Short version:
Simple analogy: you have a high quality audio source - say your own recorded music at sampled at 24bit/96KHz playing your favorite music instrument. You KNOW how that sounds when you recorded it back in the studio. Then you made an MP3 file off that recording and you brought both files home. Here's my point (pardon me the extremities ): you played BOTH files through 5 bucks god awful Chinese counterfeited Sony-branded tiny speakers. Can you hear a difference?
Long version:
First of all, I don't use Windows, I'm done with Windows. Second of all - I've done the test with my friends, written tests. We spend the whole day playing with my audio gear. We concluded at that time that MP3 were BAD, and that came from 7 people - on written test, with blinds and all (both on the person and on the audio gear were blinded so no one can cheat). The person who made the test (the one which renames the files) and the one who's executing the test (the one who plays the files) are two different person, and each brought their own files, which we all check that those are indeed uncompressed and compressed, simply by looking at it's spectrogram (and listening it, making sure that both are the same music ofc :) ).
The difference was night and day. Once again, I don't mean to brag - but please, use better gear. I don't mean to brag, but the speakers that I use is listed at $3750 each @vintageking.com, not to mention the DAC. Do not use headphones simply because it's way inferior than loudspeakers (specifically calibrated studio monitors). Someone on the forum said to me that the effect on the speaker driver in headphones are doppler effect or something, not comb filtering - I stand corrected. My point still stands that headphones can't produce signal as accurate loudspeakers simply because it only has very tiny (and most of the time, single full range driver). High frequency needs a very stiff and light media/cone/plate to vibrate accurately. If the same media/cone/plate is vibrating both at - say 90Hz - and at the same time, parts of it is vibrating at 12KHz, the sound coming off that media/cone/plate won't be as accurate compared to a separate drivers (two or three ways better in this sense, that is IF it's designed properly).
In my short 10 years in audio (especially PC audio, I did record my own music back then - when I was in Windows - for my own private consumption), my conclusion for now is : use studio monitors, it's cheaper and more accurate. Those speakers are merciless in sense that if there's something WRONG with the sound/music, you WILL HEAR IT. Some studio monitors are more brutally honest compared to other studio monitors, but at the end of the day - you can't expect to get good results when you use low-end-uncalibrated gears. Proper studio monitors are pre-calibrated from the factory, in an anechoic chamber down to several dB fluctuations in all frequency band. If you have the money, use ADAM Audio's SX series studio monitors. If you have A LOT of money, use Barefoot sound's studio monitors. These speakers ARE MADE to show you your music/sound in (sometimes brutally) honest way.
As I said in my previous comment, use better gear. Go get a loan on ADAM Audio studio monitors, preferably A7X, it's cheap enough - then do your ABX testing to your self using that studio monitor. If you can, get RME's Audio Interface while you're at it (preferably the Fireface UFX), simply because the Audio Interface (the source) is A HUGE FACTOR for you to be able to hear ANY DIFFERENCE. You can't expect to not hear a difference in a headphone connected to your on-board audio out on your laptop, compared to 15K worth of (some) professionally calibrated Audio equipment (the speakers), and higher quality of analog output stage of higher end Audio Interface (or DAC). The difference is NIGHT and DAY.
Go get better equipment, then do the ABX test.
zzz2496