Why encode MP3s at 320kbps?

If the experts agree that it is difficult to tell the difference between encoding at 160kbps and 192kbps why encode at a higher rate? Surely encoding at 320kbps is a waste of time?

Hmmm, if you are ripping from media that you will no longer possess in the future yet don't have the space to do an exact copy a high bitrate rip is a good alternative without sacrificing quality and allowing for future encoding to a newer/better technology with little impact on quality.

But that is illegal so never mind...

With many types of music, it will be hard to diferentiate 160kps and 192kps files, but this is not always the case. The more complex and demanding the music is, the more resolution it needs. And I can't tell the difference between 48kps and 32kps, but I know they are both less than perfect!

Indeed, there are some parts of some songs that may require 320kps (especially in the car where our sound systems are better than computer speakers with noisey soundcards).

I think the best solution would be to use a variable bit rate scheme where the song is encoded at 160kps where the music isn't that complex and at upto 320kps when it can pay off. (The program analyses the music to determine when to use a higher resolution.)

But, on a home machine, where disk space is so cheap, 320kps might be good for archiving.

Well, I can tell you that I decided on using the --alt -bitrate -standard setting in Lame after much, much blind testing with the assistance of a friend. I don't know who says you can't tell the difference between 160 and 192, but I can 8 out of 10 times. I can tell the difference between 192 and --alt-bitrate-standard about 6 out of 10 times. Anything higher and I can't tell the difference at all. But everyone's going to have different results.

Did anyone ever get some flavor of lossless compression playback to work on the Empeg?

Stu

In reply to:

---

I think the best solution would be to use a variable bit rate scheme where the song is encoded at 160kps where the music isn't that complex and at upto 320kps when it can pay off. (The program analyses the music to determine when to use a higher resolution.)

---

How does VBR work, exactly? I've started using VBR for all newly-saved MP3's, (I use Sound Forge, since I'm an editor..) and it seems to work pretty-darn well. Some very-high end still has tinkle to it (like cymbols) but in very rare cases..

Mike.

How does VBR work, exactly?

MP3s are broken up into FRAMES. Each frame is a block of data representing a fixed time span. For a 16-bit stereo 44.1khz MP3 file, the time span is about 0.02612 seconds. So each frame is a short little snippet of the sound of the song.

VBR allows each of those frames to be a different encoding bit rate. For instance, one frame might be encoded at 320kbps because it has a lot of high frequency content, while another frame might be encoded at only 96kbps because it's got relatively simple low-frequency sound and doesn't need a lot of storage space to represent the sound clearly.

So VBR gives you more "bang" for your storage "buck", by allowing the file to only use up the bits when it needs the higher quality.

I need to research this some more then... I used Audiocatalyst with the LAME add-on, and 320 VBR to rip all of my CD's. And I have absolutely no question that when played back to back, the CD player sounds a lot brighter and crisper.

I suppose it could be the CD player itself adding something into the mix - but when I saw the topic of the thread, I thought to myself "only 320?".

tracerbullet-

I've noticed the same thing, in the car. Can't say I notice it really on the computer. I'd figured maybe it was something I hadn't delved into yet on the Rio, some setting, or something...

The best way to test that is to take your 320 VBR MP3 and use WinAmp to decode it and write the results out as a WAV file. Now write this WAV file and the original WAV that the MP3 was created from onto an audio CD.

You can then compare both on the same hardware, e.g. the CD player in question.

The best way to test that is to take your 320 VBR MP3 and use WinAmp to decode it and write the results out as a WAV file. Now write this WAV file and the original WAV that the MP3 was created from onto an audio CD.

You can then compare both on the same hardware, e.g. the CD player in question.

And, if at all possible, get someone else to write the audio CD, and don't have them in the room while you're testing -- so you don't know until after making your mind up, which was the original and which the encoding.

Peter

Take a look at your encoder settings to make sure that it's not set to cut off high frequencies. A lot of the default settings often cut off frequencies above about 16kHz. This may be the cause of your loss of brightness.

I used Audiocatalyst with the LAME add-on

I didn't know there WAS an add-on for Audiocatalyst that used LAME. Perhaps you meant AudioGrabber instead of Audiocatalyst?

And I have absolutely no question that when played back to back, the CD player sounds a lot brighter and crisper.

First things first. Remember that when you play a CD player, it's using a completely different set of audio output circuitry than the thing you're playing the MP3s with.

Even if you play a CD on your PC in Winamp, then play an MP3 in the same copy of WinAmp, they're going through two different signal pathways, and the CD audio is coming from the analog CD audio input connector on your sound card.

So you can't A/B a CD with an MP3 like that and expect to compare apples to apples.

The only way to compare the two is to create your encoded MP3, then use a disk-writer plug in (like the one in WinAmp) to convert the MP3 back into a WAV. Then burn that WAV as an audio CD track. Then play both the original CD and the burned WAV on an audio CD player. That's the only way to compare an encoder to an original CD track.

Now, having said that, there's one reason your MP3 might be less crisp than your CD. High-frequency rolloff. This was mentioned elsewhere in this thread, and it's in the FAQ. Most encoders will roll off extremely high frequencies unless you go into the encoder's options and deliberately change the setting. Make sure you've done this.

In reply to:

---

Now, having said that, there's one reason your MP3 might be less crisp than your CD. High-frequency rolloff. This was mentioned elsewhere in this thread, and it's in the FAQ. Most encoders will roll off extremely high frequencies unless you go into the encoder's options and deliberately change the setting. Make sure you've done this.

---

Since you seem to be the MP3 man.. :>
What's the 'Emphasis' flag setting that I see in Cool Edit and Soundforge? (I don't remember what the three settings were off hand, Other than the 'No Emphasis', I mean.)

What's the 'Emphasis' flag setting that I see

I don't know for sure, but I think it refers to the "pre-emphasis" flag used in CD burning. If so, I don't think it has a lot of bearing on this particular discussion.

I looked for a FAQ entry on this pre-emphasis flag at www.cdrfaq.org and didn't see one. Anyone have any more information on this?

Certain high frequencies were sometimes boosted on CDs to counteract the loss of high frequencies that would occur with older CD players during the digital to analog conversion process in early D/A converter circuitry present in CD players. A pre-emphasis flag would indicate that this process had been performed on the CD and those CD players with de-emphasis circuitry would act accordingly. As far as I know, this is ignored by all modern CD players and DACs. It is just an option in CD mastering software, because it is part of the Red Book standard.

Stu