Alternative iPod Firmware - Rockbox

[bankshot]

I sent the idea to Apple a few months ago, I guess they either think it's a bad idea and/or they don't care about people who complain about the lack of gapless playback.

I firmly believe that Apple does not care about this. I've sent in numerous suggestions and bug reports, only to be ignored. This also causes me to question the legend of Jobs being such a perfectionist. Wouldn't this bug a real perfectionist?

I know one person who works there (completely unaffiliated with the iPod/iTunes teams) who sympathizes with me, but obviously doesn't have any influence over this issue. The problem is that 99.999% of iPod customers aren't even aware of the problem, let alone care about it. They listen on shuffle 100% of the time, and probably wouldn't notice even if they listened to a "gapless" album in order. Too bad.

Rockbox does seem interesting, but the problem is that it isn't, and never will be, a drop-in replacement for the iPod firmware. It doesn't use the iTunes database on the iPod, so no sync with iTunes and no updates of metadata like last-played-time and play count. I've come to rely on these for building smart playlists, and as much as gaps annoy the heck out of me, I can't bring myself to give them up to get gapless. Rockbox also doesn't play protected AAC (which admittedly only accounts for < 100 of my 5,000 tracks).

 

[shamino]

The problem is that 99.999% of iPod customers aren't even aware of the problem, let alone care about it. They listen on shuffle 100% of the time, and probably wouldn't notice even if they listened to a "gapless" album in order. Too bad.

Bingo! And I'm included in that list.

I almost never listen to tracks in sequence. I have smart playlists that load my 4G iPod mini with random tracks from my Mac's 40G library, and the iPod always plays in shuffle mode. This mode of operation is one of the key reasons for buying an iPod. If I want sequential playback, I can play the CD I ripped the tracks from.

Gapless playback would be nice, but I don't think it will ever be a priority for Apple, unless a lot of customers start requesting it. (And a small number of people repeatedly requesting the same thing isn't the same thing.)

More important than the iPod, I wish my car stereo (a JVC Arsenal model) would offer gapless playback on MP3 CDs, since I often do listen to those discs sequentially. But fractional-second pauses on my Pink Floyd discs are hardly going to be enough to get me to buy new hardware or re-rip the tracks.

 

[bloodycape]

As far as rockbox is concerned the one thing holding me back from an iPod besides the herd mentality is the lack of other codecs. AAC/MP3 is nice an all, but I do have some FLAC and a fair amount of OGG, not to mention some really oddball stuff from old computer games and such that no player ever will be able to hande so I just transcode to MP3.. midi support would be kinda cool too, but now I'm really dreaming!

You should check out stuff from neuros they are runing a linux based firmware that pretty much has native support for all the codec you probably use. Also there is Cowon who also has support the similar codec support to the neuros out the box or with a simple firmware upgrade.

 

[bloodycape]

So.... getting back to the original article...

This Rockbox program adds a few not-very-popular file formats, deletes complatibility with iTMS purchases, and replaces the user interface with one that will confuse the typical user.

In other words, it turns an iPod into a player no different from the thousands of other music players you can buy. There's a reason nobody buys players from Creative, Rio, Sony, Panasonic, Dell or any of the others. And it's not because protected-AAC compatibility is a must-have feature.

This software is a bad joke. I'm sure some die-hard hacker types will use it and love it and try to convert the world, but it's never going to be anything more than a curiosity for the rest of the world.

I dont think that is true at all. If you look at many Asians countries mainly Japan and South Korea, the best selling player isn't the ipod but it is something else.
I hear sandisk doesn't have the best interface in their mp3 players but yet they are #2 in the North American markets. Rio was the only one you mentioned that is gapless and has sound quality that is just as good as the ipod. Their only mistake was bad marketing.

 

[shamino]

I dont think that is true at all. If you look at many Asians countries mainly Japan and South Korea, the best selling player isn't the ipod but it is something else.

I've seen that same statistic. It's my understanding that this is because of a huge price gap between the iPod and locally-grown varieties. Partly due to manufacturing costs, and partly due to government subsidies/tariffs.

I hear sandisk doesn't have the best interface in their mp3 players but yet they are #2 in the North American markets.

Yep. They jumped almost immediately to #2 when they entered the market. Mostly because they make their own flash chips, and so they can get a better price than anyone else.

Apple competes based on ease of use, industrial design and features (with ease of use being the most important). The rest of the world pretty much competes on price alone (except for some absolute disasters, which don't compete at all.)

 

[zach]

sorry, but this is false. there is no reason to go 128->CD->256, other than to waste more time and/or deceive yourself. you cannot go from 128->CD and magically recover lost information. coverting from CD->256 does not give you any more information than you already had in the 128.

please read up on various formats and converting before posting such nonsense.

actually, you're wrong.

if the conversion was 128AAC->CDDA->256AAC, you'd be totally correct: you'd gain absolutely nothing, and barring non-corrected read errors, you'd end up with the exact same audio quality.

however, your analysis does NOT hold true from 128AAC->CDDA->256MP3. in fact, ripping using ANY algorithm other than the original AAC will lose sound quality UNLESS it is lossless. all the algorithms compress using different mechanisms. the OP is actually right in stating that while a CD burnt from 128AAC will lose considerable quality ripped at 128MP3, it will lose quite a bit less ripped in 256 MP3.

back to rockbox... with the exception of two features, i have absolutely no interest.. but why in hell can't Apple support FLAC and gapless? while i usually encode my music in apple lossless, i have no idea if i'm really getting lossless.. iTunes is a notoriously bad ripper, and the same may hold true for apple lossless. and MUCH of the music i listen to (pink floyd, tool, various trance/electronica, live shows) can only truly be experienced gapless. i'd actually PREFER ability to rip apple lossless as one file and add markers, as an earlier poster stated, but they could at least give us FLAC and gapless. it'd be absurdly easy to add gapless, especially since the iPod has two processors, it could decode the streams realtime and splice them, rather than having to decode ahead of time like my computer has to with only one core.

but i digress.

 

[zach]

More important than the iPod, I wish my car stereo (a JVC Arsenal model) would offer gapless playback on MP3 CDs, since I often do listen to those discs sequentially. But fractional-second pauses on my Pink Floyd discs are hardly going to be enough to get me to buy new hardware or re-rip the tracks.

gapless playback on MP3s is not really an option, as the files have an unspecified empty space at the end.. gapless refers to seamless playback of lossless codecs such as FLAC, APE, or apple lossless.

 

[bankshot]

gapless playback on MP3s is not really an option, as the files have an unspecified empty space at the end.. gapless refers to seamless playback of lossless codecs such as FLAC, APE, or apple lossless.

Not quite true. There are a couple of options for gapless with MP3 and similar formats. For one, an intelligently written player could look for silence at the end of the last frame and remove it. I've experimented with this some myself just to see if it could be done and had pretty good results. Second, many MP3 encoders are now putting the exact end time of the track into the tags so that a player aware of this could read that and simply not play any samples past that. Either method works.

 

[lostless]

In Case of you have not noticed, that AAC is gapless. Youll never know it unless you play it on a psp. Amazingly, the psp can play aac gapless that are ripped from itunes. Yet Mp3 does not play gaplessly on the psp. Im saying if apple only allowed gapless on aac, i would be satisfied.

 

[kugino]

actually, you're wrong.

if the conversion was 128AAC->CDDA->256AAC, you'd be totally correct: you'd gain absolutely nothing, and barring non-corrected read errors, you'd end up with the exact same audio quality.

however, your analysis does NOT hold true from 128AAC->CDDA->256MP3. in fact, ripping using ANY algorithm other than the original AAC will lose sound quality UNLESS it is lossless. all the algorithms compress using different mechanisms. the OP is actually right in stating that while a CD burnt from 128AAC will lose considerable quality ripped at 128MP3, it will lose quite a bit less ripped in 256 MP3.

why am i wrong again? sure, i understand that AAC->AAC is better than AAC->mp3, but going from 128aac->CD->256mp3 or 256AAC will not give you anything more than what was already in the 128AAC...isn't that what i said in my original post?

 

[Kufat]

gapless playback on MP3s is not really an option, as the files have an unspecified empty space at the end.. gapless refers to seamless playback of lossless codecs such as FLAC, APE, or apple lossless.

This is true according to the original MP3 specification, but MP3s encoded with LAME have an additional field in the header that indicates their precise length. (It might just indicate how much of the last frame is empty; I'm not sure.)

I've used Rockbox on an iPod 5G. Since I listen to album-oriented rock and live music, I wouldn't purchase an MP3 player without gapless playback. (On the other hand, I don't think I've intentionally used a shuffle function for many years. Shuffling's not for everyone.) Most of my extensive live music collection is in FLAC and SHN, and I don't feel like converting it.
The UI definitely isn't as smooth as Apple's and there are still bugs around, but if I ever got another iPod, installing Rockbox would be the first thing I did.

 

[EvilBeans]

OGG, FLAC and gapless from a Rio Karma owner.

Not quite true. There are a couple of options for gapless with MP3 and similar formats. For one, an intelligently written player could look for silence at the end of the last frame and remove it. I've experimented with this some myself just to see if it could be done and had pretty good results. Second, many MP3 encoders are now putting the exact end time of the track into the tags so that a player aware of this could read that and simply not play any samples past that. Either method works.

Hey folks. I stumbled across a link to this thread on another forum and thought I'd chime in.

As bankshot mentioned, there are ways to adjust for MP3 frame gaps. The 2 ways bankshot described are exactly how the Rio Karma handles MP3: For tracks encoded using LAME, it reads the track length in the header and starts the following track accordingly. For tracks without track length info, it checks for a sudden silence in the very last frame and starts playing the next track accordingly. While I have no experience with Rockbox, I'm guessing they're doing the same thing.

Someone asked how you can play the next track gaplessly if you have to read the header info first? Easy. I think the term is "double buffering". Basically, start buffering the next track into memory while the current track is still playing.

In my experience, though, this kind of adjustment for MP3 tracks works great, but not always perfect. This is where OGG comes into play. OGG Vorbis (and FLAC, too) is natively gapless, i.e. no frame-gap issue as with mp3. Every CD I own that has any kind of gapless transition in it I encode to OGG, and the gapless transitions on my Karma are perfect.

BTW: It looks like OGG is gradually turning from a niche codec to a mainstream codec. From what I understand, the difference in audio quality between OGG and some of the other codecs at high bit rates isn't that huge, however it's at lower bit rates where it really excels. This would explain why a lot of newer flash players support it, even if they don't play gapless.



Disclaimer: No judgements about the iPod are intended or implied.

 

[LxMx]

This is true according to the original MP3 specification, but MP3s encoded with LAME have an additional field in the header that indicates their precise length. (It might just indicate how much of the last frame is empty; I'm not sure.)

That's right. LAME MP3's have supported gapless playback for some time now.

In Case of you have not noticed, that AAC is gapless. Youll never know it unless you play it on a psp. Amazingly, the psp can play aac gapless that are ripped from itunes. Yet Mp3 does not play gaplessly on the psp. Im saying if apple only allowed gapless on aac, i would be satisfied.

Officially, the AAC spec makes no mention of gapless playback. The only encoder I know of that actually does support gapless is Nero, and even then it uses the same workaround that LAME MP3 uses.

See http://en.wikipedia.org/wiki/Gapless_playback#Format_support.

For me, LAME MP3 is the way to go. You get a mature format that has pretty much universal support, as well as high quality files. LAME MP3 consistently scores equal to, and sometimes better than AAC at bitrates of 192Kbps and above in blind listening tests. (See http://www.rjamorim.com/test/multiformat128/results.html, it's a 128Kbps test, but it illustrates my point that AAC isn't that much better than MP3). To my ears, LAME MP3 with "alt preset standard" is indistinguishable from the source.

Rockbox supports playing LAME MP3's in gapless with no problems. The only time I ever use the Apple firmware is to transfer new songs over to my iPod.

 

[shamino]

if the conversion was 128AAC->CDDA->256AAC, you'd be totally correct: you'd gain absolutely nothing, and barring non-corrected read errors, you'd end up with the exact same audio quality.

Actually, it will end up worse. Recompressing, even with the same CODEC at a higher bitrate causes more data to be thrown away with each generation.

Take any track you like. Use the "Convert to AAC" option in iTunes. Run it again on the result, and again a few more times. You'll notice that each subsequent generation sounds worse than the one before. In some cases (especially tracks with crowd noise, I've found) you notice severe degradation after only the second generation.

 

[Evangelion]

So.... getting back to the original article...

This Rockbox program adds a few not-very-popular file formats, deletes complatibility with iTMS purchases, and replaces the user interface with one that will confuse the typical user.

"Typical users" are NOT going to replace the firmware of their mp3-player. Rockbox is primarily meant for geeks

n other words, it turns an iPod into a player no different from the thousands of other music players you can buy.

No it doesn't. Granted, I don't use RockBox (yet), but if I installed it on my Mini, it would still be the Mini. I don't use iTunes at all, and I still find my Mini to be kick-ass piece of hardware.

This software is a bad joke. I'm sure some die-hard hacker types will use it and love it and try to convert the world, but it's never going to be anything more than a curiosity for the rest of the world.

If someone wants to use this firmware, I REALLY fail to see why you need to get your panties in a bunch over it. Really, what is your problem? Some people decided to write a new firmware for bunch of mp3-players (including the iPod). And because of that, we have people coming in and saying "this is a bad joke!". Don't use it then, problem solved as far as you are concerned. But if someone wants to spend their time doing something like this, and if someone else wants to use their firmware it does NOT concern you, so **** already.

 

[matticus008]

This Rockbox program adds a few not-very-popular file formats, deletes complatibility with iTMS purchases, and replaces the user interface with one that will confuse the typical user.

No it doesn't. You can keep the original firmware on the iPod and switch into Rockbox when you need to play other formats or want to listen to a gapless album.

In other words, it turns an iPod into a player no different from the thousands of other music players you can buy.

No, it expands the use of an iPod into the most feature-complete music player on the market, at the cost of some MB of storage.

This software is a bad joke. I'm sure some die-hard hacker types will use it and love it and try to convert the world, but it's never going to be anything more than a curiosity for the rest of the world.

The software doesn't do anything that that most of the rest of the world needs, wants, or cares about. It is by nature for audiophiles and geeks and tinkerers...and for those of us who can benefit from it, it's anything but a bad joke. (Die-hard hacker types think this is too easy, anyway).

Did anyone see the Arctic Desert theme? It looks pretty amazing, if you ask me.

 

[shamino]

If someone wants to use this firmware, I REALLY fail to see why you need to get your panties in a bunch over it. Really, what is your problem? Some people decided to write a new firmware for bunch of mp3-players (including the iPod). And because of that, we have people coming in and saying "this is a bad joke!". Don't use it then, problem solved as far as you are concerned. But if someone wants to spend their time doing something like this, and if someone else wants to use their firmware it does NOT concern you, so **** already.

Wow. Such a violent reaction to a simple observation. One would think you've got your life's savings riding on this hack or something.

Lay off the caffeine and knock off the personal attacks. If you can't make a point without trying to start a flame war, then take your own advice.

 

[Apple Shmapple]

You're impressed because Windows XP has Spider Solitaire? Boy are you misguided...

And who are you to judge what can and can't be used to impress a computer user? I've seen people choose a computer on a lot more superficial grounds than the software.

 

[milatchi]

It seems cool that this has been done, but I think I'll stick to the good ole' Apple default.

 

[Evangelion]

Wow. Such a violent reaction to a simple observation.

Because your "observation" was totally uncalled for? Not to mention lame?

One would think you've got your life's savings riding on this hack or something.

No. I just get annoyed when some people decide to do something, and then we have completely unrelated people come along and attack them for no good reason. What if you decided to make some piece of music (for example) and decided to hand it out to anyone interested. Then someone came along and said "Your music sucks! In fact, it's nothing but a bad joke!". Whoa, thank you for that piece of constructive criticism!

Don't like their project? Fine, don't use it then.

Lay off the caffeine and knock off the personal attacks.

What "personal attacks"? Please point them out. "Personal attacks" would mean that I called you "stupid", "ugly" or something like that. I didn't do anything of the sort.

If you can't make a point without trying to start a flame war, then take your own advice.

You didn't offer any constructive criticism. You just marched in and told that "This project is nothing but a bad joke!". How's that for a "point"?

 

[LethalWolfe]

why am i wrong again? sure, i understand that AAC->AAC is better than AAC->mp3, but going from 128aac->CD->256mp3 or 256AAC will not give you anything more than what was already in the 128AAC...isn't that what i said in my original post?

I think the hair splitting is starting w/AAC and ending up w/MP3. A 128AAC should sound better than a 128MP3. So going 128AAC->CD->256MP3 should sound better than going 128AAC->CD->128MP3. But, for example, the converse should not be true. 128MP3->CD->256AAC shouldn't sound better than 128MP3->CD->128AAC.


Lethal