Answers to most of your conversion questions

Avi to DVD and other video conversions, such as wmv to DVD, mts to DVD, mkv to DVD and more. 2 pass encoding, multi-core support, and always more files supported.

Moderators: Maggie, JJ, ckhouston, Phil, Forum admin

User avatar
VSO Team Member
Posts: 5913
Joined: Mon Apr 02, 2007 1:39 pm
Location: France

Answers to most of your conversion questions

Post by Claire » Thu Jan 14, 2010 9:39 am

Worried about quality?

What encoding settings for best quality output:
It is simple: use the "Automatic" as the encoding option, read the 2 pass section below to decide when to use 2 pass and the Image filters section for information on them.

With the Automatic option you can trust ConvertXtoDVD to provide best quality for you regardless of the size of the final size output.

If you want to understand why read below:

In the case of ConvertXtoDVD it is a misconception to believe that a DVD that is not filled up or written to the very edge of the disk is bad conversion or that the result is of low or bad quality.

ConvertXtoDVD encodes in a different way than most video converters. ConvertXtoDVD does not try to fill a DVD during each conversion. With the special converting engine of ConvertXtoDVD filling the DVD would not give better quality it would just fill up space. So there is no point filling the DVD if it is just to "fill" the DVD. In the case of ConvertXtoDVD it is perfectly normal that not all your disks are not filled.

Encoding method
ConvertXtoDVD uses an encoding method called CQ, contant quantization factor encode. Most users are familiar with constant bitrate (CBR) encodes and traditional variable bitrate (VBR) encodes. With a CQ encode size is not proof of quality. Only your eyes can tell you if the result is respectable or not. In other words ConvertXtoDVD is a quality based encoder which make an economic use of the bitrate with a very simple principle: when it's not needed to use higher bitrate, convertx will not use it.

For more explanations about ConvertXtoDVD encoding method please see: this post

The common mistake is to believe that BIGGER SIZE / BITRATE will produce bigger quality. THIS IS NOT TRUE IN ANY CASE.

For the hard headed I might as well make this easier for you and tell you now, that setting "Short projects" as your encoding option your output size will be bigger than with the other options listed, however---this does not mean your result will be any better at all, in fact it maybe worse. Once again, size is not proof of quality in the case of ConvertXtoDVD a better use of bitrate might have been done by using a different encoding option.

Encoding option advisor
The encoding option advisor makes a decision to obtain the best visual quality, and doesn't care to fill up completely the DVD.

The question to ask ourself is "The encoded result is looking good or not"? If it's looking good with the currently used size, why it should take more? As long as you answer NO to the second question, then it's pointless to say that it don't take enough space.


Keep in mind: ConvertXtoDVD cannot invent quality where it does not already exist. You cannot convert a mediocre file and expect to have even better quality after the conversion. It's common sense for instance to think that if the source file used fit in 1 or 2 CD, there is no reason why it should take a full DVD -5 or even DVD-9 when converted, no matter how efficient the input format is.

Image filters
If you are worried about quality, first change the image filter used in the settings, under the Video Processing tab. The default setttings is at Linear (VSO) which is a good filter - optimized for speed for a faster conversion but other filters will provide a better quality but can increase the conversion time.
See this post for more information about image filters, and image filter comparisons + and this one, often suggested is the filter "Lanczos".

High Definition Content + ConvertXtoDVD
HD content to DVD, a lot of MKV files (amoung other files) contain HD content and users are surprised about how big the difference in size the result is after conversion with ConvertXtoDVD. You must remember that the simple fact of resizing the image to DVD standard (720 x 480 NTSC, 720 x 576 PAL) can reduce the file size significantly ---> (1920*800)->(720*428) almost dividing by 3 times the size. Also keep in mind the MKV or other file type container often contains content that will not be processed by ConvertXtoDVD, extra audio and subtitle files for example.

Unless you have a blu-ray player, blu-ray player, + BDRE disks, or a media player and a HD TV you will not be able to watch your HD files on TV in their original HD resolution. VSO is working on HD solutions....

2 pass
2 pass will not improve quality enough to be worthwhile if you use the Automatic encode option except for projects long enough that the DVD cannot be kept from being filled. So you might get some improvement with 2 pass for those longer than about 170 or 180 minutes on a DVD5 disc. But any improvement is reduced to nearly zero for projects as long as 5 hours.

Consider keeping projects below 240 minutes or putting them on a DVD9 disc instead if you are really concerned about quality.

Long projects
Some users want to put a lot of TV series episodes on a DVD5 disc and are not very concerned about quality. You can use Half D1 resolution instead of Full D1 for projects longer than about 5 hours or SIF for those longer than about 7 hours to prevent block pixilation. Play the converted files on your computer before burning them to see if quality is acceptable when experimenting, and consider that conversions to low resolution like Half D1 and SIF will not look good on high definition TVs. You may want to put fewer episodes on a DVD5 at Full D1 resolution for better quality, or use DVD9 if it is important to put as many episodes as possible on one DVD.

Read ... 13008.html if you want a more detailed explanation.

User avatar
Jacques (VSO)
VSO Team Member
Posts: 70
Joined: Fri May 11, 2007 9:31 am

Re: Answers to most of your conversion questions

Post by Jacques (VSO) » Mon Feb 07, 2011 10:39 am

There are constantly reoccurring questions on the forum about the final size created by ConvertXtoDVD.

1) Why doesn’t the conversion fill the disk?
Many factors may cause this, those 2 are the most common :
* The total duration of the video is small - less than 60 min. of video will NOT fill a DVD-5, no matter the source. This is DVD Specs.
* Re-encoding video from source of almost the same screen resolution, and with similar encoding algorythm (mpeg-x) is very efficient without loss of quality. For instance, the average 720*480 MPEG-4 350 mB source of a TV episode will most commonly take less than or up to twice the original size (700 mB). So if one want to put on a DVD-5 4 episodes of such source, it'll produce a DVD of about 700mB * 4 = 2.8 gB, leaving more than 1 gB free on the converted DVD. This is perfectly normal and is not a sign of reduced quality. It's just that the encoder take all the space it need for the best quality, and in this case there is still some space left on the target.

2) How come the output size cannot be known in advance?
ConvertX evaluate the output size based on information that can be obtained quickly from the input files. Those information are often accurate enough but sometime only an extensive study of a source file may provide a good answer. So, output size is evaluated by convertX, but it is a higher limit recommendation, and the final result may be less than what is set.

3) How come the desired final output size cannot be a setting entered by user ?
The desired output size is a setting entered by user (Target size), however because the most important is to produce a conversion that will not be bigger than the desired output size and that will be playable on every DVD reader (so match the DVD specs), and because of what is explained in #1, convertX may produce less or up to the required size. Once again, if ConvertX produce less size than the desired final size, it is always to match the best quality within the DVD specs.

4) How come, depending on the number of file converted the size of each file is different?
On a Episode DVD, ConvertX adapt itself to the remaining size on disk after each episode. So if one episode took less size than expected, it allocate more size for the remaining episodes. So, the allocated size for each episode will depend on the total duration on the DVD - if you put 4 hours, each episode will have half the space than if you put only 2 hours.

In all encoded size related question, it is often believed that if the output will be bigger, the quality will be better. This is not true on DVD. The specification that makes a DVD readable on any player introduce an absolute limit in bitrate that we cannot overpass. The encoding choice of ConvertX priviledge quality over size accuracy, which make convertX take decision that will improve quality, but may lead to produce less than available size.

We are always compared to other solutions that uses FFmpeg or not and that produce bigger output size for the same files converted. We cannot speculate on the choices made by our competitors in matter of quality, however we can say that our primary choice is the quality of the output video, and for that we have carefully tuned FFmpeg library in our own way. If there is still space left on the target size, we don't fill it with data that will not increase the quality and may even produce DVD's not compliant with DVD specs.
VSO Software Team Member