Are you an audiophile?

Maybe everyone posting here should add what kind of audio setup they have a home or the place they're posting from. Because at least I think that most of you simply can't be audiophile because lacking a decent audio setup.
Like mentioned before, you possibly won't tell the difference between a mp3 and a lossless encoding if you're listening with cheap pc speakers or the earphones from you iPod.

My audio setup:
- Philips DVP 5505S DVD-standalone (also capable of playing SACD and DVD-audio with meridian lossless) [this one is main output for music]
- Soundblaster Live! Player 1024 with kX Project drivers [does the output when sitting in front of the pc]
- 2x BT-928 headphone amp (heavily modified, so output quality is quite decent) [one b-tech is for the philips, one is for the sb live]
- Teufel Concept E Magnum speaker set (this is a 5.1 surround set, not including a decoder) [hooked up at the philips b-tech, which does pass-through]
- Sennheiser HD570 Symphony headphone [hooked up a the sblive b-tech]
- Sennheiser RS140 wireless headphone [hooked up at the philips b-tech, second port]
- AKG K701 headphone [hooked up at philips b-tech, first port]
- iRiver iMP-550 portable CD-player-hybrid (this one gets the data from cd, ergo reads mp3-cds, cdda and also a bit of vorbis)
- Sony MDR-EX 81 SLB in-ear-phone [hooked up at iMP-550, the original iRiver headphone broke some time ago, defect membrane]

Thinking about building:
- DIY PIMETA headphone amplifier

And I vote for 'no' (because I don't own a STAX *g*)

cyay
liquid

Jam it back in, in the dark.

If you're into DIY Piccolo maybe I can interest you in this:
http://www.tangentsoft.net/audio/pimeta/ (or any other DIY amp on the site)

Currently I don't have enough free time for another DIY project (still have to finish a few things on the second b-tech like socketed opamp and new power connector)... so the PIMETA has to wait

Most amazing jew boots

Hehe, I should support that because it's hardware made in Germany *g*

Most amazing jew boots

Originally Posted by Kairyu

To put it simply, the more complex the audio/waveform is the more resolution you'll need in each frame in a mp3 to retain the overall sound quality.

Replace 'resolution' by bits. The problem is entropy : you need a certain amount of bits to represent a block of information. This is a result of information theory, which always applies if you want to represent information.
That's also the reason why you can't compress data (lossless) to arbitrary sizes.

Now take away the lossless component. That introduces an error vector measuring the error between the original data and the data reconstructed from the compressed data. If you lower the amount of bits below the minimum-value given by the entropy you get a non-zero error vector. And error obviously increases if you reduce the bit amount more and more.

Originally Posted by Kairyu

Kinda like a movie frame being shown in 640x480 pixels as compared to a higher resolution movie frame like 1024x768 pixels. Of course you'll see the difference, the same would go for audio resolution in a mp3 file.

But we're talking about bitrate here, not resolution. The effect of low bitrate in a movie keeping the resolution constant are pixel errors, like macro block artifacts, ringing, etc. - also sharpness reduces and the image looks blurrier then the original. But the blur is a specific effect of MPEG encoding, more specific: the cosinus transformation.

Originally Posted by Kairyu

Sampling rates are a different story but is still one of the major factors in the filesize. Sampling rates set the frequency range at which the file will encode at.

Uhm, not directly. Maximum frequency range is determined by the sampling rate, but the sampling rate alone only tells you how times per second the analogue signal was sampled. Obviously the digital data is more exact (compared to the analogue source signal) if sampling rate is increased.
Now you can take the next step and ask yourself what maximum frequency you can reconstruct from such a digital stream with sampling frequency X. And the Nyquist-Shannon theorem tells you that the frequency is X/2. So for reconstruction of a signal from 0 to 21kHz you need 2*21kHz sampling frequency.

Originally Posted by Kairyu

If the mp3 file had a sampling rate of 96kHz it will be able to capture more sounds (ie stuff outside of your hearing threshold) for you but at the cost of larger frames per second.

The enlarged frequency range is a nice sideeffect but you also get more precise signal reconstruction in the lower frequency range. So it's not only a waste of bits but does have its purpose.


cu
liquid

I am a dolphin, do you want me on your body?