Why I have to Dither for a CD but I don't for an iPod ?

[Logicno8]

I bounced my small portion of the song using, of course, Bounce option and it made me wonder about one thing. I haven't really pay attention but, we all know that If we want to burn a CD for listening we need to use Dithering.

Now...I have my sessions mostly at 44.100Hz/24Bit.

When bounce my songs (or portions like I did yesterday) I usually put them in my mp3 and listen them in the city and think about everything what is happening there amongst all the other famous bands songs in my mp3 as well.

I exported it as a MP3 and 320kbps (assuming that's the best quality for mp3) and I thought about ...how can I listen to that .mp3 file in my iPod or any mp3 device (without any dithering procedure) but I HAVE to Dither my files If they were about to be burned on a CD ? mp3 is mp3 right ? Playing on a CD or in my mp3 device.

So I can listen to a mp3 song exported as MP3 44.100Hz/24Bit in my device but I can't on a CD. I have to make it 44.100Hz/16Bit plus Dithering.

Why is that ?

[David Nahmani]

we all know that If we want to burn a CD for listening we need to use Dithering.(...) I HAVE to Dither my files If they were about to be burned on a CD ?

 

You don't HAVE to do anything. Dithering is an option, it's your choice whether or not to use it.

 

So I can listen to a mp3 song exported as MP3 44.100Hz/24Bit in my device but I can't on a CD. I have to make it 44.100Hz/16Bit plus Dithering.

Why is that ?

 

Again, the dithering is an option. You can only burn a 16 bit audio file onto a CD because that's what red book CDs use. http://en.wikipedia.org/wiki/Red_Book_%28audio_Compact_Disc_standard%29

 

mp3s, AAC, don't have to conform to the same red book standard.

[Logicno8]

we all know that If we want to burn a CD for listening we need to use Dithering.(...) I HAVE to Dither my files If they were about to be burned on a CD ?

 

You don't HAVE to do anything. Dithering is an option, it's your choice whether or not to use it.

 

So I can listen to a mp3 song exported as MP3 44.100Hz/24Bit in my device but I can't on a CD. I have to make it 44.100Hz/16Bit plus Dithering.

Why is that ?

 

Again, the dithering is an option. You can only burn a 16 bit audio file onto a CD because that's what red book CDs use. http://en.wikipedia.org/wiki/Red_Book_%28audio_Compact_Disc_standard%29

 

mp3s, AAC, don't have to conform to the same red book standard.

 

Oh...okay. Thank you David, nice explanation. So I misunderstood this one. I thought it is a MUST. So CDDA (Red Book) is where Dithering should come into play ? I mean, people keep complaining how mp3's are destroying quality of a recorded material, mostly, as I get it, because they lowering sample rate and bit depth in order to make them burned onto a CD (as a regular albums selling in stores). So Hi-Fi playback systems are incapable of reproducing higher quality than 44.100Hz/16Bit ? And people want to avoid (or lessen) "quantization error" and then they use Dithering feature to get over it with smallest quality loss, am I right ?

And than again, mp3 playback devices (as iPod) are CAPABLE of "reading" mp3 files which are bounced as 44.100 or 48.000 and 24 Bits right ?

I mean I didn't notice any difference when bounced mp3 right from my project (not changing sample rate or bit depth thus no dithering I did) to my desktop and listening through iTunes player and through my mp3 device later.

[David Nahmani]

Logicno8, I have to say, I'm not sure how you come to your conclusion or where your statements come from?

 

So CDDA (Red Book) is where Dithering should come into play?

No. Dithering is an operation you can use when reducing an audio signal's bit depth.

 

I mean, people keep complaining how mp3's are destroying quality of a recorded material, mostly, as I get it, because they lowering sample rate and bit depth in order to make them burned onto a CD (as a regular albums selling in stores).

No. mp3 have nothing to do with CDs, and they have a lower quality than a PCM file because they use lossy compression.

 

So Hi-Fi playback systems are incapable of reproducing higher quality than 44.100Hz/16Bit ?

Different Hi-Fi systems can playback different supports, different formats, etc... some of them can play mp3s, DAT tapes, PCM files exceeding 44.1kHz/16bits.

 

And people want to avoid (or lessen) "quantization error" and then they use Dithering feature to get over it with smallest quality loss, am I right ?

Kinda... if you want. If you ask me, you can't avoid or lessen quantization error, but you can hide it. That's what dithering does.

 

I mean I didn't notice any difference when bounced mp3 right from my project (not changing sample rate or bit depth thus no dithering I did) to my desktop and listening through iTunes player and through my mp3 device later.

Not sure why you would expect a difference. It's the same file.

[Logicno8]

David...I don't have any conclusion, that's why I;m asking you.

So...I THOUGHT (bu I wasn't right obviously) that CD's contain mp3 files but I used CD's for listening music long time ago so I forgot that they don't.

PCM's are Wave, Aiff, Caf, SDII (that's what Logic offers) and if we go to compressed file formats such as mp3 of course we can (but it's not mandatory obviously) use Dithering. Or we can leave mp3 as our project setting and export it as whatever happened to be our sample rate and bit depth.

Now I realized that Dithering IS NOT a mandatory at any time.

SO when people are going from say 48.000/24 to 44.100/16 they do that because they want to be sure that they file will be playable at any Hi-Fi system right ? ( I'm asking this because you said that some Hi-Fi systems can reproduce higher than that but not all of them).

Yes Dithering is introducing noise to a recording and masking "quantization error" not lessening or removing it. My bad. I knew that.

[Rev. Juda Sleaze]

Maybe you might find this snippet from an old thread helpful:

 

Could there be any benefit to dithering audio before converting it to mp3?

That's an interesting question, actually. There are no dither options in Logic when you bounce to mp3 because mp3 has it's own algorithms for how it deals with dynamics, particularly on the low level end of things. However, there may be some instances where the application of dither, used as a plug-in, may give a better results when bouncing the 32 bit stream to mp3. Since dither effectively multiplies the low level signals upward, this can potentially give the mp3 codec a more defined signal to chew on. This can also depend on the particular mp3 codec too. Trial and error are your only methods for determining if it has any value.

 

I tend to bounce 16bit dithered .aifs into iTunes, then if I want an mp3 version I just convert the 16bit file in iTunes. Sounds good to me.

 

The full thread: http://www.logicprohelp.com/viewtopic.php?t=52692&highlight=dithering

[Rev. Juda Sleaze]

PCM's are Wave, Aiff, Caf, SDII (that's what Logic offers) and if we go to compressed file formats such as mp3 of course we can (but it's not mandatory obviously) use Dithering. Or we can leave mp3 as our project setting and export it as whatever happened to be our sample rate and bit depth.

 

It won't be the sample rate and bit-depth of the project; it will be the sample rate and bit-depth of the .mp3 you exported

 

SO when people are going from say 48.000/24 to 44.100/16 they do that because they want to be sure that they file will be playable at any Hi-Fi system right ?

 

Not really. They would do that if, as part of the mastering process, they were bouncing to CD quality audio. It depends on what your target platform is, eg. CD, mp3, 24bit .wav, et al. Although if you were bouncing to CD quality, you may well want to check out how mp3s encoded from that PCM file would sound at varying bit rates, as consumers would be likely to import their CD tracks into an mp3 player.

 

That's my layman's understanding of it anyway.

[Logicno8]

PCM's are Wave, Aiff, Caf, SDII (that's what Logic offers) and if we go to compressed file formats such as mp3 of course we can (but it's not mandatory obviously) use Dithering. Or we can leave mp3 as our project setting and export it as whatever happened to be our sample rate and bit depth.

 

It won't be the sample rate and bit-depth of the project; it will be the sample rate and bit-depth of the .mp3 you exported

 

SO when people are going from say 48.000/24 to 44.100/16 they do that because they want to be sure that they file will be playable at any Hi-Fi system right ?

 

Not really. They would do that if, as part of the mastering process, they were bouncing to CD quality audio. It depends on what your target platform is, eg. CD, mp3, 24bit .wav, et al. Although if you were bouncing to CD quality, you may well want to check out how mp3s encoded from that PCM file would sound at varying bit rates, as consumers would be likely to import their CD tracks into an mp3 player.

 

That's my layman's understanding of it anyway.

 

Thanks for upper thread. It was good question that about dithering audio before converting to mp3.

However...you wrote "It won't be the sample rate and bit-depth of the project; it will be the sample rate and bit-depth of the .mp3 you exported"

Well I can choose 320 kbps as my highest quality of mp3 files. It' pretty straight forward

[Rev. Juda Sleaze]

However...you wrote "It won't be the sample rate and bit-depth of the project; it will be the sample rate and bit-depth of the .mp3 you exported"

Well I can choose 320 kbps as my highest quality of mp3 files.

 

Note the bit of your quote I put in bold type. You could bounce to mp3 of all manner of bit-rates, I was just trying to be clear.

 

It' pretty straight forward

 

Cool, then you're good to go

[Logicno8]

However...you wrote "It won't be the sample rate and bit-depth of the project; it will be the sample rate and bit-depth of the .mp3 you exported"

Well I can choose 320 kbps as my highest quality of mp3 files.

 

Note the bit of your quote I put in bold type. You could bounce to mp3 of all manner of bit-rates, I was just trying to be clear.

 

It' pretty straight forward

 

Cool, then you're good to go

 

Ok, I got you. So it's really good to have these sessions saved or just wave files (or stams) in full recording resolution since you never know when you will need an mp3 for internet or mono wave for TV or radio, or which ever file it happened to be needed.

[David Nahmani]

Ok, I got you. So it's really good to have these sessions saved or just wave files (or stams) in full recording resolution since you never know when you will need an mp3 for internet or mono wave for TV or radio, or which ever file it happened to be needed.

 

Definitely. That's called archiving.

[Logicno8]

Ok, I got you. So it's really good to have these sessions saved or just wave files (or stams) in full recording resolution since you never know when you will need an mp3 for internet or mono wave for TV or radio, or which ever file it happened to be needed.

 

Definitely. That's called archiving.

 

Yes, that's called "something I missed to learn and now I have to RE-record some songs again"

I mean really, with HDs getting bigger and cheaper, as soon as I come to LA next year and get some decent money to spend I'll make HD archive backup for every single note I record and leave it there in it's "session resolution" until the end of time 8)