camerri

Have the ML stereo in the 2IS and have a question.

I want to create some DVD audio cds from MP3s that I have downloaded/ripped to my computer. Currently I make MP3 discs on CDs using compressed MP3s at 192kbs (about 10-11 full albums). My questions is this: If I use these same MP3s at 192kbs and stick them onto an uncompressed DVDA disc (about 9-10 full albums), would there be a difference in sound quality since they are both coming from a compressed music file, and why?

I want to say the DVDA sounds better to me, but that may just be my hopeful thinking. I'm sure it would be better to extract WAV files straight from the CD to burn to the DVDA, but that is more inconvenient and some disc are downloaded from iTunes.

Thanks for your help.

Robert_J

The medium the audio files are being stored on does not determine the quality of the audio, the source of the files determines the quality. Even if you are converting them from MP3 to true DVD Audio format, it will not help. You can't magically add in the information that was lost during the original lossy compression. Even if you rip .wav files from CD to DVD-A, it will not sound any different from the CD. True DVD-A discs are mastered from high quality digital recordings.

Also, a car is not a great environment for audio anyway. I have never heard DVD-A in a car but I don't think it would be much better than CD considering the poor quality of OEM speakers and electonics in all cars. Get some Mark Levison home amps and a pair of Eggleston Ivy speakers in a room with proper acoustical treatments. A quality mastered DVD-A should be a moving experience.

-Robert

camerri

Maybe I should have been more clear.

I am not expecting true DVD-A quality sound to come from a 192kbs music file. The file is what it is and I understand that. I also understand that a WAV file made from a CD still doesn't make DVD-A quality. However, when you burn an MP3 onto a CD, it stays compressed let's say 4MB size, but when making a DVD-A disc the file is 45MB uncompressed (at least from the original 192kbs file).

Does this make a difference in sound quality, that is, does uncompressing/converting the file for the player produce a better quality sound than letting the stereo uncompress/convert the file. I am simply wondering, should I make MP3 cds with 100 songs on them or DVD-A discs with 100 songs on them. My goal is to get the best quality while maximizing the amount of music I can put on one disc.

Hopefully that makes better sense. Thanks.

jtgs400

iTrader: (1)

Whether it's 4mb or 45mb they're the same. When that file went from 45 down to 4, it loss quite a bit of information. Even when you uncompress that file back to 45 the lost information can never be regained. They will sound exactly the same.

If you want the best sq, rip the songs into .wav files and make the dvd-a disc using those wav files. Atleast then the sq will be the same as the original cd but with the capacity of a dvd.

alk3997

An interesting question. Yes, given that everything is equal, the sound should not improve. But, I've found that not everything is equal in the comparison you mention. A review of the signal/data paths is helpful.

For DVD-Audio in the Lexus, the player has to read the disc (at up to approximately 9 mbps) the PCM counts are converted back to analog, they are sent through a filter (the higher the sample frequency, the less the filter should impact the sound) and then sent through the amplifier and then the speakers.

For an MP3 through the CD, the data is read, an MP3 to PCM conversion is done (with lossy results), the PCM counts are converted to analog, the analog is filtered (in this case the filters probably change the upper end a bit because of the lower sample frequency) and then the sound is sent to the amplifier and the speakers.

When you convert the MP3 to a wave file on your computer, you are essentially eliminating the MP3 to PCM conversion in the Lexus. You are doing that before hand. Since this is a lossy operation and done by different chips (Lexus) / program (PC) the resultant PCM could be different (likely is different). So, it is possible for the sound to be different.

Another are where you could have difference is the filters. The DVD-Audio filtering is likely higher quality than a CD filter.

Notice in none of above do I say it would be better. In the high fidelity world, the real test would be which version is closer to the *original* PCM file that then became an MP3 file. That would be impossible for me to say without files to compare.

BTW, in the old days of receiver comparison shopping, stores quickly found out that when your average consumer compared different receivers, most consumers thought that the receiver that was louder sounded better and the one with more bass sounded better.

Where that is relevant here is that if the DVD-Audio source is slightly louder, then there could be a perceived difference from the CD/MP3 source even though the files are very similar.

It just shows that comparison of two seemly exact audio can produce different results. Hope that helps explain what could be happening.

BTW, if you feel that the DVD-Audio is better, then continue to make DVD-Audio discs. You can also fit a lot more on a DVD-Audio disc than a CD.

Andy

camerri

Great answer! I knew it wasn't as simple as it seemed, but that's all I knew. You explained it very well. To me, the DVD-A sounds as though it has a little less low-end than the MP3 discs, with a little crisper (maybe louder) high-end. But again that is my perception.

...but then I would be back to MP3's again.

Thanks for the reply.

camerri

One more question. By what method would one test this?

Thanks.

alk3997

Good question. If I were to test this I would try two methods (depending upon money and equipment, of course).

1) Take a known 96kHz, 24bit audio wave (PCM) file and burn a DVD-Audio disc with it. Also convert the wave file to an MP3 on your home computer (choose appropriate MP3 bit rate to mimic the files you have). I'm going to assume a mono file. Then make appropriate MP3 CD and converted DVD-Audio as if the original wave file did not exist. This takes care of the test input.

2) Now to measure the output...Either 1) take the speaker output wires, convert to a line level and correct for impedance or 2) place a high quality microphone at a relatively close distance to the speaker. Using the chosen output, then record the original wave DVD-Audio, the MP3 CD output and the DVD-Audio output. BTW, if the speaker output is too complicated, finding a line level output from the pre-amp would be the next best method but you may miss some accoustics/DSP effects that way.

3) Convert all outputs to a wave file (96/24) (if not done already) and compare the files. When subtracting the PCM waves, you'll see any differences as a new wave file. To subtract all you have to do is to phase invert *one* file and then mix the two files together. The "winner" is the smallest difference to the original wave file.

The second method I would use is a double-blind test involving a set of people. You would need to make sure that no one (including anyone in the vehicle) could tell which source is being used. A panel of people would be then selected which would rate the output of both types during A/B comparisons. Requires a lot of friends and a good test setup in a car. Very subjective however.

Andy

tcho82

iTrader: (1)

Have you been successful in getting this to work? As far as I'm aware there is no way to get a burned DVD-A disc to work in the ML system. Correct me if I'm wrong.