Help - Search - Members - Calendar
Full Version: is 128kbps 128000 or 131072bps ?
Hydrogenaudio Forums > Lossy Audio Compression > AAC > AAC - General
sven_Bent
whats is the most correct value for 128kbits Nero encode AAC files... 128000 or 131072 ?

I've always used 128000 since i believe Ivan told that it was the correct value. However Winamp is telling me the files is only 125kbits.

i know its the old 1kilo = 1000 or 1024 issue. but what value its the most correct to use?
kornchild2002
Stick with using 128kbps. Keep in mind that the AAC format, by nature, is never truly CBR. So a 128kbps CBR file can actually have an overall average bitrate of 125kbps or even 129kbps. This is the nature of the format and both the Nero and iTunes AAC encoders experience this behavior. Unless you are limited by CPU power, you should really be using one of the VBR presets with Nero's AAC encoder. That way you get more efficiency out of the encoder.
sven_Bent
QUOTE(kornchild2002 @ Mar 10 2008, 21:36) *

Stick with using 128kbps.


eehhh. That what i'm doing

QUOTE(kornchild2002 @ Mar 10 2008, 21:36) *

Keep in mind that the AAC format, by nature, is never truly CBR.


I'm using 2paas vbr

Do you have an answer for my question?

kornchild2002
QUOTE(sven_Bent @ Mar 10 2008, 13:43) *

Do you have an answer for my question?


What? The most correct value to use? To use in what? If you are talking about the correctly value to use in encoding then I am not sure. The conversion factor is that 1024 bytes = 1 kilobyte, that is what the software world uses. Memory manufacturers use the standard of 1000 bytes = 1 kilobyte. I would think that the Nero AAC encoder, being software, already uses the 1024 bytes = 1 kilobyte standard. So 128000 should actually come out to 128kbps (then again, you are using 2-pass instead of CBR or VBR so that means that the overall average bitrate will come out lower than 128kbps anyway).
skamp
Data rates (hence, bitrates) are always expressed in powers of 10, not 2.
128 kilobits per second = 128,000 bits per second.
It's easy to check: Divide the filesize in bits (bytes * 8) by the length in seconds. Due to the added size of headers, the resulting number should be a little over 128,000, not 131,072.
menno
QUOTE(sven_Bent @ Mar 10 2008, 12:43) *

QUOTE(kornchild2002 @ Mar 10 2008, 21:36) *

Keep in mind that the AAC format, by nature, is never truly CBR.


I'm using 2paas vbr

Do you have an answer for my question?


AAC can be completely CBR. Even with a big bit reservoir you should still get close to the requested bitrate (if everything is implemented properly).
Same for 2pass VBR, as you know exactly how many bits are available and how you should distribute them in the 2nd pass.
Next version of the Nero encoder will do this better than current versions. Right now 2pass usually ends up a little bit under the requested bitrate (indeed usually 125000bps for 128000bps) and CBR and ABR go over the requested bitrate.

So to answer your question: use 128000.
sven_Bent
QUOTE(menno @ Mar 10 2008, 22:10) *

Next version of the Nero encoder will do this better than current versions. Right now 2pass usually ends up a little bit under the requested bitrate (indeed usually 125000bps for 128000bps) and CBR and ABR go over the requested bitrate.

So to answer your question: use 128000.


Thanx for the answer. i will stil go and use 128000

It's funny that it's actual the encoder going under bitrate. I thougt it was the winamp plugin using 1024 divisior instead of 1000. 128000/1024 is exactly 125.

Having a slight mental issue about numbers, i will now wait for the next release.
tgoose
QUOTE(kornchild2002 @ Mar 10 2008, 20:49) *

QUOTE(sven_Bent @ Mar 10 2008, 13:43) *

Do you have an answer for my question?


What? The most correct value to use? To use in what? If you are talking about the correctly value to use in encoding then I am not sure. The conversion factor is that 1024 bytes = 1 kilobyte, that is what the software world uses. Memory manufacturers use the standard of 1000 bytes = 1 kilobyte. I would think that the Nero AAC encoder, being software, already uses the 1024 bytes = 1 kilobyte standard. So 128000 should actually come out to 128kbps (then again, you are using 2-pass instead of CBR or VBR so that means that the overall average bitrate will come out lower than 128kbps anyway).

This is bits, not bytes.
probedb
Just remember there are also 8 bits in a byte, not 10.
pdq
There are 10 kinds of people in the world - those that understand binary math, and those that don't.
sven_Bent
[quote name='menno' date='Mar 10 2008, 22:10' post='552027']
ve an answer for my question?
[/quote]
Right now 2pass usually ends up a little bit under the requested bitrate (indeed usually 125000bps for 128000bps) and CBR and ABR go over the requested bitrate.
[/quote]

just did some calculations

track length 2m30s
track size 2.411.693 bytes

thats (2.411.693 *8)bits / (2x60+30)seconds
which again is 19293544bits /150seconds = 128623,626666 bits /second

divide by 1000 = 128,6kbits
divide by 1024 = 125,6kbits


So the file IS actually 128000bit/s
and the winamp decoder most use a 1024bit divisor for kilo and not 1000
~*McoreD*~
QUOTE(pdq @ Mar 11 2008, 23:01) *

There are 10 kinds of people in the world - those that understand binary math, and those that don't.


LoL. Indeed.

If WinAMP shows 125 number then it should have the units Kibit/s

128 Kb/s = 128000 b/s = 125 Kibit/s
benski
QUOTE(sven_Bent @ Mar 22 2008, 06:23) *

So the file IS actually 128000bit/s
and the winamp decoder most use a 1024bit divisor for kilo and not 1000


It's definitely using 1000. Perhaps you're using some really old version of Winamp or a third party MP4 plugin (like the outdated FAAD-based one on rarewares).

Also, MP4 file format has a field where an encoder can store a pre-calculated bitrate value for a track. Winamp (and likely many other players) takes this value on faith. It's possible that the encoder has stored a value of 125000 in this field.
This is a "lo-fi" version of our main content. To view the full version with more information, formatting and images, please click here.
Invision Power Board © 2001-2008 Invision Power Services, Inc.