Skip to main content ↑ ↓

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: iTunes 10 AAC 160kbps VBR fail? (Read 10005 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

iTunes 10 AAC 160kbps VBR fail?

Hey all,
I'm a long time lurker but since i didn't have anything to contribute I hadn't registered until now...
I recently updated to iTunes 10, and so have started ripping my CDs using it.
I noticed that the common convention around here is that 160kbps VBR iTunes AAC is reasonably transparent, and some even suggests 128kbps VBR is transparent 'enough' already.
However, I recently bought a new CD, Linkin Park's A Thousand Suns (which is a really cool album by the way, not their usual blah blah screamo stuff...) and noticed that the new iTunes fails in at least one of the tracks at 160kbps VBR!
Now, I don't use state of the art equipment... I listen primarily on an iPhone 3GS using an Ultimate Ears Super.Fi 5 Pro... But on track 8 "Waiting for the End", the opening has a distorted electric guitar, and when encoded on 160kbps VBR, it has some very obvious high pitch artifacts.
I did a blind test 10 times against a lossless and picked it out every single time, so I'm definite it's not placebo. It disappears somewhat on 192kbps but it's still there, and on 256kbps, sometimes I can hear it, sometimes I can't - I'm assuming that's just my mind playing tricks on me.

Now, I'm just wondering, is this common? Anyone else experience this problem when using iTunes? Is there a better, more efficient encoder out there?
See, the codec fails on a very common sounding sound - electric guitar - pretty big flaw to me? Now I'm wondering if 160kbps VBR is good enough... But I read that AAC was designed to be effective at lower bitrates??

Thanks all in advance

iTunes 10 AAC 160kbps VBR fail?

Reply #1 –
Its possible after all this is lossy compression. However i expect such cases to be exceptional  - where you hear huge artifact at 160k. It also depend on your genre but if such bad cases happen a lot (you actually hear artifacts in casual listening ) then this is bad and you either use 192k instead or another format.

The most robust format at these bitrates in my experience is MPC. The default setting averages around 160 .. 180 k and is usually transparent in blind testing and can rival 320k mp3. Artifacts are usually subtle and very rarely heard outside abx tests.

iTunes 10 AAC 160kbps VBR fail?

Reply #2 –
The problem is that MPC is never officially supported on portables because no one licenses it. Ogg Vorbis seems a bit better in this regard (and it's among the best in quality vs. bitrate) but I'm not sure it plays on Apple stuff.

>ppolicar: Have you tried the Nero AAC encoder? If you're around or under age of 20, you may have superior hearing. Waiting about 10 years should "fix" the issue.

iTunes 10 AAC 160kbps VBR fail?

Reply #3 –
The problem is that vorbis at 160 k fails on other samples so you are in the same boat as aac 160. MPC standard is designed not to fail on normal music. If one can use rockbox it is a seriously good option for mid~high bitrates.

iTunes 10 AAC 160kbps VBR fail?

Reply #4 –
Before this thread moves into nuance discussion of codecs, I think it important , in order to help the OP, that  he/she would elaborate on:

Quote from: ppolicar on
I did a blind test 10 times against a lossless and picked it out every single time, so I'm definite it's not placebo.


How was this done?
Using what software?
Done using the phone as media player?

terry

iTunes 10 AAC 160kbps VBR fail?

Reply #5 –
I noticed artifacts too when using itunes aac (128 kbps VBR) and I was able to easily abx it with the original flac 8 out of 8 times (I used foobar2k).
At 160 kbps the situation was a little better. The strange thing is that I didn't have the same problem with nero aac q 0.40 even though itunes encoder should be as good as nero if not better...

My equipment in the test was fairly lo-fi  foobar2k, realtek alc 260 with wasapi dll set to 24 bit output and sony mdr-ex35lp in-ear headphones

The track was the 2009 remaster of "Come together" 44.1kHz 24 bit

iTunes 10 AAC 160kbps VBR fail?

Reply #6 –
hey all thanks for the very quick replies:

1st things first:
how i did the blind test:
1. ripped the track using apple lossless.
2. ripped 160kbps aac version, with a different name
3. created duplicates (5x)
4. made a playlist of all tracks.
5. put tracks into iphone, played playlist on shuffle and repeat, whilst covering the track name with my finger
6. played the song in shuffle
7. after each song is heard, i choose whether or not its lossless, checked the file name and see if im right.
8. determined im right every single time.

primitive, but sufficient proof for me that im not kidding myself

i can provide a snippet if you guys want, but how do i do this? (ie. whats the best place to upload etc?)

I AM 20!! haha...so i guess thats part of the problem...
but really, I only encountered this now, ive been ripping at 160kbps vbr on itunes for around 2 years now, under the perception that theyre practically lossless...
i feel like my favourite toy just broke..
I also just came across a thread that apparently after a certain version of itunes aac's codec sucks....is it possible that itunes 10 broke the aac codec and hence my sweet spot setting??
http://forums.ilounge.com/digital-audio-fo...my-opinion.html

and regarding Nero AAC, no I havent tried it but the consensus seems to me that itunes and nero are as good as each other? am I wrong?

what settings are you guys using? if youre using itunes aac to rip into your iphone devices? (i dont think apple iPhone supports MPC and OGG? so that sort of rules those formats out...how's MP3 going these days??)

iTunes 10 AAC 160kbps VBR fail?

Reply #7 –
OP wrote,
Quote
1st things first:
how i did the blind test:
1. ripped the track using apple lossless.
2. ripped 160kbps aac version, with a different name
3. created duplicates (5x)
4. made a playlist of all tracks.
5. put tracks into iphone, played playlist on shuffle and repeat, whilst covering the track name with my finger
6. played the song in shuffle
7. after each song is heard, i choose whether or not its lossless, checked the file name and see if im right.
8. determined im right every single time.


Guess you missed the very first response from kornchild2002, where he thanks that poster for using foobar2k to properly ABX..

iTunes 10 AAC 160kbps VBR fail?

Reply #8 –
so youre saying i somehow could know whether or not the next track on the shuffled playlist will be lossless or not?

iTunes 10 AAC 160kbps VBR fail?

Reply #9 –
Not saying that at all..But hey, you never know, the proof is in the pudding..Use f2k and submit the results..
That's all I'm saying..Do it properly as is always suggested around these parts..

iTunes 10 AAC 160kbps VBR fail?

Reply #10 –
i would if i could, but i dont have a windows machine...im a mac ... and fubar2k's website says its windows only...

but what do you use personally for encoding?

and how do i upload clips onto here again?

thanks

iTunes 10 AAC 160kbps VBR fail?

Reply #11 –
Sorry not a Mac user, but no doubt someone will chime in..Lately I've been using QuickTime 7.6.7 (qtaacenc) via foobar2k @ ~256kbps VBR..
Good luck!.

iTunes 10 AAC 160kbps VBR fail?

Reply #14 –
You could try out QuickTime's True VBR AAC encoder by using Max.

It's possible to use LAME on the Mac version of iTunes, which is a highly tuned Mp3 encoder which produces transparent results at around 180kpbs - 220kbps with most music. Also iTunes Mp3 encoder is not too bad at 192kbps with Joint Stereo enabled.
"I never thought I'd see this much candy in one mission!"

iTunes 10 AAC 160kbps VBR fail?

Reply #15 –
sweet as guys...
ill try all of the above soon and post back
just out of curiosity, what do you enthusiasts use as your AAC setting for your lossy rips??

thanks

iTunes 10 AAC 160kbps VBR fail?

Reply #16 –
Quote from: /mnt on
It's possible to use LAME on the Mac version of iTunes, which is a highly tuned Mp3 encoder which produces transparent results at around 180kpbs - 220kbps with most music.
The problem is that LAME at this setting equals to about 18kHz roll-off low-pass filtering which might sound offending to his 20yo ears (I remember hearing over 20kHz at that age). I've played a bit with the iTunes AAC encoder recently and found out that at 192kbit setting it includes stuff up to 20kHz while at 160kbit it does much more aggressive low-passing (it cuts even to my hearing range).

iTunes 10 AAC 160kbps VBR fail?

Reply #17 –
Quote from: Martel on
The problem is that LAME at this setting equals to about 18kHz roll-off low-pass filtering which might sound offending to his 20yo ears (I remember hearing over 20kHz at that age).

As part of a complex music signal?

iTunes 10 AAC 160kbps VBR fail?

Reply #18 –
Quote from: /mnt on
You could try out QuickTime's True VBR AAC encoder by using Max.

It's possible to use LAME on the Mac version of iTunes, which is a highly tuned Mp3 encoder which produces transparent results at around 180kpbs - 220kbps with most music. Also iTunes Mp3 encoder is not too bad at 192kbps with Joint Stereo enabled.


While MAX is a decent product  XLD is superior and is kept up to date far more frequently. http://tmkk.hp.infoseek.co.jp/xld/index_e.html

iTunes 10 AAC 160kbps VBR fail?

Reply #19 –
Quote
1. ripped the track using apple lossless.
2. ripped 160kbps aac version, with a different name


Could this be a ripping error? To be sure it's not, just convert the lossless file to AAC.

iTunes 10 AAC 160kbps VBR fail?

Reply #20 –
Quote from: 2t0nEg on
Guess you missed the very first response from kornchild2002, where he thanks that poster for using foobar2k to properly ABX..


What?  I haven't even responded to the OP yet or participated in this thread.  Boom!  Random name drop.

In all honesty, I doubt there is really anything that drastic going on here.  I too purchased that album (though I highly regret it, those bastards) and tried the track at 160kbps VBR iTunes AAC on my system.  I conducted a quick blind ABX test with my Bose headphones and did not pass.  Then again, I did this in a period of about 10 minutes as my legendary copy of Halo: Reach is sitting their begging to be played.  I could have concentrated a bit more to really try to pinpoint something but I didn't.  I just gave it a casual listen.

iTunes 10 AAC 160kbps VBR fail?

Reply #22 –
Quote from: Slipstreem on
As part of a complex music signal?
There's usually just hi-hat and cymbals there (and a bit of harmonics from electric guitar effect) so it's often easier to pick out than I would like it to be.
 

iTunes 10 AAC 160kbps VBR fail?

Reply #23 –
Quote from: greynol on
Did you check the ilounge link ppolicar gave?


Ah, no I didn't.  I was looking in this thread the whole time.  It also doesn't help that the iLounge thread they linked to was started in 2007.  I believe Apple has updated their AAC encoder quite a few times since then.