Differences in quality between ffmpeggui and TMPGEnc AC3 encoders?

[redwudz]

This question came up in another post. I'm just wondering if some of you have used both AC3 encoders and have an opinion or some information about the differences in quality, etc.

[celtic_druid]

TMPGEnc should be better since the one in ffmpeg is not properly tuned. Not sure if anyone has bothered doing any actual tests though. Maybe over at hydrogenaudio?

[guns1inger]

Major difference - one is Dolby certified as compliant, hence the licencing fee, the other isn't.

Does it make a whole lot of difference to the home made punter - not really. I have commercial studio discs with very dodgy Dolby Digital audio (thanks, Warners) which I'm sure were encoded with certified equipment.

The other question to ask yourself is - will my source show up any differences ? If you source is MP3 to begin with, then the short answer is no.

[redwudz]

Good point about the source, guns1inger. Most of the ones I convert are MP3 and the quality has already been destroyed.

Thanks for that hydrogen audio site, celtic_druid, it looks interesting.

I couldn't find much in the way of comparisons between the two programs. Apparently the freeware AC3 encoders are probably all based on same original encoder. The payware encoders are licensed by Dolby and thus the cost. They also seem to have a more compliant AC3 output and probably better quality if you have a good quality source file. But for my purposes with MP3 conversions, ffmpeggui is probably sufficient most of the time. 8)

[Wile_E]

A problem with the TMPGEnc AC3 encoder, is that it doesn't have any parameters to set it up properly. From what I've read, for home use, most commerical dvd's should have the Dialogue Normalization parameter set to -31db. A proper encoder boosts softer sounds and lowering loud sounds, so everything is easy to hear in a home theater environment. I'm not sure about ffmpeg, but Vegas allows you to set certain parameters before encoding to AC3. Is it worth worrying about for home use? It's up to you to decide.

Here are some links that explain AC3 encoding.

Doom9 guide - How to properly encode Dolby Digital Audio(AC3).
Hometheaterhifi.com - Dialogue Normalization: Friend or Foe

[bobkart]

It seems that this could be answered quantitatively, at least for specific instances of encoded material (and target bitrate). I have some simple WAV file tools I've written, and using these I can create a WAV file that represents the differences between any two other (same-length) WAV files. So one could pick a test WAV file, put it through each of the converters, then convert those results back to WAVs, and difference them against the original. Then all that is needed is some way to count those differences and report them. That last piece I could build in an hour. (Of course such a simplistic measuring of differences is not likely to reflect real-world human hearing differences very well.)

But really, my original question was simply IF there where known differences in the quality of these encoders, and that's been answered clearly enough. Thanks guys.