[quote = ROBSQIX]
You mean studio masters or 24/96 based files?
Actually most studio files are recorded at 24/96 anyway.
You can fill up your system with 24/96 but if you do the research 16/44.1 is more then enough.
ANY digial audio based of an analog waveform is always a sampling of the original material. At what point is good enough
is the question. Most consider 16/44.1 audio to be that point... Some look for 24/96....choice is yours..
[/quote]
For playback, I found that I was able to distinguish between high quality lossy (eg 320 kbps MP3 or AAC) and Lossless with some effort. Telling the difference between 96/24 and 44.1/16 required perfect listening conditions (eg, turning off all fans/noise sources in my apartment -- including moving my computer well outside of the listening room and listening with a USB dac with a long USB cable...). Even then, I could only just barely tell the difference in ABX (about 70% of the time) If I focused completely on artifacts -- rather than the music itself.
However, I still find I like to have the highest quality source possible in my archive -- either 24/96, or lossless CD rips. That way if I want to, I can make lossy rips for my car (reads MP3 CDs), or to put on my phone (I'll be listening on poorer quality headphones in noisy environments -- negating any advantage of the higher quality source), without suffering "double losses" from say converting an MP3 to an AAC. Having a higher quality source is also more important if doing any DSP operations -- such as eq or phase to compensate for the acoustic of the listening room. (For example, I actually compensated for the not-very-flat frequency response of my car stereo by burning specially eq'd CD's specifically for the car) While the difference between lossless and lossy may not be easily audible by itself for straight playback (because the lossy codecs are based on psychoacoustics), the "losses" can be exaggerated greatly by digital filtering etc.
NOTE: I did some online hearing tests and found that my hearing sensitivity drops off pretty rapidly above 18.5kHz, and I was only able to distinguish between 0.2 dB sound intensity difference for pink noise in a double-blind test -- this means I likely won't be able ot hear much benefit of higher frequency response and/or lower noise floor of 96/24, anyway.