[Patrickhat2001]

I have a pecular problem on my system--namely it' s that wav files I use (ripped via EAC) don't sound as good as most of the MP3 files I've downloaded. The big problem with the wavs is that they don't seem to have the warmth of many of the MP3s--in all honesty they can sound a little, well honky and thin. Now, of course I've done the obvious which is to make sure that I, indeed, can tell the difference between MP3 and wavs. I've ripped many wav files via EAC and compared them to the same songs ripped in the same manner and encoded as MP3s via LAME (320K CBR). In these situations the wav files always win. But compared to many of the MP3s I've downloaded (I admit by illicit means) the wav files that I've ripped myself seem to lose out--they have better detail and transparency but many times just sound off--kinda honky and thin and, in all honesty, unnatural sounding. Maybe even a bit, I don't know, processed sounding. The MP3s I've made from these wav files also sound like this. I'm I doing something wrong? I have noticed that many of the MP3s I download seem to have a slightly louder volume level which leads me to think that they've been dynamically compressed. Could this be what is causing the difference? Or could I have set something up incorrectly in EAC? I've already made the mistake of not turning off the cache for the for my CD-ROM drive before. Believe it or not but forgeting to turn the cache off caused all of my wav files ripped in such a manner to sound slightly smeered throughout the frequency range and harsh on the top end. Turning the cache off immediately cured this problem but the above mentioned problems still remain. Am I forgetting any to tweak any of the settings in EAC? I've checked the tutorials but everything seems to be set correctly excepting the offsets which I haven't bothered setting up since (as the tutorials say, anyway) their not supposed to change the sound. Anyway does anyone know what could be going wrong?

As for my system I play all my files through Foobar at 24 bit padded to 32 bit decoding at a sample rate of 88.2 Khz (with slow mode enabled). I also use the ASIO plugin (set to time critical and buffer at 0). In all three of my current soundcards--Audiophile 24/96, Revolution 7.1 and Terratec EWX 2496--this problems rears its head. Could I just be experiencing the supposed graininess of mid-fi digital sources?

 

[Wodgy]

MP3 compression in general is biased towards a slightly warmer sound. At very low bitrates (128kbps or lower) this is very pronounced because usually all frequencies above 16kHz are simply discarded. However, even higher bitrate MP3s tend to be warmer sounding. (The opposite seems to be true of WMA, which tends to be more analytical sounding.) It depends on what kind of sound you like. There was someone who posted last week about how the MP3s he was using on his iPod were too warm and so he had to go lossless.

I know you say that you've tried making homemade MP3s and you say they don't sound better than the WAVs, but I suspect that if you did this test again but as a blind test you might find that you even prefer your homemade MP3s.

Redbook CD audio can easily be brittle or harsh sounding. It's just the nature of the beast. The better your source, the more this effect seems to be minimized and so the sound tends to appear warmer. This isn't a hard and fast rule by any means, but it's been my experience that when I've built these things myself, a clean power supply (especially to op-amps) makes a lot of difference. Witness how much improvement the RME modding crew got by just taking the op-amps out of the signal path (it was too difficult to improve the power supply on that card). On the other hand, the E-MU that everyone's talking about seems to have a better power supply by default and Iron_Dreamer didn't notice major changes when bypassing the op-amps. Likewise, Sbulack noted a major increase in sound quality by adding an inexpensive external DAC (the Piccolo) to his Terratec EWX.

I'm not going to recommend any specific upgrades. Just thought I'd add my two cents to explaining the "warmer MP3s" phenomenon.

 

[pedxing]

I have the opposite experience with mp3's. I find my mp3's grainy and riddlfed with artifacts. I have also discovered that mp3's have a tendancy to make my music sound more mono than stereo. I think the stereo separation is lost.

 

[asdfeproiu9]

That is probably due to the fact that sometimes MP3 converts the stereo into joint stereo during the encoding process.

 

[fewtch]

Quote:

Originally Posted by asdfeproiu9 That is probably due to the fact that sometimes MP3 converts the stereo into joint stereo during the encoding process.

Joint stereo is an encoding method that (when decoded) translates back to full stereo again. There's a common misconception that it's somehow different than regular stereo, or fiddles with stereo separation... nothing could be further from the truth. Many lossless algorithms use joint stereo encoding as well to help increase compression.

 

[tiberian]

i have an opposite experience too, mp3 is less defined and detailed than wav in my rig...it just doesnt sound right.

but yeah uncompressed wav files do sound bright and harsh if your system is not well matched...EQing will help tho.

 

[go_vtec]

Hmm, I'm with the others. Wave encoded music (lossless) sounds fuller and warmer than MP3. Although at 320kbps, it's getting rather difficult to tell mp3s from wave apart.

 

[pedxing]

Quote:

Originally Posted by fewtch Joint stereo is an encoding method that (when decoded) translates back to full stereo again. There's a common misconception that it's somehow different than regular stereo, or fiddles with stereo separation... nothing could be further from the truth. Many lossless algorithms use joint stereo encoding as well to help increase compression.

Does this mean this loss of stereo separation is inherent in mp3 and not made any worse with joint stereo turned on?

 

[fewtch]

Quote:

Originally Posted by pedxing Does this mean this loss of stereo separation is inherent in mp3 and not made any worse with joint stereo turned on?

If there's loss of stereo separation with an MP3, then it was encoded with a poor encoder. There is really no single "MP3" in terms of all of them being alike, encoder quality plays a major part in how the end product sounds.

 

[Spent&Bent]

Quote:

Originally Posted by fewtch If there's loss of stereo separation with an MP3, then it was encoded with a poor encoder. There is really no single "MP3" in terms of all of them being alike, encoder quality plays a major part in how the end product sounds.

Yes, remember xing encoder? I actually used that thing a lot at first when it came out, not realizing what I was throwing away encoding stuff with that thing

.

 

[Iron_Dreamer]

Quote:

Originally Posted by Patrickhat2001 Could I just be experiencing the supposed graininess of mid-fi digital sources?

I think so. When I went from the Delta410 to the stock RME, I was amazed by the much fuller, more powerful sound, especially in the bass. I felt as if I was hearing real bass for the first time. That is also the type of sound I've heard from the EMU (modded or stock), Cary 308, Grace 901 DAC, etc. The thin sound you're complaining of is present with all soundcards I've used other than the EMU and RME. Why not sell two of your cards for one better one, and eliminate your problem?