Screen Capture Codecs for presentation slides?

[senze]

Hello,

can anyone tell me what's the difference between a Screen capture codec and a "normal" codec like the MPEG4 codecs?
For a video of presentation slides (e.g. transcoded from Powerpoint) should I use screen capture codecs or normal ones?
BTW, what is the difference between still image compression and the compression of presentation slides? None? Might it be the case that all the Motion Vector prediction stuff is useless for presentation slides, because there isn't any motion?

Best Regards

[mariush]

Screen capture codecs are usually lossless and are tuned to detect that some portions on the screen change less often than others and compress only the differences. They're often smart about detecting the content of the image and separating it into blocks of solid color, text, lineart, full color and they treat these differently.

General codecs such as the mpeg 4 codecs are lossy at various degrees (it's like making a screenshot and saving it as a very high quality JPG) and they're more tuned to analyze how humans perceive motion to compress certain areas they determine humans won't notice easily. Screen captures don't have a lot of motion except some small animations so some poorly tuned or badly configured codecs can generate low quality videos due to this. The codecs in general are not that good at maintaining the quality at sharp/sudden changes of the capture image, for example when you change from a black and white slide to a full color slide.
Also, due to the nature of the compression behind these codecs, sharp text is often hard to compress while preserving its quality - sometimes sharp, small letter size text gets blurry or ugly.

Of course, some video encoders such as x264 (which encodes video into mpeg 4 avc or also known as h264) have a lot of configuration options and they can be configured (even have presets) for such types of videos and they get very close to the quality of specialized screen capture codecs..

[senze]

Hi,

You said, the codec treats text, background, .... differently. Do you know a source where the processes are explained in more detail?

might you tell me on which settings I should look in x264 when trying to get some good settings for recording presentation slides, I have seen that there is a still image preset?

With the use of x264 for this purpose I have a problem:
When I try to use a program for capturing e-Lectures like camtasia or lecturnity I have different sources e.g. camera filming the person, mic recording the speech and the screen recorder recording the desktop. I'm already recording the person with the camera in x264 VfW, now when I try to simultaneously record the desktop with x264 I have to use the same settings as for the person.

Do you know a possibilty to record simultaneously with the same codec but with different codec settings?

[chainring]

I don't have answers to your specific questions, however I have been on the lookout for a really good (and free!) screen recorder program. Blueberry Flashback Express is very cool. Of course, they have a paid versions and those have some great features. The free one is quite advanced and is better than anything I've tried so far, even better than most paid ones I've tried. GDI mode (capture presentations), MPEG2 mode (capture video), scheduled recording, simple time based recording (record for "x" minutes), file size based recording (record only until file size is "x"), export to any codec that registers with Windows.

I'm quite shocked they're giving that much away for free.