128k vs 320k MP3's, can you hear a difference?

Face
Face Posts: 14,340
edited March 2008 in The Clubhouse
"He who fights with monsters should look to it that he himself does not become a monster. And when you gaze long into an abyss the abyss also gazes into you." Friedrich Nietzsche
Post edited by Face on

Comments

  • ben62670
    ben62670 Posts: 15,969
    edited March 2008
    I guessed right. Sounded slightly better. My puter is using the Accurian amp modded, and the Ipod thingy speakers. Not a reference system for sure, and I don't know what their player is. Now on a system with the 2 samples I think it would be more apparent.
    Please. Please contact me a ben62670 @ yahoo.com. Make sure to include who you are, and you are from Polk so I don't delete your email. Also I am now physically unable to work on any projects. If you need help let these guys know. There are many people who will help if you let them know where you are.
    Thanks
    Ben
  • Gaara
    Gaara Posts: 2,415
    edited March 2008
    Not to scientific, you only do the test once with a sound clip you have never heard before. I guessed right without ever listening to the clips. Interesting comparison though.

    If anyone uses Foobar you can setup a similar comparison. Foobar can convert one of your favorite songs to 128kbps mp3 and 320kbps mp3 then compare them using Foobar's ABX. Can be ear opening.

    P.S. Mike did you end up selling those cables? Let me know
  • honda cber
    honda cber Posts: 267
    edited March 2008
    slightly better? on some material, it makes a HUGE difference. 320 doesnt seem to have as much weird crap going on up top. i particularly notice on very dynamic material or stuff that has lots of high frequencies (those fizzy-swirly cymbals really get on my nerves at times).
    a
    gear list:
    1 down, 4 up....
  • sucks2beme
    sucks2beme Posts: 5,601
    edited March 2008
    There was a survey done of cd VS. Mpeg at different bitrates.
    Randomly selected test listeners were given songs at high, low, and
    standard cd, and asked to pick which was which. Almost all the
    test subjects could spot the cd, but the different bitrates stumped them.
    I can't find the link right now, but the results were interesting.
    Audiophiles should fare much better, but it shows just how few people
    pay real attention to sound quality.
    "The legitimate powers of government extend to such acts only as are injurious to others. But it does me no injury for my neighbour to say there are twenty gods, or no god. It neither picks my pocket nor breaks my leg." --Thomas Jefferson
  • Joe08867
    Joe08867 Posts: 3,919
    edited March 2008
    I missed it. Of course I am using really cheesy speakers at work to do this. I need to upgrade.

    I don't think it is a good test either. There isn't much going on in that clip. Beside the singing there is almost no music and the echo can mask alot.

    On my Ipod I have both 128 and 320. There is a difference for sure. 128 can get gritty in the high end to me.
  • devani
    devani Posts: 1,497
    edited March 2008
    there's a difference for sure....4-5MB is not enough to reproduce music the way it was intended..
    Video: LG 55LN5100/Samsung LNT4065F
    Receiver: HK AVR445
    Source: OPPO BDP-93
    HT: POLK SPEAKERS RTi6, FXi3, CSi5, VTF-3 MK2
    2Ch system: MC2105, AR-XA, AR-2A, AR9, BX-300, OPPO BDP-83
  • michael_w
    michael_w Posts: 2,813
    edited March 2008
    I could barely hear a difference at work but I'm using one headphone with my laptop and they're the stock ipod headphones smilesickmi8.gif (and yes I guessed correctly). And people laugh when I don't use lossless for my ipod :rolleyes: . Cheap headphones don't do it justice.
  • Music Joe
    Music Joe Posts: 459
    edited March 2008
    I thought they sounded the same and just guessed...and guessed wrong.

    I'm sure I hear a differences of 128 vs 320k on the main system :o
  • zingo
    zingo Posts: 11,258
    edited March 2008
    I like those fun little internet things. Nice post Face.
  • AsSiMiLaTeD
    AsSiMiLaTeD Posts: 11,725
    edited March 2008
    In that clip to me it's most noticeable at the orchestra hit at the very end...I guessed right the first time easily
  • audiobliss
    audiobliss Posts: 12,518
    edited March 2008
    Dang.

    Do I hafta sell my system, now? :(
    Jstas wrote: »
    Simple question. If you had a cool million bucks, what would you do with it?
    Wonder WTF happened to the rest of my money.
    In Use
    PS3, Yamaha CDR-HD1300, Plex, Amazon Fire TV Gen 2
    Pioneer Elite VSX-52, Parasound HCA-1000A
    Klipsch RF-82ii, RC-62ii, RS-42ii, RW-10d
    Epson 8700UB

    In Storage
    [Home Audio]
    Rotel RCD-02, Yamaha KX-W900U, Sony ST-S500ES, Denon DP-7F
    Pro-Ject Phono Box MKII, Parasound P/HP-850, ASL Wave 20 monoblocks
    Klipsch RF-35, RB-51ii

    [Car Audio]
    Pioneer Premier DEH-P860MP, Memphis 16-MCA3004, Boston Acoustic RC520
  • Refefer
    Refefer Posts: 1,280
    edited March 2008
    Got it right. The clarity difference was very tangible, and the airiness didn't feel genuine on the 128kbps one. Further more, the dynamics seemed slower and more distorted.

    I was using a pair of Sennheiser HD-280 Pros through my Rotel preamp, so that probably made it a lot easier to hear than through computer speakers :D
    Lovin that music year after year.

    Main 2 Channel System

    Polk SDA-1B,
    Promitheus Audio TVC SE,
    Rotel RB-980BX,
    OPPO DV-970HD,
    Lite Audio DAC AH,
    IXOS XHA305 Interconnects


    Computer Rig

    Polk SDA CRS+,
    Creek Audio 5350 SE,
    Morrow Audio MA1 Interconnect,
    HRT Music Streamer II
  • jcaut
    jcaut Posts: 1,849
    edited March 2008
    I can almost always pick out a 128kbps mp3 file from a 320kbps. With other lossy formats like wma it's harder to tell, IMO, but on most material mp3 really stinks at 128. And the better recorded the original material, the worse it stinks at 128/mp3.

    This might sound contrary to what would seem to be logical, but I think that on poorly recorded or mastered material -- like most modern, popular-type music, with heavy dynamic compression-- good quality mp3 (especially VBR)sometimes sounds better than the uncompressed versions. I think it has to do with the way the compression works. Even though most of us think of it as a low quality format, you have to be mindful that there has been much effort put into maintaining the perceived audio quality.

    When a recording is converted to mp3, most of the signal above some frequency threshold (like 16kHz) is discarded. Now some of the dynamic peaks with high frequency content above that threshold are allowed to remain. Their high frequency content then stands out above the artifically created silence, making the recording seem more dynamic.

    Lets see if I can show what I'm talking about:

    This actually is not a bad recording, but.. Anyway, here's a spectral view of the uncompressed version. The vertical scale is frequency and the intensity of colors shows the relative energy content. You see most of the energy is concentrated below 5000 Hz, but there are lots of peaks up into the 20kHz range and a constant background noise (blue- very low level - almost inaudible).

    Now look at the same view of the mp3 (compressed) version (this is VBR from LAME, the "standard" preset in version 3.97). You can see that much of the material above about 16-17kHz is gone, replaced by silence, but some of the high frequency peaks remain. They're dynamic-sounding against the silence and help to mask the fact that a lot of the actual material has been thrown away.

    That's not the best of examples of what I was trying to describe, but it's the first thing I came up with and I don't feel like searching out other examples right now. Anybody agree with my theory? Edit, added another picture showing only the things that are different on the two previous pictures.
  • wizzy
    wizzy Posts: 867
    edited March 2008
    Absolutely ... I do 192Kbps minimum. Actually I just use the extreme or standard preset of LAME depending on what I am encoding.

    The best encoder is LAME which is incorporated into various things, I might suggest CDEX with the latest LAME DLL.

    Most people will use the standard preset which will result in a VBR MP3 with an ABR of ~190kbps ... the extreme preset for me usually ends up in the ~230kbps average VBR

    http://wiki.hydrogenaudio.org/index.php?title=LAME
  • BaggedLancer
    BaggedLancer Posts: 6,371
    edited March 2008
    I got it first try using my stinky Dell 5.1 computer speakers. Pretty significant difference.
  • mantis
    mantis Posts: 17,185
    edited March 2008
    When I fist got my ipod, I used 128k. I thought it sounded thin and edgy on top. then I discovered the ability to go 256k. that sounded more like the cd when playing back the Ipod. Now I use Apple Lossless which varies depending on the song. It's suppost to be exaclty the same as the orignal recording. On the Ipod this format sounds the clearest.

    So in my opnion yes you can hear the difference with the different recordings. Some recordings you can hear a bigger difference. Some older recordings at 256k sound the same as lossless.
    Dan
    My personal quest is to save to world of bad audio, one thread at a time.
  • Bamadude
    Bamadude Posts: 245
    edited March 2008
    I streamed some mp3's from my PC to xbox hooked to my main system and can tell a huge difference between 128kbs and 320kbps. Couldn't tell as much on my PC speakers.

    Here are some graphs that help show what different compression levels are doing in the upper frequencies (red line is CD footprint). It explains the tinny wobbly cymbal sound with higher compression, I can't stand it.

    Daniel
    AVR: Pioneer VSX-84TXSi (RIP - lightening) / Amp: Sunfire Cinema Grand / Klipsh R-10B Sounbar, LC65fx / Sub: Elemental Designs LT/1300 / TV: Panasonic TH-50PH9UK /SIZE]