Video performance requirements analysis?

[Kendall Shaw]

Part of the question in the first post is more or less how does one do benchmarking. Beyond that how do you interpret the results and use the information to point out what you must change to correct the problems.

Different players rendering the video differently would also depend on hd ram etc. A player rendering the video more efficiently than another is a matter of how it performs in relation to those factors.

[Walter Soyka]

[Kendall Shaw] “Part of the question in the first post is more or less how does one do benchmarking. Beyond that how do you interpret the results and use the information to point out what you must change to correct the problems.”

I think your insistence on a general solution to this problem makes it practically unsolveable. As Chris pointed out, I completely ignored the software stack in my basic list of failure modes. This adds dozes, hundreds, maybe thousands of variables.

I think it’s much more fruitful to discuss the actual failure conditions you’re seeing. Of the zillion possible failure modes, there probably only a few practical ones, and addressing those will cover very nearly all playback cases.

Walter Soyka
Designer & Mad Scientist at Keen Live [link]
Motion Graphics, Widescreen Events, Presentation Design, and Consulting
@keenlive   |   RenderBreak [blog]   |   Profile [LinkedIn]

[Kendall Shaw]

I don’t usually think of problems in terms of being solvable or unsolvable. In this case, when I talk about knowing what I am doing, I don’t mean total awareness. I don’t think that ever happens. I mean knowing what I am doing to some subjectively significant degree.

The choice is not between blindly dealing with each crisis in isolation vs. solving all possible problems. I am seeking a trajectory that is in the problem solving direction.

There is a fundamental difference between the two approaches. In some contexts the difference is obvious, more and more it is less obvious in relation to software because the world largely switched from learning concepts to imitating procedures around the time of y2k.

It’s expected that a doctor will know some signs of a heart attack. The analogy would be knowing some signs of problem video.

It’s not expected that the doctor knows all possible signs of a heart attack. Similarly, it’s not reasonable to expect to know all possible factors that can result in problem video.

If the doctor can only deal with individual cases in isolation, it doesn’t work well. The patient is grasping at his chest, let’s try slapping him on the back. Oh, that didn’t work. Let’s try submerging him in an ice bath, etc. etc. The number of times this results in a dead patient might suggest that this approach is less than optimal. It could lead people to seek general knowledge that can make one prepared to deal with situations that have not yet occurred in a way that is practical.

The switch from concepts to imitation can be seen in technical documentation which used to be reference material but is almost entirely random lists of procedures to imitate in order to perform some task.

For each case in isolation, it requires less effort to follow the procedure than to learn a concept. In other areas outside of the software industry the problems with this idea are too obvious to even be debatable. We teach 4 year old kids the value of learning abstractions.

There are some related aspects to the trend that are good. Extremes in either direction are not practical.

[Walter Soyka]

If you’d like to learn the concepts underpinning compression, there are a number of books that cover the area in depth. A quick Amazon search on “video compression” will give you a number to choose from. I read Ben Waggoner’s book a long time ago; it should still cover the basics, but newer books would have better coverage of the current landscape.

Walter Soyka
Designer & Mad Scientist at Keen Live [link]
Motion Graphics, Widescreen Events, Presentation Design, and Consulting
@keenlive   |   RenderBreak [blog]   |   Profile [LinkedIn]

[Kendall Shaw]

Thanks. I just glanced at it (the ben wagonner compression book) on safari books and it looks very good.

[Ivan Myles]

Just getting caught up on this thread and I think it wound up in a good place. Ultimately, the best way to understand how different variables affect the end result is to 1) develop a stronger understanding of the compression process, 2) become familar with encoding guidelines/practices, and 3) perform systematic tests. Encoding is well-suited to testing because it is easy to vary one parameter while holding others constant. Over time, compressionists learn how to improve not only the encoding settings, but also upstream activities like editing, compositing, colorization, camera movements, et cetera to ensure smooth playback.