Some basic Audio conversion question

[efiste2]

Hi, apologies for being a lamen, but I have a basic question, im coverting FLAC files to M4A so they play on my iPod in the best quality possible. I use the vey good AVS Audio Converter and set it to the following.

sample rate 44100Khz

Bitrate 512Kbps

However the Generic setting of BEST QUALITY is 256Kbps.

Is it correct to say the higher the KBPS the better the converted file will be ?

What is the most important factor in converting audio formats?

Thanks for any advice !

[hello_hello]

Is it correct to say the higher the KBPS the better the converted file will be ?

In theory yes, the higher the bitrate the better the quality, but for the popular encoders AAC tends to become "transparent" at around 128k, so going too much higher probably won't provide a quality difference you can hear. 99% of the time. 512kbps is, I think, the maximum bitrate allowed for stereo AAC.

I know nothing about AVS Audio Converter or the encoder it uses, but the commonly used AAC encoders all have a quality encoding mode with a variable bitrate as MP3 encoders do. You specify the quality and the bitrate varies as required. In theory you use more bits where you need them and less where you don't so it's more efficient. VBR audio with a 128k average bitrate should be higher quality than 128k constant bitrate. I have no idea if ipods support variable bitrate AAC. Most devices do.

These are the popular AAC encoders. They're all command line encoders but free encoding GUIs generally support one or more of them.
http://wiki.hydrogenaud.io/index.php?title=Apple_AAC
http://wiki.hydrogenaud.io/index.php?title=Nero_AAC
http://wiki.hydrogenaud.io/index.php?title=Fraunhofer_FDK_AAC

The Nero encoder can be downloaded from Nero's website, the Apple encoder requires files you need to extract from the itunes or quicktime installers, FDKAAC can be found on the internet and there's a front end for the non-free Fraunhofer encoder commonly know as FhGAAC. It requires you to extract files from the latest Winamp installer.

[newpball]

Is it correct to say the higher the KBPS the better the converted file will be ?

In theory yes, the higher the bitrate the better the quality, but for the popular encoders AAC tends to become "transparent" at around 128k, so going too much higher probably won't provide a quality difference you can hear. 99% of the time. 512kbps is, I think, the maximum bitrate allowed for stereo AAC.

The problem is often the sources used for testing. Sound engineers* nowadays poison their sources with all endless 'improving' plugins, limiters and compressors. The original is like a wild flower while the results produced by sound engineers is something incomparable:

Before:


After:



* { yes there are exceptions, surely there are sound engineers who care about fidelity first, but they are hard to find }


Make a 96/24 live recording with some matched Earthworks QTC50 and now see if you can tell the difference!

[hello_hello]

Is it correct to say the higher the KBPS the better the converted file will be ?

In theory yes, the higher the bitrate the better the quality, but for the popular encoders AAC tends to become "transparent" at around 128k, so going too much higher probably won't provide a quality difference you can hear. 99% of the time. 512kbps is, I think, the maximum bitrate allowed for stereo AAC.

The problem is often the sources used for testing. Sound engineers* nowadays poison their sources with all endless 'improving' plugins, limiters and compressors. The original is like a wild flower while the results produced by sound engineers is something incomparable:

Why the necessity to illustrate everything? It's more annoying than typing with the caps lock on.

Make a 96/24 live recording with some matched Earthworks QTC50 and now see if you can tell the difference!

If you're trying to claim there's an obvious quality difference between 96/24 and 44.1/16 that'll deplete any taking newpball seriously stamina I had built up in a single hit. Until someone's able to prove otherwise I'm keeping those claims in the same box as the claims regarding the quality differences according to the brand of hard drive an audio file resides on. It's quite a testimont to how well the brain can convince you of something when you're expecting it to. This stuff is so funny it brings tears to my eyes.

http://www.enjoythemusic.com/hificritic/vol5_no3/listening_to_storage.htm
Harris thought the Hitachi sounded very ethereal, almost out of phase, and rated it lowest; the Seagate was sharper with a more thumpy bass, slightly brighter with a slight tendency to sibilance. Both lacked much drive in presenting the Madonna track, and were certainly 'mushy' compared with the best sound quality we'd heard from the QNAP stable.
Drive three (a solid state type) gave a far from subtle shift in tone and soundstaging. I thought that here this Kingston SSD spread the stage wider, could really pull apart the multi-track layers, and certainly led in blackness too, sounding agreeably quieter than it had any right to. Yet there was also a dull flatness to its presentation, even a graying of timbre.

Oh.... The 96/24 vs 44.1/16 thing.....

The Emperor's New Sampling Rate

Audibility of a CD-Standard A/DA/A Loop Inserted into High-Resolution Audio Playback
PDF: http://drewdaniels.com/audible.pdf

24-Bit vs. 16-Bit Audio Test - Part II: RESULTS & CONCLUSIONS

Of course for some people..... ah...... that's how inventing imaginary people in order to contradict them works..... I get it now.... for some people pesky stuff like evidence should be ignored, and often they believe the very act of testing influences the results of the test, and an ABX test is like the audio equivalent of a double-slit experiment, because you know audio sounds better when you're not listening to it, but the act of listening makes impossible to tell, and apparently some people even believe ABX tests operate under a unique variation of the uncertainty principle, where the more precisely you try to determine the accuracy of someone's hearing, the less accurate their hearing becomes.

Anyway, the recent lossy codec listening tests I've seen use quite varied types of audio samples, because.... and it may surprise you..... the idea is to test the ability to encode vastly different types of audio, and not just a specific badly engineered type that won't sound any different after it's been encoded.... or however you're imagining it works..... because that'd obviously defeat the purpose of testing in the first place and therefore it'd be monumentally stupid. They also tend to perform listening tests at fairly low bitrates these days (ie 96k), because for AAC at least, once you go much higher it starts to become too hard to reliably test, and I suspect that'd be largely due to the transparency thing I mentioned.
http://listening-test.coresv.net/results.htm

* { yes there are exceptions, surely there are sound engineers who care about fidelity first, but they are hard to find }

Apparently caring can effect your hearing too, even if you're a fidelity loving engineer. It makes me laugh and laugh.....
Rick Rubin Explains What Mastering for iTunes Means
Of course, that's not what mastering for iTunes means officially, but it was his method at the time. Still, someone tested his "night and day" theory and called fidelity caring bullshit. Maybe null tests operate in the realm of quantum mechanics and that explains why they're not valid either, yet fidelity caring sound engineers know better. As does newpball, of course.

Some more fun reading. Lots of information for you to ignore.
24/192 Music Downloads
All signals with content entirely below the Nyquist frequency (half the sampling rate) are captured perfectly and completely by sampling; an infinite sampling rate is not required. Sampling doesn't affect frequency response or phase. The analog signal can be reconstructed losslessly, smoothly, and with the exact timing of the original analog signal.

If you're careful, you can completely ignore the link on that page to a video demonstrating the conversion from analogue to digital and back again. I thought I'd warn you in advance in case it doesn't agree with your opinion and you need to pretend you didn't see it.

Footnotes
As one frustrated poster wrote,
"[The Sampling Theorem] hasn't been invented to explain how digital audio works, it's the other way around. Digital Audio was invented from the theorem, if you don't believe the theorem then you can't believe in digital audio either!!"