Has nero changed the channel order in version 7?

[Rockaria]

This thread is made in a special respect to SeeMoreDigital (http://forum.doom9.org/showthread.php?p=744098#post744098).
It's not "my situation", it's everybody's situation!

It's vital we establish whether Nero have made a mistake with their AAC codec using their own encoder first.

Once this has been properly established we can then take the appropriate action!

If we discover Nero have made a mistake at their end. It would be a waste of everybody's time here developing "third party" fixes. Even more so if Nero correct the problem on the next release!

Now.... has anybody taking part in this thread, checked to see whether the Nero7 version of Recode2 generates 6Ch AAC streams the wrong way round. If so can you please confirm your findings?


Cheers

I wish EVERYBODY contribute resolving the issue.

[bond]

there is only one allowed channel order for aac streams defined by the specs. either you use the correct one or your encoder is incompliant, buggy, crap...

[SeeMoreDigital]

At last.... a voice of reason....


We really need to confirm whether Nero7 is handling the channel mapping correctly within Recode2....

Could somebody please use this PAL MPEG-2+AC3 12sec Sample (http://81.98.148.105/Uploaded_Files/Doom9_Forum_files/PAL_MPEG-2+AC3_12sec_Sample.7z) as your source and e-mail me your results please?


Many thanks

[Rockaria]

you use the correct one or your encoder...==>
At last.... a voice of reason....
Bond, I take 'you' as anybody and 'your encoder' as any encoding environment which encodes wrong channel order. Please correct me if I am misunderstanding...

[tebasuna51]

-Test using Nero Burning Room 6 -> Extras -> Encode files with a Test_Chan_5.1.ac3
Nero 6 can't accept wav6chan as input file, I use WaveWizard and Nero Frontends for the wav.

I agree with Bond in "there is only one allowed channel order for aac streams defined by the specs", but other question is the channel order expected by the (buggy?) aac Nero encoder.
This is the expected order:

Nero6
Order Order Order
AC3 AAC WAV
----- ----- -----
Chan 0 FL -> C <- FL
Chan 1 C -> FL <- FR
Chan 2 FR -> FR <- C
Chan 3 SL -> SL <- LFE
Chan 4 SR -> SR <- SL
Chan 5 LFE -> LFE <- SR
Ac3 -> Aac always invert FL and C channels, rest OK
For wav I send to Jonhman, and is included in WaveWizard, the channel remapping:
6->6#move#inactive#multichannel wav to aac
2 0 1 4 5 3
Seems work correctly.

-With Nero Burning Room 7 -> Extras -> Encode files (Ac3 and Wav6, now accepted)

Nero7
Order Order Order
AC3 AAC WAV
----- ----- -----
Chan 0 FL -> FR <- FL
Chan 1 C -> C <- FR
Chan 2 FR -> FL <- C
Chan 3 SL -> SR <- LFE
Chan 4 SR -> LFE <- SL
Chan 5 LFE -> SL <- SR
Now in Ac3 -> Aac only the C channel is OK
For wav need a wav6 remapped like I say in my post about BeSweet:
http://forum.doom9.org/showthread.php?p=733463#post733463

-Sorry SeeMoreDigital, but my Recode2 version is limited to stereo audio encoding (5.1 disabled), and can't send you the test.
The downmixed (no dpl) aac stereo seems ok. FL-SL-C to FL, and FR-SR-C to FR

[Rockaria]

Thanks for the input again tebasuna51. I agree with you in general but have a bit differernt point of views.
I believe the wav(or PCM) is the basic format to decode-to and to encode/play-from. So there is logically no direct connection between AC3 & AAC in transcoding.

My picture is AC3->WAV(PCM) memory buffer or file->AAC. So every decoder is responsible of aligning the channel order to WAV(PCM) format.
And every encoder is reponsible of aligning the channels to it's own format from the WAV format channels. It includes other features of WAV such bit size, sampling rate.

The aac.dll which I guess used in both decoding & encoding seems to be responsible for the channel assignment in both process, which is not the role of any launchers or tools.
And the channel mappings are different in the two versions as we have experienced & stated already.

[dimzon]

My picture is AC3->WAV(PCM) memory buffer or file->AAC. So every decoder is responsible of aligning the channel order to WAV(PCM) format.
And every encoder is reponsible of aligning the channels to it's own format from the WAV format channels. It includes other features of WAV such bit size, sampling rate.

The aac.dll which I guess used in both decoding & encoding seems to be responsible for the channel assignment in both process, which is not the role of any launchers or tools.
And the channel mappings are different in the two versions as we have experienced & stated already.
Agreed by 100%
:goodpost: :sly:

[SeeMoreDigital]

Sorry SeeMoreDigital, but my Recode2 version is limited to stereo audio encoding (5.1 disabled), and can't send you the test.
The downmixed (no dpl) aac stereo seems ok. FL-SL-C to FL, and FR-SR-C to FRThanks for taking the time anyway....

Can anybody else help out with my request please?


Cheers

[Wilbert]

Ac3 -> Aac always invert FL and C channels, rest OK
That's how it should be. Nero6 uses the correct channel ordening, and Nero7 the wrong one.

http://www.hydrogenaudio.org/forums/index.php?showtopic=10986

[tebasuna51]

I believe the wav(or PCM) is the basic format to decode-to and to encode/play-from. So there is logically no direct connection between AC3 & AAC in transcoding.

My picture is AC3->WAV(PCM) memory buffer or file->AAC. So every decoder is responsible of aligning the channel order to WAV(PCM) format.
And every encoder is reponsible of aligning the channels to it's own format from the WAV format channels. It includes other features of WAV such bit size, sampling rate.

The aac.dll which I guess used in both decoding & encoding seems to be responsible for the channel assignment in both process, which is not the role of any launchers or tools.
And the channel mappings are different in the two versions as we have experienced & stated already.
I agree 100% also.

Only want show a tool 100% Nero (Extras -> Encode files) that ignore these principles. Well know input formats, like ac3 and wav6, are encoded to aac without the correct channel remapping.

And, Wilbert, this is also for Nero6. For the end user, the transcode tool must be transparent.

[Rockaria]

..And the channel mappings are different in the two versions as we have experienced & stated already.in the two versions==>between the two versions

v6 aac.dll alignes & decodes to wav format channel order correctly and plays well on almost all decoders including h/w so it looks compliant (to the standard & spec)
There would be no problem using the v7 in a closed(paired encoder & media & decoder) environment only and looks incompliant so far.

[SeeMoreDigital]

Is there nobody here able to generate an Nero7/Recode2 encode?


Cheers

[zzspectrez]

Is there nobody here able to generate an Nero7/Recode2 encode?


Cheers
http://www.mytempdir.com/301384


zzSPECTREz

[SeeMoreDigital]

http://www.mytempdir.com/301384Hi zzspectrez,

Thanks for the Nero7/Recode2 sample... I was able to download it this morning :)

Anyway, I've just checked it and the channels have been encoded and are presented in the correct order.

I guess at the end of the day, Nero's only concern will be making sure the channels are presented in the correct order when encoding with Recode2.... Which they appear to be!

Meaning if, the channels are encoded in the wrong order when using third party tools (to do the encoding). Why should they be bothered :eek:


Cheers

[Elias]

Why did Nero have to go and screw with the channel order? Is it a bug, or did they do it with purpose?

[tebasuna51]

Meaning if, the channels are encoded in the wrong order when using third party tools (to do the encoding). Why should they be bothered
Nero Burning Rom ->Extras -> Encode files, isn't a third party tool and encode in wrong order.

[SeeMoreDigital]

Nero Burning Rom ->Extras -> Encode files, isn't a third party tool and encode in wrong order.Can you e-mail an encoded sample please. Generated from my source!

Can you also explain the process you used to generate the sample because as far as I know, unlike Recode2, the Nero application you mention does not accept direct input of .VOB's (or .AC3 streams for that matter). Meaning the source has to be converted using an "third-party" tool first.....


Cheers

[tebasuna51]

@SeeMoreDigital
I send you the e-mail requested. Of course you need believe it is generated with a new install of Nero 7 without changing Nero dll, this is:
aac.dll v3.0.0.3
aacenc32.dll v4.2.1.0

You can verify the channel mapping with the second table in my first post (Nero7).

The Nero Burning Rom -> Extras -> Encode files accept now wav 6 channels, and for me is enough to say is a 100% Nero tool. This is not true for Nero 6 because it not accept wav 6 and for ac3 transcode need the "third-party" Mausau's plugin nxMyAC3.dll (like Nero 7).

Then this method for ac3 -> aac is not a 100% Nero tool, but confirm the internal channel order expected for the encoder.

[bond]

(GarfTop) N7 encoder expects
(GarfTop) LF RF C LFE LR RR
(GarfTop) that's _INPUT_ order
(GarfTop) N7 has a difference INPUT ordering, and that is LF RF C LFE LR RR
(GarfTop) (for AAC encoder)
(GarfTop) AC3 has nothing to do with this
(GarfTop) N7 encoder expects the order I mentioned
(GarfTop) and will output in correct AAC order

fyi, the official channel order of 5.1 wav is LF, RF, C, LFE, LR, RR and you see it matches what nero7 expects, so imho nero6 did it wrongly (as it needed a wrong input .wav) and nero7 does it correctly now

that is of course only if nero7 creates a correct aac channel mapping (C, LF, RF, LR, RR, LFE) from a correct wav (LF, RF, C, LFE, LR, RR), so does nero7 do this?

[Rockaria]

14496-3 (http://sound.media.mit.edu/mpeg4/audio/documents/) and P18 on 2203.pdf (ftp://ftp.tnt.uni-hannover.de/pub/MPEG/audio/mpeg4/documents/w2203/w2203.pdf) states for 6(5.1)ch which seems to be the actual order in the content body, (C, LF, RF, LR, RR, LFE) :
center front speaker,
left, right front speakers,
left surround, right surround rear speakers,
front low frequency effects speaker

IMHO, what (GarfTop) says seems to be what it is supposed to be.
Based on my(our) tests, the v6 aac.dll expected FL_FR_C_LFE_SL_SR(wav standard) and the v7 aac.dll asked the remapping of FR_C_FL_SR_LFE_SL to be played correctly on many many ;) players.

[bond]

well thats what he said after i asked him

as they are hardly around on doom9, i guess you should ask him, or ivan dimkovic, on hydrogenaudio

[Rockaria]

i guess you should ask him, or ivan dimkovic, on hydrogenaudio Well, I have no problem myself about the issue here in this thread.
Everything is actually to be derived from the standard and verified by personal experiences. I do not doubt clearly stated experiences if it concures my own or common sense.

I believed v6 encoded contents plays correcly on h/w decoders because it was said by somebody else's experience, should I doubt it?

[edit]
Before tebasuna51's last post, I actually posted & deleted the message right away because I didn't want anybody to get confused or any mistakes left uncorrected by original issuer before it goes too hard.
I verified the emails delivered to everybody in this thread by tebasuna51' post after that.

There seems to have been no system to verify if the attached mp4 file is v7, recoded by recode 2 or not or even if it is recoded with different combinations of nero dlls.
The current v7 nero encoder(aacenc32.dll) writes a 'tool' meta tag as 'nero 4.2.1.0'.