gapless playback not so gapless

[whitebrotha]

It seems to me that the gapless playback ain't so gapless. I still can hear the gaps (altough they become much harder to notice). I check so both "seamless playback" and "remove leding.." is marked but still the problem remains. Is it maybe because i don't use the crossfade (only the gap remover) ??

[hedge]

It seems to me that the gapless playback ain't so gapless. I still can hear the gaps (altough they become much harder to notice). I check so both "seamless playback" and "remove leding.." is marked but still the problem remains. Is it maybe because i don't use the crossfade (only the gap remover) ??
I take it you're playing mp3's. normal mp3's by there nature do not support gapless playback, and 'hacks' like those used in qcd and winamp don't dispose of the gap entirely, they just make it a lot less noticeable. If you encode files with newer version of LAME on the other hand, then you can get completely gapless playback by using a player that supports it (which at this stage is really only foobar, or if you use case's MAD input plugin for qcd...).

But yes, you also might like to try upping the buffer's in the pref's, that might get it a little better (and if its happening with non-mp3 files like ogg's, then this is the reason why its not working properly).

[Roj]

I take it you're playing mp3's. normal mp3's by there nature do not support gapless playback, and 'hacks' like those used in qcd and winamp don't dispose of the gap entirely, they just make it a lot less noticeable. If you encode files with newer version of LAME on the other hand, then you can get completely gapless playback by using a player that supports it (which at this stage is really only foobar, or if you use case's MAD input plugin for qcd...).

But yes, you also might like to try upping the buffer's in the pref's, that might get it a little better (and if its happening with non-mp3 files like ogg's, then this is the reason why its not working properly).

Actually, gapless playback her never worked with either the stock XAudio plugin or Shaohao's MAD plugin. It does work with case's MAD plugin.

[whitebrotha]

when i playback lame-encoded mp3's in foobar2000 and use its "remove gaps" and "skip silence" dsp plugins, i have no problem, same goes when i playback it in billy [a small player from http://www.sheepfriends.com/]. And i encode with the gapless switch...

[Roj]

when i playback lame-encoded mp3's in foobar2000 and use its "remove gaps" and "skip silence" dsp plugins, i have no problem, same goes when i playback it in billy [a small player from http://www.sheepfriends.com/]. And i encode with the gapless switch...

The current crop of playback plugins don't really suport gapless. There is a MAD plugin by case that is in "extended beta" that supports gapless play and the BASS plugin that is currently under development (I hope) also supports that function. Hopefully at least one of those will reach final status within a relatively short march of days.

Another solution would be to use the Winamp Plugin Manager and one of the mpg123 variants (I don't like the sound quality - too harsh and tinny for my taste - but you may like it) that supports that function.

[whitebrotha]

actually i don't like mp3 at all. I prefer .mpc, .ogg & .flac when it comes to ripping my own music (i only download mp3).

[Roj]

actually i don't like mp3 at all. I prefer .mpc, .ogg & .flac when it comes to ripping my own music (i only download mp3).

I do mp3, ogg, ape and flac. If I had to pick one that was used least, it unfortunately would be ogg. mp3 gets used because of the several portable mp3 players in the house (both flash and CD-based), my car deck and internal mp3 server (no access allowed to anything outside my net). My minimum standard is 192-bit / Stereo / HQ / non-Xing (LAME variants and FhG are acceptable).

// soapox on //

I dismissed mpc as a format having no future a long while back. I predicted it would fade away due to lack of acceptance and because its proprietary closed nature would prove a financial obstacle to hardware vendor acceptance (the author used code that required licensing fees). As it turned out, time proved me right (I have a feel for this sort of thing :) ). Development stopped when the dev threw in the towel at least a year or so ago. I also predicted that ogg audio quality would rise to equal and then eclipse mpc and that too is well down the road to becoming fact.

The big obstacle to widespread hardware vendor acceptance of ogg today is something I didn't foresee back then: DRM. That greed-driven filth has messed up the whole open nature of digital music. Because of that abomination we likely now will not see widespread acceptance of an open format that by nature is not intended to be "secured" by the greedy thieces which run the entertainment industry (read: "RIAA").

// soapbox off //

[whitebrotha]

well, Musepack actually recently got updated to version 1.15 from 1.14 and it's excellent for for especially rock, jazz & techno (if found less useful for hiphop, classical & opera).

But since it lacks hardware support it's not my favorite choice.

[m0rbidini]

Musepack is Open Source (LGPL) and so far no one was able to find any valid patent in it (Frank Klemm and others at Hydrogenaudio tried to find but didn't succed).

Years of listening tests carried out in r3mix and Hydrogenaudio make it possible to say that Musepack is still the format with the smallest number of problematic samples and the one that scales best in medium/high bitrates.

I'm a big supporter of Vorbis but Musepack still manages to achieve transparency at typically lower bitrates than Vorbis (once again, check HA for recent tests).

Regarding DRM: it's no problem to apply it to Vorbis, contrary to what you may think.

Cya

[Roj]

Musepack is Open Source (LGPL) and so far no one was able to find any valid patent in it (Frank Klemm and others at Hydrogenaudio tried to find but didn't succed).

Where did the reports of it containing commercial code start then? I remember hearing this several years ago... Was the source changed a la LAME to remove commercial code? Also, take note that it's not a PATENT you should be looking for but a COPYRIGHT. A patent is for an expression of an idea while a copyright is for the code necessary.

Years of listening tests carried out in r3mix and Hydrogenaudio make it possible to say that Musepack is still the format with the smallest number of problematic samples and the one that scales best in medium/high bitrates.

True enough but it still isn't supported by any hardware with no prospects for that on the horizon either.

Regarding DRM: it's no problem to apply it to Vorbis, contrary to what you may think.

Well and good but it's also unlikely that it will happen. It's far more more likely to happen with a proprietary format than something open source.

[Aaron]

Well and good but it's also unlikely that it will happen. It's far more more likely to happen with a proprieatary format than something open source.

Roj is right. Vorbis is an open format so extremely unlikely it'll have any DRM associated with it. The whole reason it was developed was in protest over patents with mp3 encoding (as far as I understand).

Aaron

[m0rbidini]

Where did the reports of it containing commercial code start then? I remember hearing this several years ago... Was the source changed a la LAME to remove commercial code? Also, take note that it's not a PATENT you should be looking for but a COPYRIGHT. A patent is for an expression of an idea while a copyright is for the code necessary.

It was believed that Musepack used patented technologies (Philips?), being based on MPEG-1 Layer-2 / MP2. But then some on HA went looking for this and no one was able to find one that didn't expire. Regarding the patent/copyright note: LGPL covers the copyright issue.

Well and good but it's also unlikely that it will happen. It's far more more likely to happen with a proprietary format than something open source.

Roj is right. Vorbis is an open format so extremely unlikely it'll have any DRM associated with it. The whole reason it was developed was in protest over patents with mp3 encoding (as far as I understand).
http://www.sidespace.com/products/medias/
http://xml.coverpages.org/EricksonOpenDRM20020902.pdf

Cya

[Roj]

It was believed that Musepack used patented technologies (Philips?), being based on MPEG-1 Layer-2 / MP2. But then some on HA went looking for this and no one was able to find one that didn't expire.

Did the author himself have any comment? I mean, he should know, right?

Regarding the patent/copyright note: LGPL covers the copyright issue.

If it's his, yes.

http://www.sidespace.com/products/medias/ (http://www.sidespace.com/products/medias/)
http://xml.coverpages.org/EricksonOpenDRM20020902.pdf (http://xml.coverpages.org/EricksonOpenDRM20020902.pdf)

More DRM *filth*, this time from the open source community?

I am *deeply* disappointed - I thought that *they* of all people would know better. Mind you, it is being embraced by HP and few have more proprietary interests than they do... (regional cartridges for their printers?)

[Aaron]

http://www.sidespace.com/products/medias/
http://xml.coverpages.org/EricksonOpenDRM20020902.pdf

Cya

Thanks for the links. I stand corrected.

That sort of thing sucks big time.

Aaron

[Hanzo]

So that means they are turning OGG into a DRM format ???
Or will it be like non-DRM Ogg's and DRM-Ogg ???

Anyhow it sucks...

[Aaron]

So that means they are turning OGG into a DRM format ???
Or will it be like non-DRM Ogg's and DRM-Ogg ???

Anyhow it sucks...

I doubt it - have a look at this slashdot article.
http://slashdot.org/articles/03/04/03/2059223.shtml?tid=141

This comment in particular....

"I assure you that SideSpace is in no way, shape or form affiliated with the Xiph.Org Foundation, who make Ogg Vorbis and other royalty-free multimedia codecs.

Emmett Plant
CEO, Xiph.Org Foundation [xiph.org]"

The DRM thing is not driven by the Vorbis people (the 'they' you refer to) themselves. Rather these SideSpace lot.

Aaron

[Roj]

I doubt it - have a look at this slashdot article.
http://slashdot.org/articles/03/04/03/2059223.shtml?tid=141

This comment in particular....

"I assure you that SideSpace is in no way, shape or form affiliated with the Xiph.Org Foundation, who make Ogg Vorbis and other royalty-free multimedia codecs.

Emmett Plant
CEO, Xiph.Org Foundation [xiph.org]"

The DRM thing is not driven by the Vorbis people (the 'they' you refer to) themselves. Rather these SideSpace lot.

Aaron

Um, the HP logo in the second link is a dead giveaway. Open source? Not in the traditional sense. That's like IBM getting behind Linux and Sun behind OpenOffice. It's happened but each has an ulterior motive and "free and open" have zero to do with it.

[m0rbidini]

Did the author himself have any comment? I mean, he should know, right?
From the official site (http://www.musepack.net):
It is based on the MPEG-1 Layer-2 / MP2 algorithms, but has rapidly developed and vastly improved and is now at an advanced stage in which it contains heavily optimized and patentless code.
So that means they are turning OGG into a DRM format ???
Or will it be like non-DRM Ogg's and DRM-Ogg ???
The latter. It's possible to apply DRM to any format, as a general rule.

Cya

[Roj]

From the official site (http://www.musepack.net/):

In other words, it's a similar situation to LAME. Hopefully the code is completely replaced with no vestiges remaining. Whether or not HA people could find any references is irrelevant if a judge in a copyright case finds code he deems identical. That's the real test. Patents only apply to ideas - not code. The open source LGPL won't protect anyone if that's the case.

The latter. It's possible to apply DRM to any format, as a general rule.

True enough - I just think it's pointless. Sooner or later the majors will have to learn that:

1) Any DRM can and will be cracked; at the very least Moore's Law is against them.

2) The money is better spent finding ways to attract customers, not repel them

If they don't get those concepts, they'll go out of business over time - plain and simple.

[Hanzo]

1) Any DRM can and will be cracked; at the very least Moore's Law is against them.

2) The money is better spent finding ways to attract customers, not repel them

Amen ! I don't think they get that anyhow in any multimedia playback the binary data, encrypted, secured, tunneled, compressed or whatever must go back to analog so humans can use it. And that is the weak point of any protection scheme.

Anyhow someone will sooner or later crack it so you don't have to worry with analog...

:cheeky:

[m0rbidini]

In other words, it's a similar situation to LAME.

LAME is also LGPL but contains patented code. What happened is that they removed all ISO code from it (version 3.81) in order to not infringe copyrights, but the patent issues remain. So, it should be clear that in the patent issue, the situations are completely different. Distributing binaries of LAME is illegal in many countries (similar to XviD), whether or not those that do it end up having problems with it. I only posted for clarity, since some confuse patent issues with copyright.

[Roj]

LAME is also LGPL but contains patented code. What happened is that they removed all ISO code from it (version 3.81) in order to not infringe copyrights, but the patent issues remain. So, it should be clear that in the patent issue, the situations are completely different. Distributing binaries of LAME is illegal in many countries (similar to XviD), whether or not those that do it end up having problems with it. I only posted for clarity, since some confuse patent issues with copyright.

I see - I think. Patents are generally for an idea while a copyright is for the mechanics behind an idea. You can patent an algorithm but you copyright the code to express that algorithm.

[madjo]

I see - I think. Patents are generally for an idea while a copyright is for the mechanics behind an idea. You can patent an algorithm but you copyright the code to express that algorithm.
I hope you meant "You can't patent an algorithm" :)

but if that is true, then why is there such a fuss about software patents?
Of which, I might add, I'm against...
but okay, that is entirely off topic... sorry 'bout that...

[rorythedog]

I hope you meant "You can't patent an algorithm" :)

but if that is true, then why is there such a fuss about software patents?
Of which, I might add, I'm against...
but okay, that is entirely off topic... sorry 'bout that...

No, he means "you can patent an algorithm". I think. :evolved:

[Aaron]

Software patents :(