Sansa Clip+ vs iPod Touch Detailed Measurements & Review

[Enigmatic]

Originally Posted by meetooman

ABX is very easy to understand.

Now if you imagine you can create 2 different streams from wave-mp3/128k and wave-mp3/192k. Comparing them via ABX may be insane (absurdity).

If ABX was very easy for you to understand, you would never attempt this.

But am I that sure it would be completely meaningless without trying?

It is meaningless.

You all can say this software is useless, but I will say how it leaves me helpless without closing doors.

We are not saying that this software is useless. We are saying that this software is useless for examining the quality of the files from lossy encoders. And it was explained to you why this is so.

[NwAvGuy]

I think I can sympathize with meetooman's point of view. It's certainly reasonable to wonder if there's an objective way to evaluate lossy audio compression. Unfortunately, there isn't.

Ultimately, it's what is removed, not how much is removed, that defines the subtle differences between high quality lossy CODEC's and their settings. And, to evaluate the "what", you need a human brain doing the evaluation. And, ideally, a whole bunch of human brains (i.e. a listening panel or group) to establish a consensus.

Audio DiffMaker can only tell you the "how much" which, I suspect, largely correlates with the compressed bit rate when comparing CODECs. So while it might be interesting on a few levels, it doesn't tell you which CODEC sounds best.

The good news is anyone can do their own CODEC listening tests with Foobar 2K and the ABX plugin. You can build a Foobar playlist of some of your favorite most revealing musical tracks representing the various genres you listen to. And you run those tracks through several different encoders or the same encoder with different settings.

You can then compare the lossy files to the uncompressed versions and get your ABX score using your best headphones. If you find a lossy encoder you can't distinguish from the original tracks, you've answered your question using a relatively worst case selection of music.

Finally, I would humbly suggest this is getting off topic for this particular thread. Another thread in a different forum here, or another site (such as HydrogenAudio) might be a better place to discuss CODEC evaluation.

[a_tumiwa]

@NwAvGuy: i think your ipod touch 3G graph is unloaded test, could you pls create a graph using a 16ohm load like what dfkt always did?? i think its more accurate? i feel that itouch 3G is not as flat as sansa clip, its make me suprise when i saw the graph.

sorry for my bad english :P

[WalkGood]

@ a_tumiwa IIRC there's a resistive 15 ohm load used in the tests.