You know you're an audiophile when... - Page 316  

Have not heard that before.

 

Or, depending on the age of the iPod, medium impedance headphones. My SRH440 hisses very faintly with my 5G iPod. Granted, it was new in 2005, and I imagine they've improved since  then, but it's still notable.

The iphone 4 do hiss a bit

Clipping is TERRIBLE! I purchased 2CELLO's album after liking what they had on YouTube and was severely disappointed with the poor mastering of the album. I don't even listen to it any more because the SRH940 is so revealing.

I tried playing it through the car speakers one day just to see what happens and my mom even acknowledged the terrible "recording" since the speakers made a very noticeable buzzing distortion noise.
Edited by miceblue - 11/29/12 at 12:00am

I'm pretty lucky, I guess. There's nothing in my collection with obvious clipping, though there are a few things with distortion.

 

The worst thing in my collection is the audio from a very crappy YT video of an arrangement of the Gabriel's Oboe theme, conducted by Morricone himself, and performed at the UN headquarters in New York in 2007. Now where are you going to find that? I've looked everywhere for some way to buy a legitimate recording of this. Obviously somebody taped it from live television and uploaded it to YT, whereupon it was subsequently deleted and then reuploaded by somebody else, incurring a further generational loss. It sounds like utter crap, but I can still enjoy it. Barely, but I can.

...when you purchase an artist's music through iTunes only because they sell it through there and upon purchasing a song you immediately notice how @#$% it sounds. No seriously, it sounds terrible. I don't think 256 kbps VBR AAC does the job for me. It sounds nearly identical to the artist's official video on YouTube. :(

Omi Gawd, byound beautiful

Hey man i'll go to! Although Sadly... 1 Dollar=88 Yen... so 388,500 yen can= $441... meaning Buying Jp is expensive... but we could just I dunno... LIVE THERE, come back to the Us with AWESOME CANS, and buy AMerican with our Yen xD

You know your an audiophile when you Boy Cott iTunes, Because ALAC is a pain to transcode...

 

You also know you an audiophile when Anemic Bass, is a reality check for Lucid dreams... [Because you KNOW your Sony Xb700s are TOTALLY FAR CANNONS... if they are Anemic your dreaming ;3] 

 

Also, you know your an audiophile when you RUN through a mall to try a new set of cans, in addition... Red Hot Chili Peppers Instrumentals... from UMPP[1987]... PURE FUNK MAGIC! 

Edited by Mshenay - 11/29/12 at 8:45pm

It'll be the mastering - not the container. If you want to find out for sure, get a well mastered CD rip or HQ Flac file. Create a converted copy at 256aac. Use Foobar2000 + abx plugin. Open both files, apply replaygain to volume match, then run the abx test. Do at least 15 iterations. If you're game, post the log. Pretty sure you won't be able to tell flax from 256aac when both are same volume and the test is blind. It's enlightening to say the least.

The issue with the quality of the tracks is more than likely to be the mastering. If it's been poorly mastered - it won't matter what container it's in.

You've got a point there. Darn it... :/

I purchased a different song from the same artist (different album) and it sounds significantly better.

 

 

And no I'm not one of those people who claims there is a huge audible difference between FLAC and 256 kbps AAC/320 kbps MP3/etc. etc.

Wheee!

^ Warning: boring video, but you get the point

Edited by miceblue - 11/29/12 at 9:26pm

@miceblue
Good on you for taking it, and thanks for posting a pretty good how-to.

Your 7/10 is actually a statistical fail. Next time try to do 15 iterations - the more you do, the better the results bear out. Ideally, to prove you weren't guessing, you need to get under 5% (pretty hard to do when you get 15 iterations - for most of us impossible at aac256 which is pretty much transparent).

Also - try it with the "hide results" box checked. Then its a true blind.

You know you an audiophile when you realize, lossess is maybe 10% better than lossy [if your rig can let you hear it] and 96/24 is only about 5% better then 16/44.1  and that's only if you can trick your brain into hearing it ;3

...When your mother gifts you a shower curtain because "(she) knew you'd like it because it had headphones on it!"

So Head-Fi, please help me determine which headphones are on my shower curtain.

IIRC, there is no audible difference between those two.

 

Correct. The bitness just determines the total available dynamic range, which is already large enough with 16-bit to encompass pretty much the quietest sounds any human can hear through the loudest sounds any human should ever safely hear. In other words, if you calibrate the volume from your listening position so that the quietest possible sound encoded is the softest sound you can hear, the loudest possible encoded sound will be ear-splittingly loud (theoretically 96dB above the quietest sound). That same range with 24-bit is theoretically 144dB, meaning the loudest sound encoded would almost certainly deafen you if you listened for any length of time.

 

Worth noting is there's some encoding/decoding overhead, so the 16-bit number is more realistically 90dB in actual practice. Also, this does not mean that the loudest sound you can ever get from 16-bit is 90dB from your listening position--quite obviously you can amplify the signal to any level you desire, and people manage to deafen themselves with their music just fine at 16-bit.

 

Bitness has nothing to do with fidelity. The reason studios use 24-bit is because, along with the wider dynamic range, the noise floor is much lower. This is important when stacking many tracks on top of one another in mixing, as the noise is cumulative.

 

The sampling rate has a little more to do with theoretical sound quality, except that even there you're still dealing with a studio technicality. Dithering noise (the result of a shaping algorithm applied to the cumulative sampling error) is shifted to the highest frequency ranges of your sample rate, which for 44.1k stereo is somewhere between 10kHz and 20kHz (I believe there's a mathematical theorem that gives the exact range). This is theoretically audible, and once again, cumulative. Using a higher sample rate ensures that this dithering noise ends up outside the audible frequency range for humans and doesn't interfere in mastering.

 

Once all this is done, the track is mixed down to 16-bit, 44.1k. Since this conversion only happens once at the very end, the artifacts (raised noise floor and dithering noise) are acceptably faint. Since the overwhelming majority of everything people record will never use a dynamic range greater than 96dB (I guarantee you none of your music does), you don't lose anything by having a theoretically smaller dynamic bandwidth.

 

 

 

This excursion into digital audio obscurity is brought to you by me forgetting to shut off my alarm again and thus being woken up at seven in the morning for no good reason whatsoever. And you know you're an audiophile when, after reading all the above, you think, "But I can still hear a difference!"

Edited by Argyris - 11/30/12 at 5:03am