Question about matrices

[Insane2986]

I notice my matrices give ranges. Ex: High_High(4000+), High_Low(3500-4000), Medium_High(3200-4000), etc. Are these ranges for the average bitrate? Or is the lowest bitrate supposed to fall within this range too?


Also which matrix should I choose for this?

- CCE 2.70.2.4 encoder selected.
- VTS_01: 2,604,449 sectors.
-- ANGLE and/or INTERLEAVING is present.
-- Scanning and writing .D2V & .AVS files
- Processed 71.7MB ILVU section...
- Processed 153.2MB ILVU section...
- Processed 343.5MB ILVU section...
-- Processed 237,737 frames.
-- Building .AVS and .ECL files
- Reduction Level for DVD-5: 78.9%
- Overall Bitrate : 3,752/3,002Kbs (if 3,002 is typical, what is 3,752??)
- Space for Video : 3,633,070KB
- HIGH/LOW/TYPICAL Bitrates: 4,892/903/3,002 Kbs

Thanks in advance

[l8nights]

my understanding is :the different matrices are adopted by your encoders when a segment falls into one of these catagories, high high,high low, low,very low.

[Rippraff]

Are these ranges for the average bitrate?
Yes.
(if 3,002 is typical, what is 3,752??)
Take a look here (http://forum.doom9.org/showthread.php?t=100919). ;)

Cu Rippraff

[archaeo]

Also which matrix should I choose for this?



I wouldn't answer that question based strictly on bitrate...
It's a starting point, but is not the only thing involved with deciding which matrix to use. Two other things I find important are:
1) the source characteristics - what level of detail, action, color, shifts, fades etc.., does the film have? In other words is it a rather simple film or is there 'complexity' to it? What's the overall quality of the image like? If this film relies on a lot of detail for it's best viewing, you'll want to be aware of that.
2)The other issue I find that that helps me to decide is to extract the original matrix from the source (using Rockas Matrix Editor) to see what they used to create the dvd. This matters for the same reason as the first point because matrices are designed to hold/lose certain levels of hi and low frequenccy detail in order to gain compressability. If they used, for example, the Fox Home Entertainment matrix (which holds a lot of detail) and this movie would benefit from retaining that, I would tend to choose a matrix that balances compressability with holding that detail. On the other hand, if they have used the standard mpeg matrix it may not be worth it to try to salvage detail that has already been lost.

At a 78% compression and with 3000 average bitrate, you may benefit from looking at these things before deciding. My own experience has shown me that matrices don't always make a huge difference in many cases - but I have run across enough cases where they have mattered and have helped me to keep as much quality as possible, compared to if I had just chosen the defaults. There are some threads around here that cover these in more detail than I can, and they may help you get beyond just 'bitrate based' decisions.

[SDragoon]

My (somewhat) related question...

What bitrates do the categories of Main Feature, Low, and Very Low matrices in DVD-RB represent?

[jdobbs]

When the average bitrate drops below 3000Kbs the "Low" matrix is used. When it drops below 2000Kbs the "Very Low" matrix is used.

[SDragoon]

When the average bitrate drops below 3000Kbs the "Low" matrix is used. When it drops below 2000Kbs the "Very Low" matrix is used.

Thanks!

[dialysis1]

When a movie is to be compressed at around 90%, would it be wise to use the same matrix that was used to make the movie?

[jdobbs]

I'd use the encoder default.

[archaeo]

I'd use the encoder default.

Is this because at such a low compression rate, you feel matrices won't matter?

[jdobbs]

Most encoders have been designed around their default matrix -- so IMHO it's better to use that one for most sources. The exceptions are when you have "special" sources like animation, computer graphics, low bitrates, etc.