Maximum values: Which is the best converting method?



Is there a difference on selecting with
-Constant Average Bitrate “2048KBits/s” for all the “parts” of the DVD or
-Constant Quality “125” for all the parts.

It is obvious that I want the maximum quality possible (the size doesn’t matter)…so selecting the maximum in both the converting methods (Constant Average and Constant Quality), which is the best in terms of quality?


Usually variable bitrate gets things done better. It uses more bitrate when it needs it and less when it doesn’t (for example in still images), so that way some bitrate doesn’t get wasted. If you check the average bitrate it checks how much it’ll need before hand. The constant quality option doesn’t.


Thank you so much for the answer and the explanation :wink: .
Now I’m going to try this interesting program with the methods you suggested :iagree: .


Reading the info on RatDVD official website it seems that to gather best quality (for unspecified file size) it is better the “Constant quality encoding”.


[B]How are the quality settings working?[/B]

The quality settings depend on encoding mode and are different for “Constant Quality” and “Constant Average Bitrate”.

In “Constant Quality” mode the video encoder uses an adaptive quantization scheme which uses a psycho visual perceptive model so that it quantizes the most where you’re the least likely to see it. The “average quant” generated by this model is what is wired to the quality setting parameter (inversed to generate an intuitive user experience). The closer to 120 (The closer the average quant is to 0) the more accurate will the pictures be and the better it’ll look obviously the bitrate will climb too. It’s a constant quality encoder and this means that it’ll carry through on the quality setting no matter what bitrate that might produce. Constant quality transcoding is advantagous quality wise over rate control schemes.

In “Constant Average Bitrate” mode you can specify the average video bitrate and with it the size of the resulting ratDVD. In this mode the bitrate of the encoding is constantly controlled and if the bitrate needed is different than the target bitrate, it is slowly adjusted to meet the average bitrate target and with it the predefined size. This adjustment is made slowly overtime to make sure that scenes that need a higher bitrate will get it. However, this slow adaption to unknown video material can lead to not meeting the target size exatly in case of special video content or very short video material (which shoudln’t happen if you convert from DVD).

[U]Quality wise it is suggested to use “Constant Quality”[/U] encoding but in certain cases (e.g. you want to put a DVD an a memory stick) you may want to use “Constant Qaulity” to meet a given size.

[B]Can the file size of the .ratDVD file be estimated in Constant quality encoding?
What is the typical file size/compression level from DVD as source? [/B]

No! If you want to estimate the destination size you have to use “Constant Average Bitrate” encoding.

Instant “Constant Quality” mode every frame in the video gets the bitrate it needs. This ensures the highest possible quality but means that the size of the converted video greatly depends on the video material. Interestingly there is almost no relation to the original DVD size - only the quantity of frames in the original is of some meaning.

[U]“Constant Quality” is the default transcoding method because I am a fan of the highest possible quality[/U] and compared to DivX, XviD, etc. in ratDVD there is no need to fit the file more or less exactly to a CD - because when it gets played back in a DVD player it gets blown up to a DVD anyways.


am… well… I guess I wasn’t setting the constant quality high enough then. I used to set it at 100 and the quality was horrible.


By the way…I used the maximum value of Constant Quality (125) of the first DVD of Dream Theater - Live at Budokan (7,52GB) and the result is a 4.25GB file (just right for a DVD-R).
Then I compressed DVD 2 (6.65GB) with Constant Average Bitrate at maximum settings (2048KBits/s) and the result was a file of 1.76GB.
It would be more interesting to make a comparison with the same DVD (I’ll do it tomorrow), but it is interesting to see how little is the file in Constant Average Bitrate even using the best setting possible.
Tomorrow I’ll compress the DVD1 in Constant Quality (at 125) to see what is the difference in terms of visual quality.


It depends on what DVD you’re coding. For me to code a 6.8 GB DVD from GITS-SAC at 1500 Kbps it made a 1.88 GB file. That’s more than your bigger DVD at better quality. It deppendes on how much movement there is and so on. For me at 1500-1700 Kbps it’s more than enough. I can’t barely tell the difference with the source and I can fit two full DVDs without taking anything out in just one 4.3 DVD±R.


Yes you are right.
A compressed DVD of for example an action movie with a lot of changing scenes, explosions etc…will be larger of a comedy movie because of the complexity and more information of the image.
The concert I’ve compressed should have less image complexity of your Ghost In the Shell –SAC, and so even if the original DVD is larger at the end after the compression it will be smaller.


Now, the only question that I can answer only tomorrow is: is the larger compressed 4.25GB file much better than the smaller file created with Constant Average Bitrate? Is it worth the extra space utilized?


Original DVD size —> 6.65GB
Compressed with Constant Quality (at 125) —> 2,69GB
Compressed with Constant Average Bitrate (at 2048KBits/s) —> 1,76GB

From what I can tell there is no significant visual difference between the two file created with the 2 methods.
So to to save space it is more convenient to use the Constant Average Bitrate, but if you absolutely want to have the best quality (even if you probably won’t notice the difference) stick with the Constant Quality.

By the way…great program :stuck_out_tongue: