determine the rendering speed. (Are you running a Core Duo or Core 2 Duo? Because the Core Duo actually isn't supported by FCPX.)
Again, one thng that would speed things up at lot is to shoot pregressive any time you have the choice - eliminating the de-interlacing step, as well as the need to use Frame Controls.
I have Core2.
I checked my Camcorder's setting. I don't seem to find progressive recording. I assume I can only shot interlaced pictures. I compared the output movies before I deinterlaced. A big difference. Before that, I can see lines when there're movements in the movies. Once I chose to deinterflace in Compressor, the lines are gone.
I think I need to encode the other movies too.