Members Megadeth Man Posted October 31, 2007 Members Share Posted October 31, 2007 Yes, you can quite easily tell the difference between a CD track and a badly encoded MP3 of it. But can you tell the difference between a CD track and a well encoded MP3? This WAV file contains 3 samples: the original CD rip, and 2 mp3 encodes of it. The samples start at 0:00, 1:00 & 2:00. Which one is the pure CD rip? Test 1: http://rapidshare.com/files/66408829/Gloden_Ears_MP3_Test.wav.html Or http://www.megaupload.com/?d=SIFPE644 Tech Info: - The track was chosen to try and expose failures in mp3 encoding and because no one here is likely to have it. - It was ripped from a CD using EAC 0.99 Beta 3 on the highest accuracy settings. - Mp3 encoding was done by Lame. - One encode uses Lame preset extreme (200-240kbps VBR) and the other uses Lame preset insane (320kbps CBR). - The clip was assembled using Cubase. - There was no other processing involved on any of the audio. - The order of the 3 samples was decided by the random number generator at random.org - There is no trickery going on here. Here is a shorter second test with a different track, in case anyone feels the first wasn't suitable! Same conditions as above. Samples start at 0:00, 0:30, 1:00. Test 2: http://rapidshare.com/files/66412562/Gloden_Ears_MP3_Test_2.wav.html OR http://www.megaupload.com/?d=1KOCXUY9 Here are lower bitrate versions of the previous 2 samples. Test 3: Same sample as Test 1 but the mp3 sections are 128 kbps & 160 kbps. http://rapidshare.com/files/66442304/Gloden_Ears_MP3_Test_3.wav.html OR http://www.megaupload.com/?d=NZOEUA9Y Test 4: Same sample as Test 2 but the mp3 sections are 128 kbps & 160 kbps. http://rapidshare.com/files/66446228/Gloden_Ears_MP3_Test_4.wav.html Or http://www.megaupload.com/?d=6MODTODD Test 5: New sample heavier on cymbals and wanking. MP3 bitrates are 128 kbps & 240 kbps. Bonus points for knowing who it is. http://rapidshare.com/files/66467396/Gloden_Ears_MP3_Test_5.wav.html Or http://www.megaupload.com/?d=VYYCK6RT Please refer all people who bitch about mp3 altering, or adding "digital nasties", to the mix to this thread. Link to comment Share on other sites More sharing options...
Members Osram Posted October 31, 2007 Members Share Posted October 31, 2007 original, 200, 320 do I win!? Link to comment Share on other sites More sharing options...
Members fendercat Posted October 31, 2007 Members Share Posted October 31, 2007 My guess is 200, original, and 300. I think the hihat sounds definitely clearer with the second one than the first. Hard to tell which is the original between 2 and 3 though. Link to comment Share on other sites More sharing options...
Members code_blue Posted October 31, 2007 Members Share Posted October 31, 2007 Should we ignore the kbps displaying on all of our music playing programs? Link to comment Share on other sites More sharing options...
Members tommythelurker Posted October 31, 2007 Members Share Posted October 31, 2007 No one is arguing that an MP3 encoded at a high bit rate doesn't sound good. The problem is that most are encoded at much lower rates. Link to comment Share on other sites More sharing options...
Members blargh Posted October 31, 2007 Members Share Posted October 31, 2007 No one is arguing that an MP3 encoded at a high bit rate doesn't sound good. The problem is that most are encoded at much lower rates. Depends on where you get them... personally, I can't tell a difference any time I've tried to do a blind test between a high bitrate mp3 and the source file, but my hearing isn't so good... Link to comment Share on other sites More sharing options...
Members Herald of Light Posted October 31, 2007 Members Share Posted October 31, 2007 That's kind of a dumb test. Lots of stuff is like 128, and made with who knows what random, crappy encoder. You're lucky if it's encoded in 192 with LAME. But anyways, I honestly can't tell the difference with most stuff, even in "lower" quality. It tends to only be certain things, in certain songs, that jump out at me, and only if I'm used to hearing the actual CD. I don't just hear a decent MP3 with no frame of reference and instantly go, "ZOMG TEH MP3NESS RUINZ TEH QUALITE3Z!!11!1!!" Link to comment Share on other sites More sharing options...
Members filthy6 Posted October 31, 2007 Members Share Posted October 31, 2007 My guess is 200, original, and 300. I think the hihat sounds definitely clearer with the second one than the first. Hard to tell which is the original between 2 and 3 though. +1 200, original, 320 Link to comment Share on other sites More sharing options...
Members futurism Posted October 31, 2007 Members Share Posted October 31, 2007 china cymbals are the true enemy of mp3's due to the huge chunk of the spectrum they take up, parts of which are heavily impacted on by the psychoacoustic processing of mp3 encoding Link to comment Share on other sites More sharing options...
Members Megadeth Man Posted October 31, 2007 Author Members Share Posted October 31, 2007 That's kind of a dumb test. Lots of stuff is like 128, and made with who knows what random, crappy encoder. You're lucky if it's encoded in 192 with LAME. But anyways, I honestly can't tell the difference with most stuff, even in "lower" quality. It tends to only be certain things, in certain songs, that jump out at me, and only if I'm used to hearing the actual CD. I don't just hear a decent MP3 with no frame of reference and instantly go, "ZOMG TEH MP3NESS RUINZ TEH QUALITE3Z!!11!1!!" The main reason for this thread is in response to regularly seeing comments in clip threads like "the mp3 sounds crap compared to the original", "the mp3 really compresses the sound and adds lots of 1010011010's", etc. There is no reason anyone who posts a clip here can't encode with Lame at a bitrate better than 128kbps. Still I may add a low bitrate version to this test now you bring it up. On a wider scale I need to spread the importance of using Lame at a decent bitrate for all mp3's. Link to comment Share on other sites More sharing options...
Members xiwiwix1 Posted October 31, 2007 Members Share Posted October 31, 2007 Before I read anything other than the first post I think it's 240, then 320, then wav for both. The first clip I used when her vocals come in, on the second clip it was when the girl started the last singing part right before the guitars did it's lasts notes. Link to comment Share on other sites More sharing options...
Members xiwiwix1 Posted October 31, 2007 Members Share Posted October 31, 2007 The problem I think with these samples is that I listen to metal and there is a WHOLE lot more going on with symbols and what not, so you really can't compare these songs to what I listen. I'm going to go back and listen to the first clip with my heads phones on. Link to comment Share on other sites More sharing options...
Members xiwiwix1 Posted October 31, 2007 Members Share Posted October 31, 2007 Damnit, almost tempted to play the clip on my studio monitors. lol!!!! This thread gets the: ghey cop Link to comment Share on other sites More sharing options...
Members xiwiwix1 Posted October 31, 2007 Members Share Posted October 31, 2007 Does Windows Media Player encode with LAME? That's what I use to encode my mp3's to 320, and I can still here a difference here on my computers speakers. But using WMP could be the problem. Link to comment Share on other sites More sharing options...
Members JamesPeters Posted October 31, 2007 Members Share Posted October 31, 2007 Does Windows Media Player encode with LAME? I highly doubt it. Not only that you'll want to check to see if the "quality" emphasis is selected, otherwise it'll probably encode with the emphasis on speed instead (speed of encoding). I usually encode with LAME with the "quality" emphasis, "stereo" (not "joint stereo"), and at least 160 kbps. Link to comment Share on other sites More sharing options...
Members Neilrocks25 Posted October 31, 2007 Members Share Posted October 31, 2007 I would say the middle one................. but the origanal song is made of samples and the digital converters make a bigger difference then an mp3 or cd rip oh and I am using my pc at work, if I used my studio pc I could give a much better reference. Link to comment Share on other sites More sharing options...
Members Guitarfreak108 Posted October 31, 2007 Members Share Posted October 31, 2007 its quite hard to tell the diff,but in my experiance anything lower than 128kbs, is when the mp3 sounds {censored}. everything above sounds great. i compared myself a 320 kb mp3 and a 192kb mp3 and there was no difference. even 128 you cant tell. but i tired a 96kb and i noticed a big diff. Link to comment Share on other sites More sharing options...
Members xiwiwix1 Posted October 31, 2007 Members Share Posted October 31, 2007 I highly doubt it. Not only that you'll want to check to see if the "quality" emphasis is selected, otherwise it'll probably encode with the emphasis on speed instead (speed of encoding). I usually encode with LAME with the "quality" emphasis, "stereo" (not "joint stereo"), and at least 160 kbps. This maybe the reason I hate mp3's. Things are looking up. Link to comment Share on other sites More sharing options...
Members Megadeth Man Posted October 31, 2007 Author Members Share Posted October 31, 2007 The problem I think with these samples is that I listen to metal and there is a WHOLE lot more going on with symbols and what not, so you really can't compare these songs to what I listen. I've added a test that's heavier on the cymbals for you in the first post. Link to comment Share on other sites More sharing options...
Members peehoo Posted October 31, 2007 Members Share Posted October 31, 2007 Like said, most stuff is like 128. And It won't make as much of a difference on PC speakers, but I know that mp3s will sound worse on my HI-FI. Anyway, as was debated before on these boards, nowadays people don't seem to be that much into hi fidelity hi-fi. I'm not talking obnoxious price cable corksniffery here, but a decent hi-fi setup will sound better than a damn iPod and people just seem not to care about it that much. It's all about portability and hipness... Link to comment Share on other sites More sharing options...
Members JBecker Posted October 31, 2007 Members Share Posted October 31, 2007 I haven't read anyone's remarks in the thread because I don't want to be biased. My guess is, with neither of the two higher bitrates, will I be able to hear a difference. I think that good LAME encoding is probably not the best measure, and I tend to hear the difference more with pieces that I'm familiar with, although I do think the particular clip chosen is a pretty good one in terms of testing. Doing just the 30 seconds of each clip, I'd have to say if I had to guess, original, 200, 300. That being said, of course there is no huge difference, which is why my library is lame insane 320. Link to comment Share on other sites More sharing options...
Members JBecker Posted October 31, 2007 Members Share Posted October 31, 2007 I'm listening to the 30 second clip with the original, 128, and 160 in just a moment: First one is not the original, maybe 128. Second one is also not original, not sure if it's better or worse than the first, I'd have to relisten. There is a clear difference, the last one is definitely the original. Link to comment Share on other sites More sharing options...
Members JBecker Posted October 31, 2007 Members Share Posted October 31, 2007 Did I kill the thread? Link to comment Share on other sites More sharing options...
Members tri99er Posted October 31, 2007 Members Share Posted October 31, 2007 Well I stopped encoding to MP3 for the simple reason that I could hear a difference with the drums, the cymbals, and hi-hats would have some artifacts left on them, I have though switched to AAC/mp4, and I get pretty damn close to an exact copy, and better file size than mp3. Link to comment Share on other sites More sharing options...
Members ThrashmetalErik Posted October 31, 2007 Members Share Posted October 31, 2007 no but i liek supporting bands I like more than ticket sales. Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.