calculate quality of a movie??

[Cazz]

Is it possible to calculate what quality (in %) going to be??


For example If I have a movie that is in some very high quality and it is in 4 min and I going to convert in H264 and most 40 Mb

It is maybe impossible to calculate that?

I just wonder if that is possible to know how much quality is going to be before I start the convert and wait many hour to see the result

[fitch.j]

i'm not sure how a Percentage number would help you in this. If you found out it would be 64.87% for example, this figure means nothing, you are still no wiser to the outcome quality of the video.

My suggestion would be to render a small portion, a few seconds or so, in order to view the quality of it, then you can much more accurately judge your quality before rendering an entire file.

[Cazz]

hmm ok, well yes I can do that

[jagabo]

Some sources compress better than others. So there is no way of saying exactly what the "quality" will be after compression. See the examples in this post:

https://forum.videohelp.com/topic357646-60.html#1903520

Some programs report a "quality factor" but this only tells you how many bits per pixel are used on average. This isn't a good predictor of visual quality. The clean video in the above post rates a Qf of 0.049 in GSpot whereas the 2pass video rates a 0.172. But the clean video has far better image quality.

[guns1inger]

When you have been doing this for quite a while you start to get a feel for what sort of bitrates are required for different videos. It is not a mathematical or precise measure, but simply an understanding of how these things work.

The universal rule is simple and very precise.

Filesize = Bitrate X Running Time.

That is the easy part.

However there are lot of factors that factor into deciding if that filesize is big enough, or if a file size target should even be set. Content is a big part of that. Fast motion, lots of grain and noise, lots of small details or lots of shadows - these things all require a lot more bitrate than clean, well lit, static footage. Then there is resolution. A lower resolution means each pixels gets more data from the bitrate, and therefore greater detail. Higher resolution footage has to spread that bitrate thinner if fitting into the same file size.

Until you can tell with reasonable accuracy simply through experience, encoding a sample is the best way to go.