adaptive bitrate vs. variable bitrate

[Amateur]

i posted this in another thread but realized it should be here so i'll start a thread in this forum in hopes of an answer

is there an audio setting that is similar to constant quality for video? i currently use adaptive bitrate @64kbps with aac profile at he-aac+ps with normalize peaks to 100 checked. if this is a good enough method to use, which is the better profile to use? he-aac+ps or he-aac (no +ps)? this guide (http://forum.doom9.org/showthread.php?s=&threadid=68300#post424070) shows the difference of he and lc aac but doesn't talk about the +ps option.

side note about this, does megui not have the ability to keep original channels if the input is 5.1 audio? whenever i have that source, i have to downmux to dolby otherwise i get an error.

edit: i noticed "variable bitrate" with a q value between 0 and 1 but changing it didn't seem to do anything. at 0.4 the file ended up being 17mb and at 0.8 the file was still at 17mb. am i doing it wrong?

edit 2: ok at 0.25 it actually reduced in size so i'm assuming the lower the number, the less bitrate it uses (opposite of video q value?). so my next question would be is this a better method to use than adaptive bitrate at a set 64kbps?

[SquallMX]

"he-aac+ps" is just for bitrates <32 Kbps, you should use "he-aac" for 40-64 Kbps, and "lc-aac" for everything else.

so i'm assuming the lower the number, the less bitrate it uses

Yes, the lower the number, the lower the quality, and the less bitrate it uses. Most people use values between 0.25-0.45, its a matter of personal choice and the quality of your speakers.

[Sharktooth]

PS is actually enabled up to 48Kbps. however bitrate/quality modes are the same for video. if you require a specific filesize/bitrate then use a bitrate constrained mode otherwise use variable bitrate (quality) mode.

[Amateur]

ah thank you for all the quick responses.

so if i choose 0.25 and lc-aac but my overall bitrate ends up at less than 80kbps, would that be a bad thing? or should i choose lc-acc since using variable bitrate may have peaks above 80kbps?

[Sharktooth]

sorry, wrong words. PS can be enabled up to 48Kbps.

[SeeMoreDigital]

I'm not convinced about the 48Kbps AAC-HE+PS limit.

When comparing AAC-LC, AAC-HE and AAC-HE+PS generated at 64Kbps, the AAC-HE+PS encode sounds fuller to me, when played in my hardware player!

[Amateur]

so what about when using variable bitrate at around 0.25 to 0.30 where my resulting average bitrate equals close to 64kbps. should i use he-aac or aac-lc? is there a downside to using aac-lc at a low bitrate? will the portions of higher bitrate sound bad if i use he-aac?

granted, i'll be watching these shows through my television at first but who knows if/when i will ever upgrade to a sound system of some sort.

[Amateur]

well i'm not trying to produce anything of professional grade. i'm just wondering the proper implementation of the profiles and method. for instance, when doing two test encodes i get two different results

1) he-aac profile using 0.28 variable bitrate gives me an audio file that is 11.3mb, 75kbps, 2 channel stereo, 24khz
2) aac-lc profile using 0.28 variable bitrate gives me an audio file that is 10.1mb, 66kbps, 2 channel stereo, 48khz

i had the same input file for each and have normalize peaks to 100 checked along with keep original channels and keep original sample rate. so why does the first file have a different hertz?

[Midzuki]

http://en.wikipedia.org/wiki/Spectral_band_replication

[Sharktooth]

@IgorC: unless nero devs changed it in the latest versions, i clearly remember PS could be enabled up to 48Kbps (not 35).

[Amateur]

Fine, then follow the opinion of SeeMoreDigital. HE-AAC v2 sounds "fuller" for him. HE-AAC v2 is better then HE-AAC v1. Because it's version V2 and 2 is always bigger than 1. Right?

"Don't forget to use deblock at +6:+6 because it's cool and much better than 0:0. Use merange 64. Why? because it's much better than 32 and 16.
Use 192 khz resampler because it's way better than 48 khz.
Dude, use 32 bit quantization of audio because it sound "fuller" for me than garbage 16 and 24 bit audio and it's way cool".

well my response was not aimed at your post two above mine which talked about the studies about the min and max bitrates for the given profile. it was aimed at the reference to another forum dedicated to audio. i thought my question would be an easy one to answer but i guess not.

i understand (somewhat) the concept of the three profiles and that +ps is for really low bitrate, he-aac is for between 35-64, and that aac-lc is for higher than 64. what i'm confused on is with variable bitrate.

i understand that to be where there is little to no sound, very little bitrate is used and when there is very complex sound, much more bitrate is used. because of this, would setting the profile to aac-lc be bad if there are portions of the file that are well below 64kbps threshold? on the flipside, would there be downside for the portions that are much greater than 64kbps if i used the he-aac profile for the file?

i wasn't trying to discredit your post in any way. i don't think you needed to get upset and delete all of your posts.

[GodofaGap]

If you use q values in NeroAacEnc it will automatically determine a profile. I wouldn't recommend overriding it, unless you notice there is something really wrong.

[Amateur]

i did see that option but since no one mentioned it i figured people were against it

[SeeMoreDigital]

What has happened to IgorC's posts?

[tebasuna51]

i did see that option but since no one mentioned it i figured people were against it
Is the recommended option, put the quality without force the profile (Automatic in MeGUI)

What has happened to IgorC's posts?
He delete their posts.