|
Korean Job Discussion Forums
"The Internet's Meeting Place for ESL/EFL Teachers from Around the World!"
|
| View previous topic :: View next topic |
| Author |
Message |
mnhnhyouh
Joined: 21 Nov 2006
Location: The Middle Kingdom
|
 Posted: Sun Dec 17, 2006 1:48 pm Post subject: |
 |
|
| Giant wrote: |
@mnhnhyouh
Dude, you need to relax and wake up a bit with MP3's. Let me explain it to you simply. |
Sorry if I came across as aggressive, but this is an area in which I have some understanding. Let me explain it in a more cogent fashion.
Hypothetical example...
Imagine a recording of an alarm clock. Gentle ticking. Then the alarm goes off. All you can hear is the alarm, but the clock is still ticking. The microphone can also pick up the ticking when the alarm is sounding.
So both the alarm and ticking are on the .wav file. The ticking is making the file bigger with information that you cannot detect. So during the initial phase of converting .wav to mp3 a mathematical model of what we can hear is used to chuck out all that we cant.
The lower the bitrate, the more information is left out. When going to very low bitrates information you can hear is starting to be lost.
Then it is compressed.
Yes, the mp3 is not as good as a .wav file, it has less information. However, only those with extensive training, or hearing damage, can tell the difference between 192k vbr mp3 and the original, and that using good headphones.
The best settings are the quality settings which use a variable bitrate. A standard 192kbps bitrate uses 192 kb for every second of audio, no matter how complex. VBR using the quality settings, like -v2 or -v0 set a quality of the resulting mp3 file in the encoder. Then the encoder uses as many bits per second to achieve that quality. More complex parts might be using 320 kb per second, less complex parts use a lot less. -v2 comes out at about 190 kbps, but songs might average 220kbps, or 120kbps. It just depends on the complexity of the track
If you think you can tell the difference between two codecs, or two bitrates, and want to test your ability, download a free music player called foobar 2000, it has an ABX function that will let you compare, in a blind fashion, two tracks. It will play one track, telling you which one it is, then the other track, then one of them again, but you wont be told which it is. You gan swap back and forth between the three. Then you have to pick it. You do a series of these, and are given a probability of your ability to tell them apart.
About the hearing damage bit, if the alarm was out of your now damaged hearing range, you would just hear constant ticking on the .wav file, but hear a gap in the ticking on the mp3.
h |
|
| Back to top |
|
 |
Giant
Joined: 14 May 2003
Location: South Korea
|
| Posted: Sun Dec 17, 2006 2:54 pm Post subject: |
|
|
No worries, and I totally agree with you, especially this part you said:
| Quote: |
Yes, the mp3 is not as good as a .wav file, it has less information. However, only those with extensive training, or hearing damage, can tell the difference between 192k vbr mp3 and the original, and that using good headphones.
|
The lower the bitrate the more info is left out, yes I know all about it, I also have some understanding about this matter as well.
All I was trying to say in my original post is that MP3 is good, especially if you convert at a higher bit rate, and most people wont be able to tell the difference.  |
|
| Back to top |
|
|
rocklee
Joined: 04 Oct 2005
Location: Seoul
|
| Posted: Sun Dec 17, 2006 3:40 pm Post subject: |
|
|
| mnhnhyouh wrote: |
| Giant wrote: |
@mnhnhyouh
Dude, you need to relax and wake up a bit with MP3's. Let me explain it to you simply. |
Sorry if I came across as aggressive, but this is an area in which I have some understanding. Let me explain it in a more cogent fashion.
Hypothetical example...
Imagine a recording of an alarm clock. Gentle ticking. Then the alarm goes off. All you can hear is the alarm, but the clock is still ticking. The microphone can also pick up the ticking when the alarm is sounding.
So both the alarm and ticking are on the .wav file. The ticking is making the file bigger with information that you cannot detect. So during the initial phase of converting .wav to mp3 a mathematical model of what we can hear is used to chuck out all that we cant.
The lower the bitrate, the more information is left out. When going to very low bitrates information you can hear is starting to be lost.
Then it is compressed.
Yes, the mp3 is not as good as a .wav file, it has less information. However, only those with extensive training, or hearing damage, can tell the difference between 192k vbr mp3 and the original, and that using good headphones.
The best settings are the quality settings which use a variable bitrate. A standard 192kbps bitrate uses 192 kb for every second of audio, no matter how complex. VBR using the quality settings, like -v2 or -v0 set a quality of the resulting mp3 file in the encoder. Then the encoder uses as many bits per second to achieve that quality. More complex parts might be using 320 kb per second, less complex parts use a lot less. -v2 comes out at about 190 kbps, but songs might average 220kbps, or 120kbps. It just depends on the complexity of the track
If you think you can tell the difference between two codecs, or two bitrates, and want to test your ability, download a free music player called foobar 2000, it has an ABX function that will let you compare, in a blind fashion, two tracks. It will play one track, telling you which one it is, then the other track, then one of them again, but you wont be told which it is. You gan swap back and forth between the three. Then you have to pick it. You do a series of these, and are given a probability of your ability to tell them apart.
About the hearing damage bit, if the alarm was out of your now damaged hearing range, you would just hear constant ticking on the .wav file, but hear a gap in the ticking on the mp3.
h |
The whole point of VBR is to keep the file size down. A 192kbps song verses a VBR version will be smaller in size but will not sound as good due to its fixed bitrate. Simple as that.
I don't know if it is because of the style of music that I listen to but I can always tell the difference between mp3s and wavs, especially when I compare something like soundstage. |
|
| Back to top |
|
|
mnhnhyouh
Joined: 21 Nov 2006
Location: The Middle Kingdom
|
| Posted: Sun Dec 17, 2006 6:58 pm Post subject: |
|
|
| rocklee wrote: |
The whole point of VBR is to keep the file size down. A 192kbps song verses a VBR version will be smaller in size but will not sound as good due to its fixed bitrate. Simple as that. |
Sorry, that is not quite true. If the VBR song averages out at 192, then it will be exactly the same size as a 192kbps song that is CBR. But as it saved bits on the less complex parts and spent them on the more complex parts, it will be of better quality. This, however, applies only if there is variability in the complexity of the track.
| rocklee wrote: |
I don't know if it is because of the style of music that I listen to but I can always tell the difference between mp3s and wavs, especially when I compare something like soundstage. |
Can you do this in a bliind setting? There are so many studies out there that demonstrate that if you know what you are comparing you can find differences. I agree with those who write these codecs that only results from blind testing are worth discussing.
And what bitrates are you talking about? Telling 64kbps is easy, however I am prepared to lay a big bet (1,000,000 won?) with you that you cannot pick 320kbps from .wav in a blind test. I will lay 500,000 won for a 256 kbps test, and 300,000 for a 192kbps test, if I am allowed to do the ripping. I will even lend you my Shure E500 or Sennheiser HD555 for this test if you dont have anything better. Please PM me if you are willing to take this bet on when I arrive in Korea. I like candy.
h |
|
| Back to top |
|
|
rocklee
Joined: 04 Oct 2005
Location: Seoul
|
| Posted: Sun Dec 17, 2006 7:35 pm Post subject: |
|
|
| Quote: |
| Sorry, that is not quite true. If the VBR song averages out at 192, then it will be exactly the same size as a 192kbps song that is CBR. But as it saved bits on the less complex parts and spent them on the more complex parts, it will be of better quality. This, however, applies only if there is variability in the complexity of the track. |
The moment an encoder saves bits on less complex parts in a song (ie. a pause or low sounds) is where VBR effectively reduces file size. I think I said that a normal song would be smaller in size compared to a VBR version, I meant to say "bigger" in size. Sorry for the confusion.
| Quote: |
| Can you do this in a bliind setting? There are so many studies out there that demonstrate that if you know what you are comparing you can find differences. I agree with those who write these codecs that only results from blind testing are worth discussing. |
If you listen to mp3s day in day out and suddenly hear an original higher quality recording of the same thing you will notice the difference. Giving someone a few samples of differently encoded sounds to judge blindly wouldn't prove anything because their ears haven't been trained to tell the difference!
I am not a sound engineer, but after listening and comparing various forms of MP3s, AAC, WAVs, Super Audio CD and DVD-Audio, mp3s and AACs are indeed at the lower end of the spectrum.
| Quote: |
| I will even lend you my Shure E500 or Sennheiser HD555 for this test if you dont have anything better |
I use a Sony MDR-CD3000, Senn. 580 and a Sharp MD33. Not great but not bad either. |
|
| Back to top |
|
|
wormholes101
Joined: 11 Mar 2003
|
| Posted: Sun Dec 17, 2006 9:47 pm Post subject: |
|
|
| mnhnhyouh wrote: |
The biggest attraction of.ogg seems to be its opensource nature. As for being able to pick them, what headphones are you using? And what bitrate mp3 files are you comparing them with? And what source equipment?
It sits in third place, IIRC. First are LAME mp3 and AAC. I prefer mp3 as players support it than AAC. .ogg works on even less players. |
I don't have special headphones. Panasonic RP DJ100. Cost about US$10 although I see them online for about US$20.
http://www.dealtime.co.uk/xPF-Panasonic-Panasonic-RP-DJ100-Monitor-Headphones-for-DJ
Personally, I like to download flac and convert to something smaller for use in my iRiver. So, I been playing around with the MP3 v0 (I think this is 245kbps) setting (most audiophiles seem to think that this is the best setting) and also ogg and comparing them. I can just tell the difference between flac and and an ogg at 80kbs. Especially at the bass end it doesn't seem so beefy.
What do you mean by source equipment?
PS. Bit surprised that you think that ogg is under supported re hardware. I haven't found this my experience. |
|
| Back to top |
|
|
mnhnhyouh
Joined: 21 Nov 2006
Location: The Middle Kingdom
|
| Posted: Sun Dec 17, 2006 9:53 pm Post subject: |
|
|
| wormholes101 wrote: |
What do you mean by source equipment? |
It is the machine you are playing them on. Each generation is better than the last, though there are still some dodgy players out there.
| wormholes101 wrote: |
PS. Bit surprised that you think that ogg is under supported re hardware. I haven't found this my experience. |
The biggest section of the market will not playback ogg, that is the iPods. iRiver do, and now iAudio. Not many others, I dont think.
| rocklee wrote: |
If you listen to mp3s day in day out and suddenly hear an original higher quality recording of the same thing you will notice the difference. Giving someone a few samples of differently encoded sounds to judge blindly wouldn't prove anything because their ears haven't been trained to tell the difference!
|
So are you willing to take the bet on? I will encode your track of choice into a 192 CBR LAME mp3 file, and we will use any computer you want and any speakers / headphones you want. If you pick it right 8 out of 10 tries, I will give you 300,000 won, otherwise you give me 300,000 won.
h |
|
| Back to top |
|
|
wormholes101
Joined: 11 Mar 2003
|
| Posted: Sun Dec 17, 2006 10:47 pm Post subject: |
|
|
| mnhnhyouh wrote: |
| wormholes101 wrote: |
What do you mean by source equipment? |
It is the machine you are playing them on. Each generation is better than the last, though there are still some dodgy players out there.
| wormholes101 wrote: |
PS. Bit surprised that you think that ogg is under supported re hardware. I haven't found this my experience. |
|
That would be my iRiver E10. |
|
| Back to top |
|
|
wormholes101
Joined: 11 Mar 2003
|
|
| Back to top |
|
|
Greekfreak
Joined: 25 May 2003
|
| Posted: Mon Dec 18, 2006 2:51 am Post subject: |
|
|
This wonderful discussion about FLAC files vs. MP3 files is enviable, but I'd appreciate some more suggestions concerning the original question.
In any event, I've ripped songs from various sources of various bitrates and unless you're talking about a song from 1950 vs. one from 2006, there really isn't much difference, particularly when you burn them to cd.
An audiophile might know, but the average music listener wouldn't.
Re: my collection--I've got possibly the biggest collection of music (mostly in Canada) of anybody in this country, and I'm not looking to upload it all to an MP3 player.
Just a unit that will hold 20 hours of music/spoken word, no frills, no hassle, user-friendly, and on the cheaper side. |
|
| Back to top |
|
|
mnhnhyouh
Joined: 21 Nov 2006
Location: The Middle Kingdom
|
| Posted: Mon Dec 18, 2006 3:19 am Post subject: |
|
|
| Greekfreak wrote: |
Re: my collection--I've got possibly the biggest collection of music (mostly in Canada) of anybody in this country, and I'm not looking to upload it all to an MP3 player. |
How big is this collection?
| Greekfreak wrote: |
| Just a unit that will hold 20 hours of music/spoken word, no frills, no hassle, user-friendly, and on the cheaper side. |
How big is the 20 hours of stuff? If it is mostly spoken word (generally 128k or less, then you dont need anything too big to hold it, if it is mostly music, then 20 hours is much bigger.
As for your original question, tell me some things :
1) HD or Flash? HD are bigger capacity and more delicate but are still pretty robust.
2) You seem to be ruling out software to put music on?
3) Will you be wanting to drag music off to a computer?
4) What constitutes reasonable as it applies to memory and battery life?
h
h |
|
| Back to top |
|
|
rocklee
Joined: 04 Oct 2005
Location: Seoul
|
| Posted: Mon Dec 18, 2006 7:44 am Post subject: |
|
|
Haha...yes. Mind you, I didn't buy it for that much no way. I got it when people didn't know anything about it which was way back. Hardly used either. |
|
| Back to top |
|
|
rocklee
Joined: 04 Oct 2005
Location: Seoul
|
| Posted: Mon Dec 18, 2006 7:49 am Post subject: |
|
|
| mnhnhyouh wrote: |
So are you willing to take the bet on? I will encode your track of choice into a 192 CBR LAME mp3 file, and we will use any computer you want and any speakers / headphones you want. If you pick it right 8 out of 10 tries, I will give you 300,000 won, otherwise you give me 300,000 won.
h |
I'll go one better. If I get it right then you have to cut your arms off. Seriously 
If I have the time I will encode an mp3 and wav and let you hear the difference. |
|
| Back to top |
|
|
mnhnhyouh
Joined: 21 Nov 2006
Location: The Middle Kingdom
|
| Posted: Mon Dec 18, 2006 11:31 am Post subject: |
|
|
I have a number of CDs and thousands of mp3 and am good at encoding them. It seems that you are reluctant to take the bet?
h |
|
| Back to top |
|
|
rocklee
Joined: 04 Oct 2005
Location: Seoul
|
| Posted: Mon Dec 18, 2006 7:52 pm Post subject: |
|
|
| Its not that I'm reluctant, I don't even gamble. Besides I also know how to encode mp3s and wavs to make them sound exactly the same. |
|
| Back to top |
|
|
|
|
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
|
|
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
