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: What is going on with Vorbis? (or Xiph overall) (Read 91999 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

What is going on with Vorbis? (or Xiph overall)

Reply #25
The last thing Ogg Vorbis needs now is Ogg Vorbis II, it would fragment its already small market share.

>they work on it someting like a hobby... probably they don't even
> know the potential that their codecs have in the world... is a shame...

You would expect more enthusiasm if it was a hobby, atleast someone would spend a Saturday afternoon on it.

To sum up Ogg Vorbis, it is a ship without a captain, with engine trouble, there are a few people in the boiler room auToV, Blacksword, etc hammering away to keep the engines turning, but with no one up on the bridge, that ship is not going anywhere.

What is going on with Vorbis? (or Xiph overall)

Reply #26
The last thing Ogg Vorbis needs now is Ogg Vorbis II, it would fragment its already small market share.

>they work on it someting like a hobby... probably they don't even
> know the potential that their codecs have in the world... is a shame...

You would expect more enthusiasm if it was a hobby, atleast someone would spend a Saturday afternoon on it.

To sum up Ogg Vorbis, it is a ship without a captain, with engine trouble, there are a few people in the boiler room auToV, Blacksword, etc hammering away to keep the engines turning, but with no one up on the bridge, that ship is not going anywhere.


Unfortunately I must say I agree. It's sad, I have more than half of my music collection in Ogg Vorbis and I fear within the next few years hardware support will become worse than it is today and Vorbis will be obsolete because there is no work put into it. Sure, tuning projects like aoTuV and Lancer has kept Vorbis alive for the last couple of years, but how long will they have time to put into it? I just wish there would be more talented people who could contribute...

/Kef

What is going on with Vorbis? (or Xiph overall)

Reply #27
As a successor/alternative to mp3, Vorbis has been overtaken in a remarkably short period of time by AAC, not in the least because of AAC's cheap and flexible licensing model (much better than mp3, for example).

I feel that on the whole, the audio world should be very thankful to Xiph even if Vorbis should die as an actively developed, mainstream format: by just being there, as good as it is, the project has put pressure on the competition (mainly AAC, and to a lesser extent, WMA) to keep their licensing costs low enough to prevent users (i.e., application and hardware developers) from choosing Vorbis as the 'post-mp3' codec. In the end, this means cheaper products for everyone (a free iTunes/WMP/Winamp wouldn't have been possible under an MPEG-2-type license!)

The $25.000 cap on AAC decoding license fees, for example, is a flippin' bargain for any decent-size hardware or software developer. Even Vorbis can't match that as any big company using it would want to do at least some legal due diligence (with the Vorbis no-patents claim), and your average lawyer wouldn't even step out of bed for a fee like that.

What is going on with Vorbis? (or Xiph overall)

Reply #28
Yea, all this doom and gloom about Vorbis is disconcerting to see.  Living in the here and now, tests have shown you can't get any better for at least the 80 to 128 kbps range.  Since Vorbis is an open format, 3rd party support should never wane, so Rockbox will at least ensure its availability on DAPs in the foreseeable future.  In the meantime, I'm enjoying the hardware support that is available now.  Yes, it would be nice if there was more direction from a leader in order to steer this marvel of a format into the future but hobbyist fascinations come and go.  Live and let live!  If someone steps into a leadership position at Xiph.org and incorporates the latest tunings at least, more power to them.

What is going on with Vorbis? (or Xiph overall)

Reply #29

I just know that the last version sounds great! When I play back q8.5 recordings on my audio U3 I can hear things like the natural tones of the instruments of great jazz players such as Charlie Mingus or Winston Marsillas. It sure as hell kicks MP3's ass and I honestly can't say if 320kbs AAC is better. No bad for a free lossy codec.


It's good to see that you're enthousiastic about the developments, but I have to call TOS#8 here - if you make a statement like this, you need proof (ie, ABX test results).


Well, yes, but let me clarify my original post by slight rewording: "In my opinion I honestly can't say if 320kbs AAC is better ..." I did use the personal pronoun "I" in the original post. Now if I had used the third person impersonal "it" in the original  post (ex: "It is clear that 320kbs..." or "It has been shown that 320kbs...") you would be correct.

Yes, Xiph is in bad shape, but ogg is far from waning. Hardware support for ogg is not currently wanning at all. I  went shopping for a PAP in dec 2006 and I had a surprising choice for PAPs that supported ogg. Most of the models were new to the market. Toshiba, Samsung, Iriver, iAudio etc... Just no options from from the "big 3", Zune, Creative and IPOD.

I don't thing a Vorbis II would fracture the market because there seems to be a such a strong desire for producers of content to find anything but MP3. By producers of content I mean artists(they also don't like middle men taking large chunks out of their revenue), producers, software producers(everyone from decoder writers to media players for PCs creators, etc...) and distributors(mainly online down loadable content shops like Itunes music store etc...).

I understand the point that technically ogg is not zero legal fee cost for a software producer who may want to use it. But the point is probably exaggerated. By the same token, if you think you should pay a lawyer to cover your rear because Ogg's patent may or may not be completely solid, then the same is true for AAC. How can you be sure that the AAC patent is completely solid and the 25$ licensing fee will be the only licensing fee you will be legally obligated to pay? Recent events in litigation have shown that you may not be able to assume that about any claim on any codec!

What is going on with Vorbis? (or Xiph overall)

Reply #30
Quote
Since Vorbis is an open format, 3rd party support should never wane, so Rockbox will at least ensure its availability on DAPs in the foreseeable future.


That's how percieve it is. Open source projects really never die they just slow down to a sluggish halt until they are ressurected by somebody who wants to continue to put the developement time and effort into them. The drawback is that sometimes it takes years. I don't really see Ghost as being anything more then experimental branch in the CVS due to the fact that it would probably be impossible to make it backwards compatible with any of the current source code that's out there. 

I was impressed that they did publish a research paper regarding Speex 1.2 for AES though:

http://people.xiph.org/~jm/papers/aes120_speex_vorbis.pdf
budding I.T professional

What is going on with Vorbis? (or Xiph overall)

Reply #31
I did use the personal pronoun "I" in the original post. Now if I had used the third person impersonal "it" in the original  post (ex: "It is clear that 320kbs..." or "It has been shown that 320kbs...") you would be correct.

He is correct anyways, because to put it simply: "opinions" about soundquality aren't allowed on this board. If someone makes a claim, he HAS to prove it. It is not an option, but a requirement.

To be upfront here: i doubt that you can ABX q3.0 from your stated q8.5 and are just praising your imagination here.
I am arrogant and I can afford it because I deliver.

What is going on with Vorbis? (or Xiph overall)

Reply #32
Hi! I just wanted to say there's currently some noise in the vorbis-dev mailing list. Mostly about multichannel coupling and a possible Ambisonics extension to the "standard". I tried to contribute to the discussion by explaining what the specification allows in terms of joint channel coding and by talking about solutions within the spec that seemed most practicable to me.

But we all know that's not enough for something to happen.


I'd like to pose this counter question:
What could Xiph do with its limited resources?


Cheers!
SG

What is going on with Vorbis? (or Xiph overall)

Reply #33
Get someone besides Monty to merge aoyumi's work with the trunk. And find a way to spark more interest in Vorbis, especially by programmers and hardware manufacturers, to get some improvements in it and more support.

What is going on with Vorbis? (or Xiph overall)

Reply #34
multichannel coupling
That would give vorbis a HUGE boost in usefulness.  At q0 with 6ch it ends up at around ~250kbps, that just can't compete with he-aac smashing 6ch into ~100kbps (q-2 only gets it down to ~160kbps).

If point stereo is any indication (phase stereo isn't even in current vorbis is it?)
q0
1ch 56kbps
2ch 64kbps
in theory, perhaps
3ch 80kbps
4ch 96kbps
5ch 112kbps
6ch 128kbps <- a much more reasonable number

Why arbitray channel coupling wasn't built into vorbis in the first place, I'll never understand.

Multichannel coupling for vorbis is the boost that Ogg needs right now, not so much a fully developed video codec (although I wish tarkin was more real).
Vorbis-q0-lowpass99
lame3.93.1-q5-V9-k-nspsytune

What is going on with Vorbis? (or Xiph overall)

Reply #35
Hi! I just wanted to say there's currently some noise in the vorbis-dev mailing list. Mostly about multichannel coupling and a possible Ambisonics extension to the "standard".


Nothing new there. Monty has been talking about Ambisonics since 2003.


What is going on with Vorbis? (or Xiph overall)

Reply #37
Ghost is somewhat being designed by Christopher Montgomery (Vorbis) and Jean-Marc Valin (Speex), the problem is that Valin wants that Ghost take another direction that Monty wants, but no, Speex II is not Speex with Vorbis I psymodel... AFAIK Speex 1.2 will have the vorbis psymodel.

Maybe Valin can help to clear the things about Ghost?


Let's just say we're hoping to kill two birds with one stone here. It's *way* to early to say how this thing turns out or whether it'll work at all. And yes, Ghost is very different from both Speex and Vorbis.

The issue is not Xiph's projects being dead, it is Xiph itself looking dead.


Not dead, but understaffed. There's simply not enough man-hour for the things we're trying to do. Of course, the place where it shows the most is the website because we've always had problems keeping webmasters around and it's sort of the last thing we care about when strapped for time. Hell, I don't even know how to make non-trivial changes to the Speex site 

Put the other way, Xiph.Org is no more, no less than the sum of all its contributors, plus a bit of infrastructure.

What is going on with Vorbis? (or Xiph overall)

Reply #38
I see that Xiph is mentoring projects for this year's Google Summer of Code.  One of the proposed projects is basically to be Monty's assistant in Ghost development.  There are also several Speex-related project ideas.  To me, this suggests that development at Xiph is still going strong, but like Jean-Marc said, needs more man-power.

http://wiki.xiph.org/index.php/Summer_of_Code

What is going on with Vorbis? (or Xiph overall)

Reply #39
Hi to all,
i was just searching for information about multichannel encoding in OGG and i founded this thread.
Among these great audio developer minds, i would say that i'm with who that push for a reprise of the OGG development, so i really hope something will move on.
And i can say that for me too OGG sounds REALLY great

What is going on with Vorbis? (or Xiph overall)

Reply #40
I wish Monty would be like Josh for FLAC. The latter develops and markets his little baby, while Monty cannot finish anything. He just starts everything, but cannot see it through. He cannot even bother to encourage the integration of Aoyumi-s work into Vorbis. No nurturing of some these volunteers. Nothing. I am an Vorbis user and it will stay that way because the alternatives are still not good enough for me (several reasons), but should I need some other codecs in the future I will try to avoid everything Monty is involved in.

Triza

What is going on with Vorbis? (or Xiph overall)

Reply #41
I wish Monty would be like Josh for FLAC.


If you benefit from the development of open source software, enjoy it.  Make suggestions.  Turn your friends on to it.  Use it at work.

If you aren't directly helping with the development of open source software, then please don't criticize those who are.  Whatever Xiph needs right now to move forward, I'm pretty sure it isn't value judgements.

What is going on with Vorbis? (or Xiph overall)

Reply #42
...Monty cannot finish anything. He just starts everything, but cannot see it through.


Sounds a lot like me. 
Once I figure out how something can be done, actually doing it seems to become boring. I didn't follow developments at Xiph too closely, so what I'm saying is based mostly on what's written above in this thread, but maybe it'd be a good idea to let Monty go on inventing new stuff and then move on, and have someone else finish up after him. On this line of thinking, the Summer of Code project mentioned above could bring some interesting developments.

What is going on with Vorbis? (or Xiph overall)

Reply #43
Why the heck is anyone complaining?

Trust me when I say potential vorbis users do not care about how xiph.org looks. No average joe will ever use vorbis... they don't use aac either except when it's from itunes. Those who have any interest in encoding to vorbis, are not average joes... they don't depend on any fancy homepage... you might know of this encoder called lame... notice any fancy official homepage (with binaries) for it? No, thought so. Noobs encode by pressing the rip button in their program suite. And they would sort of be pissed off if those sort of programs defaulted to "ogg vorbis"  Average users don't have any interest in vorbis, even if you tell them what it is.

Those who actually care hunt down lame, or vorbis if they care, and the recommended versions and settings.

Vorbis is still in development, and has been for a long time. Who cares if it's not monty anymore? Much respect to him but he's moved on and other capable people are now tweaking it. That happens all the time in the open source community and is the beauty of the concept. The linux world is growing, and vorbis with it. It has a decent amount of hardware support, which is actually incredible if you think about it. Many (most?) people here on HA use it. Many games use it. It is used by some anime encoders. It's a success story and it's not going anywhere soon. But yes, it will never be able to compete with AAC no matter how good it is. It caters to a niche and anyone who thought it would turn out differently would've been incredibly naive.

I am a freedom geek, so vorbis is an excellent product to me. It's top of the league in quality (for mono and stereo only, at this point admittedly) and it gives me a feelgood feeling philosophically.

Anyway I don't feel I have much to lose with vorbis. I use it for all my audio. If worst comes to worse and it dies in 5 years time, I can simply convert it to lossless without losing anymore quality. At that time I expect disk space will be a lot cheaper (100 gb bluray discs or something perhaps? ). Simple as that.

What is going on with Vorbis? (or Xiph overall)

Reply #44
you might know of this encoder called lame... notice any fancy official homepage (with binaries) for it?


Why, thank-you! (I created the LAME homepage, and for you ignorant fools, I explain: I did a minimalist design on the developers' request)

And binaries are not available because of patents, not because the lame devs are as lazy as Xiph developers. For all that it's worth, you can find sites hosting binaries at the links page.

What is going on with Vorbis? (or Xiph overall)

Reply #45
And they would sort of be pissed off if those sort of programs defaulted to "ogg vorbis".


I have used Vorbis, but I'd rather not on the whole. But I can't say it would bother me that much if an encoding program "defaulted to" it. I'd just change the setting.

What is going on with Vorbis? (or Xiph overall)

Reply #46
I think this is one of the major problems. There are some people (just average users) who read about Ogg Vorbis, they heard it sounds better then MP3 and they want to try it. So they get to the official webpage and they hope to get an official tool to convert their CDs to Ogg Vorbis. But what do they find? An outdated CLI encoder and obviously they have no idea how to use it. Then there are links to many programs and they have no idea which one to choose.

Many people need to be told what to do. Use program X to create, tag and play the Ogg Vorbis files. This is all what they need. They are confused if there are more choices.

Ogg Vorbis format has all the advantages you can imagine. So why only a few people (including me) use it? Wrong marketing and missing easy to use tools and guides. All the development is not concentrated in one places and there are no good informations for end users. If you want to be successful, you have to concentrate on end users.

Look at Firefox - if you download it and install, it is as simple as possible. No need to set up anything, just use it. If you need more functions, then you are probably an advanced user, who knows how to install the proper extension. Opera has too many functions in default install and so it confuses users. And the market share shows you clearly, what most users want - a save browser which is just working.

Another example is the Matroska project: you have an official MKVToolnix program from Moritz Bunkus, which will convert almost anything to Matroska format. Then you have an official DirectShow splitter from Haali which will enable playback in all DS players. There are also libraries for developers, so they can add matroska support to their applications. The result: Matroska is becoming more and more popular, although it has very hard competition in AVI, MP4 and TS.

An interesting bottom line is, that Matroska gains it's popularity also because of Vorbis. It is the only container (except of the dead OGM format), which can store Vorbis sound and MPEG-4 video.

What is going on with Vorbis? (or Xiph overall)

Reply #47
it will never be able to compete with AAC no matter how good it is.
I think it is doing extremely well for something that is not backed by 'the industry' and has little/no official marketing/promotion (at least I don't see any).

Quote
It's top of the league in quality (for mono and stereo only, at this point admittedly) and it gives me a feelgood feeling philosophically.
Still.. ~250kbps for 6ch audio (q0) still isn't too bad... considering I won't use any less than 256kbps mp2/ac3 for 2ch dvd audio.
Vorbis-q0-lowpass99
lame3.93.1-q5-V9-k-nspsytune

What is going on with Vorbis? (or Xiph overall)

Reply #48
Discussion about the Rarewares redesign moved to this thread.
I'm on a horse.

What is going on with Vorbis? (or Xiph overall)

Reply #49
As already said xiph.org is understaffed - and those people working on xiph.org projects have to make a living with "serious work", too.

Actually I think xiph.org is still functional. It's not a company, it's more group of people trying to provide free multimedia solutions - and they did an do succeed in their mission:

- Vorbis is still a very good audio codec. It's comparing nicely with the most widespread AAC profile (that'd be LC) and outclasses WMA and MP3.

- Theora isn't cutting edge - but it's getting its job done. It's a stable format, perfectly usable now. If you don't want to use MPEG or VC1 in your products you can opt for Theora and together with Vorbis have a complete Audio/Video solution in a free flavor.

- Speex is a very good speech codec. Flexible and powerful. It's even having wide adoption (Valve having it as codec option for in-game communication, Microsoft using it in Xbox Live etc.). It's very nature sorta moves it away from being perceived by end-users - but it's definately making people happy.

- FLAC is doing very nicely. Its compelling featureset  made it an accepted storage format for audio archival.


The other Xiph.org projects obviously are "still there", too, but perhaps not really visible to the end user.


As for Theora: Development has been slow, but it recently made important progress. It now has a fully functional decoder (merged from the experimental theora-exp thingie) and the encoder received some work, too. There's nothing better than writing a patch that get's accepted - now the encoder is using some additional flexibility which elevates it a bit from the VP3 codec (which, by the way, had a few bugs here and there that may very well impact on quality and that are fixed in Theora). And despite rumors saying otherwise: I feel Xiph.org *is* very open towards new developers. It took me like 3 days from deciding to work a bit on the encoder, learning C and getting the new encoder stuff into SVN. That may depend on the project, though - thanks to Theora's "alpha" stage (I feel it's doing better than "alpha" suggests, though) it's perhaps more open to new code than e.g. Vorbis, which needs extensive validation for any change.

There obviously are people outside of Xiph.org thinking that Ogg Vorbis/Theora is useful. It seems Opera and Mozilla will use those for the HTML5 <video> element. Open web, open formats, it's just a nice match.

As for Monty: AFAIK he's working for RedHat and is improving the media situation for Linux - he e.g. worked on the USB sound system in the past IIRC. That's for sure important, too, and will make many people happy. This sadly means that there's apparently little time to e.g. review the AoTuV changes.