By Gunnar, Tuesday, December 12th, 2006
Categories :
Speechi No tags for this post.
The majority of the video cameras on the market will provide you with an initial format in AVI, MPEG 2 or MPEG 4 (format produced by our camera Sony). It is then necessary to convert the format into FLV. The step is identical to that which should be done to insert a video FLV on a blog. To do this, there are several free and commercial tools available. We use a product called Riva FLV encoder. It is free and it accepts the formats MPEG 2 and MPEG 4.
The adjustments that work best for the Web are as follows:
- Size of film: Today 240 X 180 should be a minimum and it seems not very useful to go beyond 640 X 480. The “average” format on the Web is of 320 X 240.
- Frequency: 15 images a second at least, 25 images a second at most, 20 is a good compromise. Do not modify the frequency of the original video. If you do, it can degrade the quality.
- Band-width: For a video of good quality, 360 kbits/s is sufficient (beyond that, the viewer must have a really fast Internet connection to be able to view the video well). For videos with low resolution, 120 kbits/s can be enough (low limit).
- Band-width of the sound: Use maximum 48 kbits/s, except if you have a sound recording of excellent quality and if the quality of the output sound is important to you. For normal quality, 24 to 32 kbits/s, mono, should be sufficient.
Other alternatives to Riva (which remains our best choice “by default”, but which does not support all the “codecs”) are
- Sorenson Squeeze 4 (best quality, the most functions, but expensive and for Speechi, the additional functions are not very useful since Speechi takes care of almost all of them)
- Flash MX 2004 Pro (more complete and more practical than Riva)
- ffmpeg. This is the program which is “behind” Riva. For the “expert” users of Linux only.