I’m posting this only in hopes that it might benefit those who work with screen recorders–software that lets you record mouse movements and interaction with software (screen recording is to screen shot as still photography is to film). That means, primarily, other software trainers like me. If that’s not your line, this post might not interest you much.
My current project requires shooting lots of brief (2-4 minute) screen videos with voice overs. These eventually live in WebCT. And, to make things easy on the users, I want them in SWF format (the format created by Macromedia Flash). I use Camtasia Studio to do the captures and to render the SWFs. The problem is, while Camtasia does a fine job of capturing and exporting, it’s editing tools suck. Since I have experience with Sony Vegas, I do my captures in AVI format (rather than Camtasia’s proprietary CAMREC format, the only other option) so I can edit them in Vegas before spitting them back out and reimporting them into Camtasia to do the SWF exports. Vegas, good as it is, can’t render directly to SWF. It can import SWFs, so you can use them to create titles and such. But it’s a one-way street.
So you can see there’s a lot of room for things to go wrong and files to get really large. But here’s the best practice I’ve found so far. Though Vegas can export to a lot of audio and video formats (AVI, WMV, MPEG, Quicktime, etc.), I’m limited by what Camtasia can import (for extra credit, make a Venn diagram of that, upload it to Flickr, and link to it in the comments). I’ve done tests with AVI, WMV, and MPEG. And AVI is still the best. Mpegs are too big for no noticeable improvement in quality. WMVs, as exported from Vegas, are actually smaller than the AVIs, but they gig bigger when Camtasia turns them into SWFs.
But exporting from Vegas is only part of the issue, because the export to SWF in Camtasia offers a lot of export options. And tweaking them correctly makes a world of difference in the size and quality of the final product. So far, I’ve had the best results by setting JPEG compression to 90%, setting the frame rate to 15fps, and encoding audio to mp3 format at 22.05kHz, 40kBits/sec, stero. Any lower audio setting produces noticable high-end noise. Any more JPEG compression blurs the video more than I can tolerate. The biggest suprise was how low a frame rate you can get away with and how much it helps when it comes to file size. My original SWF export for a short movie (3:16, 712MB AVI) was 185MB. With these tweaks, it’s down to 43MB with no serious loss in quality.