A little on compression....
Each frame of a video (there are 29.97 per second) contains 640 by 480 pixels or 307,201 pixels in total per frame. Each pixel has has a color assigned to it which requires 24 bits of information per pixel.
This is equal to 3 bytes per pixel, which adds up to 115,200 bytes or the better known 900 KB (kilobytes) per frame of space on your hard drive. That means that uncompressed, 1 second of video can be as large as (29.97 x 900) 26.3 MB That translates to over 1.5 GB (gigabytes) per minute or 92 GB for an hour movie!
Compression allows us to reduce the size.
Compression basically removes redundant data from a file to reduce its size. For example rather than having a row of white pixels defined by each pixel with each being 24 bits, it can be defined by the location of like pixels in an area, their placement and color .
Two types of compression exist. Lossy and Lossless. Lossless guarantees that the compressed file is exactly the same in quality when compressed and uncompressed. One Lossless file format used often in printing work is TIFF, it only compresses exact pixels. It saves some space, but not as much as we need for video.
Lossy compression (or encoding) of a video allows the size to be MUCH smaller. The compressions used in Video such as MPEG not only compresses adjoining like pixels in a frame, but also compresses pixels which are near-like in color and location. It also does this between multiple frames in time, by defining like and near-like pixels by total number, their color, their location... and their duration over adjoining frames.
You can adjust the quality of Lossy as well. Basically highest quality, only compresses more like matches, lowest compresses higher deltas of color.
You have to weigh the trade offs on capture and compression or encoding. Encoding the term used for compression in video, so we'll use encoding, here on out.
Capturing and Encoding with XLR8...
If you choose the XtraView or ProView USB raw un-encoded choice, your video will loose zero quality, but will be very large. You must also be sure that your system can handle the throughput to the hard drive of 26.3 MB per second. To test this, simply try to copy a 30 MB file on your hard drive. If it happens in an instant, you may be okay. If you get a progress bar for a couple seconds... nope, it won't handle it. You will drop frames.
If you choose MPEG, H.264... both are very CPU intensive. Since they are great at compression, they wlll reduce the strain on the drive, but can tie up a slow CPU, because it is trying to do the encoding in "real-time" without dropping frames. The Core 2 Duo systems will most likely be okay. Older systems though, will need to test. Just, grab a few frames of high motion video and check the results.
We have found the best trade-off is to go with Photo JPeg for capture. The medium setting delivers excellent and quality, takes very little CPU horse power, delivers great compression, and handles each frame independently (unlike Motion JPeg).
Video Impressions and iDVD will handle the JPeg encoding fine for editing and creation. iMovie HD, when you import, or simply drag and drop, will re-encode to DV, which it uses for most of its output as well.
The output from any editing application is dependent on its use. When iDVD is burned it will re-encode to the formats required for DVD. Video Impressions or iMovie HD will re-encode for iPod, CD or whatever format you choose.
Our XtraView and ProView USB manuals have a detailed section on all of the encoding choices available in QuickTime.
 Attachment: XtraView_Man5.pdf (Downloaded 123 times)
____________________
Gary Dailey
Daystar Technology
http://Daystar-Tech.com, http://Daystar-Store.com
email: gary@daystartechnology.com, chat: garydailey@mac.com
|
