phillipsjk wrote:ManaUser wrote:Reminds me of a previous run-in with an audiophile I had. He claimed there is no such thing as true lossless compression, and was offended that anyone would claim otherwise.
I became a bit of an audiophile after taking a two-year electronics program. If it was me, I would agree there is no such thing as true lossless compression: Every recording device has imperfections. The sampling rate and resolution you chose also determines how much information you retain. For example, an 8khz 8 bit mono Wav file will be (approximately) telephone-quality lossless compression (the telephone system uses some power-law compression, IIRC). CD-Quality audio is 44.1KHz, 16bit stereo. All "lossless compression" means is that you can convert from one lossless format to another (of sufficient quality) without loosing data. It is still possible to lose data due to bit-rot.
(I'm really not into oneupsmanship, but for the record, my four year EE degree trumps your two year electronics program. And I don't claim to know everything; just a lot.) WADR, it seems that you don't understand what "lossless compression," or even "compression" really means in this context. "Compression" is fundementally mathematics, not audio. It is conversion of a mass of bits to another mass of bits such that the original can be regained. If the original is regained exactly then the compression is lossless; if the original is regained approximately then the compression is lossy, i.e. it looses some data integrity. It is undeniable that lossless compression exists.
When the mass of bits represents a digitized audio signal, any loss of data integrity cuaeses a loss of signal integrity, i.e. audio quality, which may or may not be acceptable, depending on a number of factors, cheifly the degree of loss and ones sensitivity to it (and sometimes ones attitude toward it.) On the other hand, lossy compression of, say, a text document is basically never acceptable. The fact that zip, tar, and other compression methods exist is proof of the existance of lossless compression. (True, zip etc. files are not well suited to the real-time decompression required for audio playback, but there are other mathematically lossless methods which are.)
Now, it
is true that every recording device has imperfections, and there is, therefore, no such thing as lossless (i.e. perfect) audio reproduction. Digitization is actually two distinct steps, sampling and quantization, which are both inherrently lossy (in terms of sound integrity) just as all methods of analog recording are. But thtat has nothing to do with compression.
dtech wrote:phillipsjk wrote:EDIT: I do get annoyed at "audiophiles" who claim the quality of the cable (past a certain minimum) make a difference in digital sound quality. Unless you are getting drop-outs, a bit is a bit.
It's just a prejudice left over from the analog days.
The signal running in the wire from the amplifier to the speaker is always analog, and a big fat cable
can make a
small differnce there. But even in the analog days, big fat cables between the components were bullshit.