MPEG2 - VBR vs. CBR

[crjackson]

Until a day or so ago, I've transcoded all my DVD video using CBR settings.

Today (using CCE 2.66, and Vegas Video) I tried re-doing some of my AVIs using VBR settings.

Either I'm doing something wrong, or nothing is to be gained by using VBR.

Can someone tell me please? Is there any advantage in using VBR over CBR? If so, what is it?

I set my min. bitrate slightly below my target average, and the max is well above my target avg.

I see no quality difference. What am I doing wrong, or did I miss the point all the way around. BTW, it's sure a hella lot faster coding CBR with what I precieve as near same quality (actually vbr seemed a little less quality).

Please clue me in. I know I'm doing something wrong... Plus, my VBRs won't play on wmp, the CBRs play on everything.

[D_Knife]

well
VBR is supposed to look the same as CBR - except that is often smaller in final size
breaking down the name of CBR - Constant Bitrate and VBR is Variable Bitrate

if you have a scene that is fairly static (i.e. not much moving ) then VBR can save you bits by reducing the datarate - while CBR on the other hand remains constantly at the specified datarate and is in effect wasting bits

CBR can be faster than VBR - but you can do single pass VBR that's almost the same speed as CBR - however most people run VBR as 2 or 3 pass - each subsequent pass refines the bitrate allocations

As a suggestion for bitrates
I create CVD's almost exclusively these days
I'm using PAL 352x576 25fps
I use CCE 2.66 with 3-pass VBR
my datarates are - 300Min 1400Avg and either 1600 for normal work or 2496 for high motion work
Some people will say you should increase the max rate well above SVCD bitrate - but I can't spot the difference so I keep the rate down low
This gives me excellent results and works quite well for music video's and the like

What do you mean wmp won't play VBR???

[furball6969]

You will gain nothing from re-encoding a CBR AVI to VBR AVI.

You will gain from encoding an uncompressed format to VBR rather than using CBR.

There are lots of discussions on the topic. Try doing a search.

[C53248]

I set my min. bitrate slightly below my target average

That's why. The minimum has to be significantly lower, not just slightly. Otherwise your Variable Bit Rate just won't vary that much and will be pretty much constant, i.e. only slightly different than Constant Bit Rate. You have to let the bitrate get below average in some places in order to be above average in other places.

The advantage of VBR is that the data is allocated more effectively. By not wasting data on simple scenes, more data can be used on complex scenes. For a given file size, the VBR should look better than the CBR.

[puertorican138]

2pass Variable Bitrate(VBR): This means that your bitrate has an minimum, maximum, and average bitrate. On this one you can set the average to a fixed setting and have it either raise to the maximum you set during high motion scenes or fall to the minimum you set when there is little motion. This setting is good for figuring out the size of the file that is being outputted and to have it reach higher bitrates yet still maintaining a lower file size. under the settings of this setting is where
you would put in everything you need to. There is also a setting, "Enable padding When falling below the minimum bitrate", if checked. forces the rate of the movie not to fall under the minimum bitrate you entered. Also this setting will go over the whole movie first then at 50% it will start to encode it, and this takes time...Allot of time! Why would anyone do this you ask. Well I do this because I know approx. what the file size will be and I feel that I will get the most out of my movie. This is mainly for MPEG-2.

[jxsam1@yahoo.com]

As D_Knife puts it very clearly, you really do not gain much quality going from CBR to VBR.

IMHO, the quality could even be worse as you are allowing the encoder to change the bitrate based on the 'activity' in a scene. High action and constantly changing scenes will require a higher bitrate in order to capture all the changes in a scene. A lower bitrate which translates to lower memory allocation for capturing the scene it's encoding.

Why can the quality be worse you may ask on using VBR? Well, if your source is encoded or captured at 8 MBits per second at constant rate and you capture the same source but this time a one pass VBR with a max bitrate of 8 MBits per second, you will tend to notice that the VBR capture will tend to be worse in quality when a scene changes quite dramatically. This is because the encoder or capturing application when using VBR and reaches a relatively static scene, it uses a lot bitrate but if it suddenly encouters a big change with a lot of action in the scene, the application will need a little time to adjust to the "right" bitrate for the motion during that scene. This adjustment, can sometimes be visible when playback of VBR's. On the other hand, the CBR capture or encoding, since you use 8 MBits per second for the entire movie, better quality can be achieved.

Using CBR will create very large files as you are wasting bits by allocating tons of memory to even scenes that does not have a whole lot of activity or action. That is why people uses VBR. VBR will allow you to set some parameters such as what is the maximum bitrate you want to encode and what is the minimum bitrate and then an average. The way it picks the bitrate as I mentioned briefly above, it all depends on the algorithm of the application. To take this further, there are 2 pass, 3 pass, 4 pass and I've even seen 5 and 6 passes (overkill) for VBR.

What do they do when you run a multiple pass VBR encoding? Well, usually the first couple of passes, the encoder or application will just analyze the video stream to detect where the scene changes are or where the major jumps from low action to fast action sections are. Knowing this, the encoder and application can determine when it finally does the encoding, to adjust the bitrate before hand or right when the scene changes thus producing better quality. 2-pass is probably what most people use. The first pass is for analyzing and then the second for encoding. It consumes a lot of time. 3-pass as D_kinfe points out, the application can fine tune itself further on where to set what bitrate thus producing an even better result.