Dirac 1.0.0 Released

First? by Anonymous Coward · 2008-09-20 06:15 · Score: -1

Maybe?

For low values of success by Naughty+Bob · 2008-09-20 06:17 · Score: -1, Redundant

The BBC have already successfully used the new codec during the Beijing Olympics...

With success like broadcasting the olympics, who needs failure?

--
"Be light, stinging, insolent and melancholy"

Re:For low values of success by Anonymous Coward · 2008-09-20 06:29 · Score: 0

Come again?
Re:For low values of success by owlnation · 2008-09-20 06:52 · Score: 0

Unfortunately using the new codec will probably not help the BBC much. Once upon a time a BBC camera department training was regarded as being one of the best in the world. BBC cameramen were highly sought after.

Nowadays, you'll be very lucky if any of them know where the iris control or white balance is on a camera. Id guess that at least about 40% of all shots from news or sports footage is overexposed by at least 2 stops.

In addition, they have been very slow to make the change to HD, most other countries' TV companies changed over far faster than they did. Much of their "quality drama" output is still shot on 16mm film rather than 24p HD too -- mainly due to resistance to change from Camera depts, and a lack of talent therein.

Why even bother with a new codec, unless it's cheaper -- no-one in the BBC seems to care a damn about visual quality.
Re:For low values of success by ATMD · 2008-09-20 07:01 · Score: 4, Insightful

Could it be that the BBC's slowness to offer HD is related to the fact that most license payers receive their broadcasts via analogue or "Freeview" digital, neither of which currently support it? I guess they have better things to spend their limited budget on.

--
Nobody else has this sig.
Re:For low values of success by lxt · 2008-09-20 07:35 · Score: 2, Insightful

16mm is a perfectly viable format for 24p HD, so your question is something of a contradiction. Sure, the quality might not be as good as it could be, but it serves a niche market for low budget HD output on 'quality' drama. Comparing the BBC's HD drama output to the stuff in the States is disingenuous, not least because HD penetration in the UK is extremely low but also because shows in the UK tend to run on lower budgets anyway than prime-time US serials.
Re:For low values of success by Anonymous Coward · 2008-09-20 07:38 · Score: 5, Insightful

In a nutshell, yes. HD is also not much of a big deal to the vast majority of television viewers either. The only reason there is such a fuss over it in the United States is mostly because they are rolling out HD and digital at the same time: most of the improvement has come from the change to digital, not HD. In Europe it's not such a big deal because we've already switched to digital. HD is "nice" but it's not the huge leap in visual quality some people would like you to believe.
Re:For low values of success by PopeRatzo · 2008-09-20 07:40 · Score: 2, Funny

Id guess that at least about 40% of all shots from news or sports footage is overexposed by at least 2 stops.
And I thought that was some kind of UK New Wave artsy Duane Hopkins/dogma thing they were going for.
Once again, I mistake incompetence for artistic innovation.

--
You are welcome on my lawn.
Re:For low values of success by ttlgDaveh · 2008-09-20 07:51 · Score: 5, Insightful

Someone's evidentially not been watching Top Gear, which features some of the best camera work on TV and film.
Re:For low values of success by RonnyJ · 2008-09-20 08:38 · Score: 2, Interesting

I strongly disagree that "most of the improvement has come from the change to digital, not HD". TVs don't magically become a higher resolution when you add a digital decoder! The main benefit(?) of digital has been more channels.
I see a huge difference in quality between SD and HD. The most damaging thing for HD that I've seen is that many retailers used to play SD content on HDTVs, which isn't particularly suited for a TFT/LCD screen and can look terrible.
Re:For low values of success by owlnation · 2008-09-20 09:24 · Score: 1

Could it be that the BBC's slowness to offer HD is related to the fact that most license payers receive their broadcasts via analogue or "Freeview" digital, neither of which currently support it? I guess they have better things to spend their limited budget on.
Who on earth modded that insightful? It makes no sense! It's completely wrong. It has nothing to do with the broadcast format, but the recording format. Using 16mm film is FAR, FAR more expensive than HD video.

It also means that other countries are less likely to buy their output. Thus ending in a downward spiral of quality, due to less money being available for production.

The changeover has nothing to do with cost, and everything to do with resistance to change and a erroneous, misguided elitism that film is better. 35mm might be for some things, but 16mm is dead tech, other than as a training medium for 35mm.
Re:For low values of success by zyzko · 2008-09-20 10:11 · Score: 1

Not to mention that Top Gear looks absolutely stunning in HD - check the Polar Special, it is definitive HD demo material.
But the grandparent has a point. Digital terrestial TV is kind-of half baked in most European countries because they went with DVB-T (MPEG 2). (As did my country, Finland). That means that without new boxes the terrestial customers are left with SD until the next "big change". And since the as in the UK and in Finland there are similiar terrestial networks the national TV companies (BBC in the UK, YLE here) are not that interested in producing HD content for cable customers. YLE did the Beijing Games in HD as a test for a small ammount of terrestial customers living withing coverage of one broadcast tower, though.
I'm not an advocate of delaying transition to digital TV, but in countries where there is a strong public tv company (like in UK and in Finland) it might have been better to wait for DVB-T2 and MPEG4 and transition to HD at the same time. Here in Finland YLE promises HD content maybe in 2016, until that it is cable and satellite and no public tv in HD I fear...
And to those who think public TV is crap and I should go for commercial cable / sat receiving anyway: What we have from YLE here is quite good actually - quality news and documentry (of course to be applied with the normal critical thinking and not believing the only source) and for an example recently they bought the rights for all the HBO shows and BBC material is also quite common. HD would be very, very nice, thankyou...
Re:For low values of success by hairyfeet · 2008-09-20 10:14 · Score: 3, Interesting

Yes,it is nice to have HD,the problem is that the big companies will ruin it by compressing the hell out of it to squeeze in more channels. personally,seeing as how many artifacts end up in the overly compressed HD,I'd personally rather have uncompressed SD than HD,thank you very much. I'm just lucky I'm on a small cableco that is going to stick with SD until they are finished upgrading their network,which they figure will take around 2 years. New servers,lots of fiber being laid,and with each new piece my Internet connection gets a little faster and snappier. But if all the providers start compressing the hell out of the HD signals I don't see HD adoption taking off. Who in the hell would want HD if they compress it so bad it looks like a low bitrate .wmv? But as always this is my 02c,YMMV

--
ACs don't waste your time replying, your posts are never seen by me.
Re:For low values of success by Relyx · 2008-09-20 12:14 · Score: 1

These days HD content is more likely to be recorded onto a high end digital video format like HDCAM. Many of the cameras are descended from the old DigiBeta camcorders seen on news cameramens' shoulders. There are also cameras such as the Sony F23, Thomson Grassvalley Viper, the Panavision Genesis and the Arri D20 which bear a closer resemblance to those from the world of film. These are mostly used on features, commercials and music videos.
Re:For low values of success by Anonymous Coward · 2008-09-20 12:48 · Score: 0

Someone's evidentially not been watching Top Gear, which features some of the best camera work on TV and film.
No no, you're right. One show out of the hundreds that the BBC produces invalidates his whole fucking point.
Re:For low values of success by GigaplexNZ · 2008-09-20 13:36 · Score: 1

The main benefit(?) of digital has been more channels.
The main benefit to consumers is less noise in the signal, drastically improving the quality. Resolution isn't the only measure of quality.
Re:For low values of success by mrchaotica · 2008-09-20 18:24 · Score: 1

TVs don't magically become a higher resolution when you add a digital decoder! The main benefit(?) of digital has been more channels.
Exactly! And "most of the improvement" is more channels! Outside of (a subset of) Slashdot and a few A/V forums, nobody really gives a shit about HD!

--
"[Regarding the 'cloud,'] ownership was what made America different than Russia." -- Woz
Re:For low values of success by Anonymous Coward · 2008-09-20 18:46 · Score: 0

Except that all that great camera work detracts from the actual cars shown. I love the show, but recently i've been steering towards Fifth Gear which lets me look at a car from specific angle for more than a fraction of a second.
Re:For low values of success by Anonymous Coward · 2008-09-20 19:12 · Score: 0

Well, the difference I notice with a good HD feed is like the difference between looking at a picture and looking into one. Subtle, but it's there. Unfortunately few of the "HD" providers really seem to actually use it. (Or know how?) HD is actually noticible on a couple Discovery HD shows and some special HD programming for PBS. (Amazingly enough.) But a lot of the other channels that say "HD" don't really seem to embrace it properly. The few shows that say "HD" on the big 3 networks seem only to be a reframed and up-sampled standard resolution image. Bigger, the colors are smoother, and it tends to fit the screen better than what the TV automatically does with a SD channel. But you notice that it fails in comparison to a good HD feed because it completely lacks the depth and sharpness.
The fact that cable companies can also garbage up a good HD feed with their compression, and that the major networks tend to scrimp on HD quality except for a few special events or some sports programming doesn't help any. I suppose blu-ray doesn't help either, a more open format for recorded media might help HD better than what networks and cable are doing.
Re:For low values of success by delt0r · 2008-09-20 19:59 · Score: 1

The same people that think youtube videos are good enough or even good. Really a lot of people don't see the artifacts. I showed my wife what some look like and she kicked me, because now she can see them too. I also find that when you up the bitrate for "artifact" free encoding, h.264/mpeg4 etc all perform about the same. The differences come out at low bitrate where I'm not all that interested.

--
If information wants to be free, why does my internet connection cost so much?
Re:For low values of success by Anonymous Coward · 2008-09-21 00:17 · Score: 0

Top Gear's camera work annoys the shit out of me with their endless colour filters, "interesting" angles and depth of field effects. It's one of the few programs where you keep noticing the camera work.
It's like they have a checklist of effects that must be in each show.
Re:For low values of success by Fred_A · 2008-09-21 00:25 · Score: 1

Especially since (in Europe at least) most people don't use wall sized TVs and hence don't care all that much about resolution.
Although maybe 70inch diagonal sets may be common in the US... Here I don't know anybody who would see the point of going beyond 35 or 40.

--

May contain traces of nut.
Made from the freshest electrons.
Re:For low values of success by GonHiDi · 2008-09-21 00:46 · Score: 1

Strictly speaking, DVB-T defines only the transmission aspects. As such, DVB-T2 may introduce notable changes but it seems to still use the MPEG-2 transport stream and thus will likely refer to the same specifications as DVB-T. It is true that as far as codecs are concerned it was not until 2005 (publication date of ETSI TS 101 154 V1.6.1) that optional support for improved AV codecs was introduced (H.264/AVC and HE AAC, with VC-1 added in version 1.8.1). However, aside from the resulting chicken-or-egg problem, this does not preclude DVB-T stations from using the newer codecs, and some are already doing so.
I too feel that the deployment is somewhat shoddy, but the theory is that there is no need to wait for a big change because buying a new decoder is relatively inexpensive. I don't quite agree because apart from being wasteful it is consumer-unfriendly to discover that an HD TV set sold as DVB-capable doesn't support some (HD) content because of the codec (beside the fact that it might not support HD DVB-T streams at all depending on whether it is HD ready, HD TV, Full HD or whatever the current marketspeak is), and that after somehow upgrading to get over that deficiency one is bound to learn that the system still doesn't support MHP or whatever technology under the DVB umbrella gets highlighted in the following months.
Re:For low values of success by WhiteHorse-The+Origi · 2008-09-21 03:50 · Score: 0

Actually HD is to SD as CD-Quality Audio is to mp3. Take a Britney spears song and listen to it on mp3 and then CD: No difference. Now take Pink Floyd and do the same: HUGE difference. The difference is in the details that the artist can convey to the audience. CD-audio is an almost identical sound as a live performance, whereas mp3(64kbps) is compressed and sounds like a live performance over the radio. I guess it doesn't matter either way if you have tin-can speakers or live in a noisy building.
I recently watched the "Lord of the Rings" director's cut from DVD at 1024x786(smaller due to fitting the screen), as well as the SD version on TV. There was a lot missing from the SD, entire characters and plot points, in fact. What's really noticeable is when you project it on the wall with a 3meter diagonal and you can see every granularity, every frame loss, etc.I suppose it doesn't matter if you're still using a 500-line CRT, but it's similar to the tin-can speaker analogy.
On the other hand, I don't think your average Joe has a HDTV set or a projector so the technology may be only for a very small crowd...
Re:For low values of success by Fred_A · 2008-09-21 04:57 · Score: 1

Someone's evidentially not been watching Top Gear, which features some of the best camera work on TV and film.
Yes, well, you know, to some of us a show about cars is about as exciting as a show about computers would be to the general public (look ! it's got wheels ! and seats ! whoohoo ! -- yawn)

--

May contain traces of nut.
Made from the freshest electrons.
Re:For low values of success by funkatron · 2008-09-21 05:08 · Score: 3, Insightful

Top Gear isn't about cars, they're just used to give the presenters something to argue about.

--
"Welcome to our world. We are the wasted youth. And we are the future too." Yes, I know these are stupid lyrics.
Re:For low values of success by Anonymous Coward · 2008-09-22 09:03 · Score: 0

Nah in the US the average is probably 30-35 at best, 52 tends to be the standard 'big box' tvs I've seen and anything bigger is usually someone overcompensating, or who has a mansion (if you can sit 10+ feet away, why not have a bigger screen?)
Me personally, I'll be happy with a 42" TV if I ever get the grand saved up. Mostly because it seems to be the sweet spot right now in price vs resolution/size.

really? by Anonymous Coward · 2008-09-20 06:18 · Score: 3, Insightful

Remember when we all used GIF until somebody came out of the closet with a patent claim. How can we be sure about this one?

Re:really? by whathappenedtomonday · 2008-09-20 06:22 · Score: 5, Informative

Read their site. From the FAQ:
Do you infringe any patents?
The short answer is that we don't know for certain, but we're pretty sure we don't.
We haven't employed armies of lawyers to trawl through the tens of thousands of video compression techniques. That's not the way to invent a successful algorithm. Instead we've tried to use techniques of long standing in novel ways.
What will you do if you infringe patents?
Code round them, first and foremost. There are many alternative techniques to each of the technologies used within Dirac.
Dirac is relatively modular (which is one reason why it's a conventional hybrid codec rather than, say, 3D wavelets) so removing or adding tools was relatively easy, even though this may mean issuing a new version of the specification.

--
I hope I didn't brain my damage.
Re:really? by Anonymous Coward · 2008-09-20 07:01 · Score: 3, Interesting

The real question is, how does it fare against good H.264 encoders e.g. x264? And how are the encoding speeds?
The few comparisons I've seen put H.264 as having the edge when it comes to both, but not by a lot.
Re:really? by David+Gerard · 2008-09-20 07:22 · Score: 4, Informative

Encoding and decoding is presently fat and slow. It's very much in development.

--
http://rocknerd.co.uk
Re:really? by whathappenedtomonday · 2008-09-20 10:22 · Score: 4, Interesting

Since you claim this I assume that you tried the 1.0.0 already - I watched the promo vid, and it says the BBC is using the codec to handle HD content over their standard def infrastructure at very low latency (a few ms, if I remember correctly).
Nonetheless, this seems to be an interesting thing to keep an eye on, because the codec specs address good compression especially for very high bandwidths, which is going to be an important issue for movie post production/processing, HD content and the likes. The promo vid is well worth watching.

--
I hope I didn't brain my damage.
Re:really? by Anonymous Coward · 2008-09-21 03:36 · Score: 0

There are hardware implementations of the I-frame only variant (Dirac Pro, aka SMPTE VC-2), that were used at the Olympics. There are currently two implementations of the full Dirac specification, "Dirac", the unoptimised reference implementation, and "Schroedinger", an optimised reimplementation. The latter is what's been integrated into VLC. It's not particularly fat or slow.

Great, but... by Anonymous Coward · 2008-09-20 06:20 · Score: 5, Funny

I tried using the Schrodinger library but I'm uncertain it works. Plus, I can't find my cat.

Re:Great, but... by Anonymous Coward · 2008-09-20 06:22 · Score: 5, Funny

Plus, I can't find my cat.
Be glad for that, as long as you dont you might still have one.
Re:Great, but... by Hangeron · 2008-09-20 06:42 · Score: 3, Funny

I had success integrating Dirac, and got a result of 1.
Re:Great, but... by Anonymous Coward · 2008-09-20 06:57 · Score: 0

TRUE dat
Re:Great, but... by S-100 · 2008-09-20 07:06 · Score: 1

Well, your cat may be OK as long as you don't look for it.
Re:Great, but... by tobiasly · 2008-09-20 07:10 · Score: 4, Funny

I tried using the Schrodinger library but I'm uncertain it works. Plus, I can't find my cat.
I hate to tell you, but your cat is dead. And/or not.
Re:Great, but... by John+Hasler · 2008-09-20 07:48 · Score: 3, Funny

You may have loaded libheisenberg instead. They sometimes become entangled.

--
Warning: this article may contain humor, sarcasm, parody, and perhaps even irony. Read at your own risk.
Re:Great, but... by coopaq · 2008-09-20 13:22 · Score: 1

You just your whole cat! Are you sure this works?
*ducks*
Re:Great, but... by Varun+Soundararajan · 2008-09-20 15:32 · Score: 1

Can you say with certainty it was cat? mow mow
Re:Great, but... by jonaskoelker · 2008-09-21 07:10 · Score: 1

You may have loaded libheisenberg instead.
Are you uncertain of that?
Re:Great, but... by danieltdp · 2008-09-21 16:27 · Score: 1

Man, if you find your cat and it is in a box, just don't open it!

--
-- dnl

Pantsents by CANON by Anonymous Coward · 2008-09-20 06:20 · Score: -1, Troll

CANON pantsentize DiraC.

0xBBCD by hey · 2008-09-20 06:22 · Score: 4, Interesting

I see the first 4 bytes are 0xBBCD.
British Broadcasting Corporation Dirac.

Re:0xBBCD by Anonymous Coward · 2008-09-20 06:35 · Score: 2, Interesting

Did you mean to say the FOURCC (which is usually not the first four bytes) is 'BBCD'? 0xBBCD is usually two bytes...
Re:0xBBCD by Anonymous Coward · 2008-09-20 06:39 · Score: 5, Insightful

Isn't that just 2 bytes? :)
*nibbles on parent's geek card*
Re:0xBBCD by i.of.the.storm · 2008-09-20 06:49 · Score: 1

Haha, nice bit of geek punnery.

--
All your base are belong to Wii.
Re:0xBBCD by Bloater · 2008-09-20 07:00 · Score: 4, Informative

the fourcc is 'drac'
Re:0xBBCD by dougisfunny · 2008-09-20 07:34 · Score: 2, Informative

The size of a byte doesn't have to be 8 bits, though it usually is.

--
This is not the funny you're looking for.
Re:0xBBCD by Yetihehe · 2008-09-20 07:54 · Score: 3, Funny

It doesn't? It's Blasphemy! In the beginning, there was word. And in word there was two bytes. In two bytes there was 16 bits. And the root saw it was good.

--
Extreme Programming - Redundant Array of Inexpensive Developers
Re:0xBBCD by Anonymous Coward · 2008-09-20 07:58 · Score: 0

Well that sure is news to me. I'm gonna punch my assembly teacher in the face next time I see him.
Re:0xBBCD by Larry+Lightbulb · 2008-09-20 10:12 · Score: 1

nYbbles on parent's geek card
Re:0xBBCD by GigaplexNZ · 2008-09-20 13:45 · Score: 1

Assembly is platform specific. When dealing with a specific platform, the number of bits in a byte doesn't change.
Re:0xBBCD by Antique+Geekmeister · 2008-09-20 14:44 · Score: 1

Oh, you'd be surprised. It's amazing what you can see in data formats such as serial communications.
Re:0xBBCD by GigaplexNZ · 2008-09-20 16:32 · Score: 1

That's communicating between different platforms. I was referring specifically to local computations.
Re:0xBBCD by Antique+Geekmeister · 2008-09-20 19:41 · Score: 1

But it's not communicating between different platforms. Or rather, it need not be. The amount of hardware on which a byte equals 8 buts, but still use '9600-7-e-1' or 9600 baud, 7-bit, even parity, one stop bit serial communications on their primary serial port is quite large, even if they're only talking to another such machine. And irrelevant of the internal computation byte size, most machines have to deal with flat ASCII text, which is typically 8-bit bytes.
My point is that byte size is also protocol dependent for all sorts of communications, even platform specific ones.
Re:0xBBCD by Yetihehe · 2008-09-20 23:25 · Score: 1

When only interface you have is serial port, every communication looks like 7bit transmission.

--
Extreme Programming - Redundant Array of Inexpensive Developers
Re:0xBBCD by Antique+Geekmeister · 2008-09-21 01:41 · Score: 1

I'm afraid not. There are quite a few 8-bit serial port devices in use. 9600-e-7-1 is common, 115200-8-n-1 is common as well and seems to work better with international keyboards. The point remains that they vary, and that communications, even between programs on the same machine, may use byte lengths not reflected by the underlying OS's architecture.

Open source overkill by mdmkolbe · 2008-09-20 06:24 · Score: 4, Interesting

From the FAQ:

What are the license conditions?

The Schrodinger software is available under any of the GPLv2, MIT or MPL licences. Libraries may also be used under LGPL.

Sounds like someone wanted there to be no question about whether it was open source.

Re:Open source overkill by damn_registrars · 2008-09-20 06:32 · Score: 5, Funny

The Schrodinger software is available under any of the GPLv2, MIT or MPL licences. Libraries may also be used under LGPL.
Sounds like someone wanted there to be no question about whether it was open source.
Sounds to me like the license exists in multiple states at once, which may be exactly the way Schrodinger would have liked it.

--
Damn_registrars has no butt-hole. Damn_registrars has no use for a butt-hole.
Re:Open source overkill by Anonymous Coward · 2008-09-20 06:34 · Score: 2, Insightful

I am not sure, but isn't MIT one good enough to relicense it to (L)GPL or MPL?
Re:Open source overkill by MrWim · 2008-09-20 06:54 · Score: 5, Informative

The GPL makes assurances regarding patents that the MIT license doesn't:

"Each contributor grants you a non-exclusive, worldwide, royalty-free patent license under the contributor's essential patent claims, to make, use, sell, offer for sale, import and otherwise run, modify and propagate the contents of its contributor version."

So if you use it as a GPL licensed library you can't get sued by the BBC or other contributors to the code.
Re:Open source overkill by Anonymous Coward · 2008-09-20 07:39 · Score: 1, Informative

Yes - GPL says that explicitly, bit MIT (copyright holders) grant you:
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so ...
So they GRANT you to do with it more less anything, even if they own patents and they can not take their word back and saying ... ops, we're sorry, we have some patents, so we will sue you now.
Of course they can not grant you immunity to third party patents, but neither GPL guys...
Re:Open source overkill by calmofthestorm · 2008-09-20 08:30 · Score: 1

Isn't that clause new to the GPLv3?

--
93rd rule of Slashdot: No matter how obvious my sarcasm is, my comment will be taken seriously by someone.
Re:Open source overkill by dapyx · 2008-09-20 09:57 · Score: 2, Informative

No, the GPLv3 has clauses about using the code in a DRM system, the anti-patents clause has been for a long time.

--
I'm sorry, the number you have dialed is an imaginary number. Please rotate your phone 90 degrees and dial again.
Re:Open source overkill by calmofthestorm · 2008-09-20 10:10 · Score: 1

I seem to recall something about a Novell-MS dealbreaker being in the GPLv3 that, if adopted by the kernel itself, would mean a lot of fireworks, but I might be thinking of something else.
I'm definitely not talking about the Tivoization clause.

--
93rd rule of Slashdot: No matter how obvious my sarcasm is, my comment will be taken seriously by someone.
Re:Open source overkill by Anonymous Coward · 2008-09-20 13:06 · Score: 0

Hmm maybe every time observation about license is made the license collapses to one specific eigenlicense? So first step in solving dispute concerning the license random roll of dice is made to decide which particular license is used?

Performance? Benefits? by Enderandrew · 2008-09-20 06:26 · Score: 3, Insightful

How does it stack up to other codecs?

Do we need another codec?

--
http://blindscribblings.com - Tasty pop-culture in conceptual fashion.

fuck muslims by Anonymous Coward · 2008-09-20 06:26 · Score: -1, Troll

i'm so glad that they are dying by the score. their little faggot religion is shit and is bringing down the human race. give up your faggot religion muslim bitches.

fuck mohammad, fuck allah, fuck islam.

raghead faggots.

Re:fuck muslims by timberwolf753 · 2008-09-20 07:18 · Score: 0

Troll much do we?

Content by whathappenedtomonday · 2008-09-20 06:27 · Score: 4, Interesting

I was wondering where I could find some vids to check out quality vs. file sizes and found this index of demo files. Looks great in VLC, quite impressive even at lower bitrates.

--
I hope I didn't brain my damage.

Underwhelming Results by Tab+is+on+Slashdot · 2008-09-20 06:27 · Score: 1

While it's very cool what the BBC is doing, and it's good to see wavelet technology being pushed, Dirac 1.0 falls extremely short in my tests (at least on animated material at medium bitrates). In the H.264 era, the quality is unacceptable. Here's hoping they'll be able to keep improving it. On the other hand, I know at least one x264 dev who's convinced that OBMC wavelets will never match the quality of MC block-based approaches without a major breakthrough.

Re:Performance? Benefits? by fuzzyfuzzyfungus · 2008-09-20 06:38 · Score: 5, Insightful

We don't need another codec, per se, we need a royalty free codec, that can be legally implemented in FOSS situations, and others without a lot of legal overhead. Assuming it isn't markedly worse than others in performance terms, Dirac qualifies. If by some miracle(class II or greater) mpeg4 were available under such terms, there wouldn't be any point to Dirac; but that isn't exactly likely.

News from OGG Theora, too! by c0l0 · 2008-09-20 06:49 · Score: 5, Informative

Dirac isn't the only royality-free, patent-unencumbered video codec there is - Xiph's OGG Theora has been around a while already, yet failed to impress quality-wise up until recently. There's some really cool development going on however, and you may see some of the results achieved over there: http://xiphmont.livejournal.com/35363.html

It's noteworthy that the changes made only affect the ENCODER, thus no changes to the DECODER (the part of a codec all applications used to play back files have included) are necessary. This bodes very well for HTML5, which will include some support for Theora on at least Mozilla (and iirc Opera) browsers.

--
:%s/Open Source/Free Software/g

YTARY!

Re:News from OGG Theora, too! by Tab+is+on+Slashdot · 2008-09-20 07:05 · Score: 5, Interesting

Theora gets a bad rap for being outdated technology, but it does have a few advantages over MPEG-4 ASP: the loop filter, adaptive block sizes, and multiple reference frames, putting it closer to H.264 than MPEG-4 ASP. With these features, it's really a pretty strong showing from Xiph, and things can only get better as the encoder nears 1.0.
Re:News from OGG Theora, too! by delt0r · 2008-09-20 07:07 · Score: 4, Insightful

This is a big point. The Encoder is far more important that the rest of the codec. Folks talk about xvid and divx as if they are codecs when really they are different encoders for mpeg4.

Both Theora and Dirac have plenty of space to move with regard to encoders.

However there is no easy way to measure "distortion" of the encoded image that matches the human visual system all that well. (unlike audio). But I expect most codecs to get better in the next few years because of encoders. (including h264).

Ironically h264 does so well because of the availability of a free, fast and good quality encoder done my the community. Not the license owners.

--
If information wants to be free, why does my internet connection cost so much?
Re:News from OGG Theora, too! by mdmkolbe · 2008-09-20 07:21 · Score: 2, Insightful

However there is no easy way to measure "distortion" of the encoded image that matches the human visual system all that well. (unlike audio).
How do you objectively measure psychoacoustic distortion? Do the same techniques not apply to vision simply due to unknown constants or is there some more fundamental reason?
Re:News from OGG Theora, too! by lxt · 2008-09-20 07:38 · Score: 1

However there is no easy way to measure "distortion" of the encoded image that matches the human visual system all that well. (unlike audio)..
I'm not sure I agree with that...and I think the fact that there are people who *can* tell the difference between a 256kbs MP3 and CD-audio and those who *can't* perhaps shows that there's no easy way to map quality of audio onto something that matches human perception. There are plenty of technical ways however, both for audio and visual. I'm not sure where you're getting this from.
Re:News from OGG Theora, too! by delt0r · 2008-09-20 07:44 · Score: 5, Informative

In sound the idea of masking works really well. That is if there is a loud sound at a particular frequency we tend to not be able to hear sounds that are a low in pitch and a bit quieter (IIRC). Its effectively masked. The other big advantage is also the linear nature of sound.

But the human visual system is a *lot* more complicated. IIRC about 1/3 or our brains are used for visual perception. Currently we use PSNR (Peek signal to noise ratio) as a measure. But this has been shown many times to be a very poor indication of what we perceive. One example is blocking. Blocking cause straight lines to form in the image and our brains lock on to them far more quickly that other artifacts.

Next is the colour and the 2d nature of a image. Then add that the eyes do a bunch of preprocessing on motion perception and its getting quite difficult. Finally we have the method of comparison. Which often involves comparing still images from the video stream. Yet if thats a high motion scene the codec might be better off encoding these frames with low quality because we can't perceive the quality loss combined with fast motion.

Lets also not forget how many people think youtube is good quality or at worse, good enough!

--
If information wants to be free, why does my internet connection cost so much?
Re:News from OGG Theora, too! by delt0r · 2008-09-20 07:55 · Score: 4, Interesting

All the R&D papers I have read and from folk in the field working on this. Its well recognized that psychoacoustic models are far more developed than psycho visual models.

I don't doubt that some people can tell the difference between flac and mp3/ogg/aac. But the true number is far less than the claimed number (do a proper blind test to really find out). Also you don't design codecs for 0.5% of the population that can hear the difference, but for the 90% that can't and the other 9.5% that don't care.

Now its a fact that PSNR is used in most encoders. Its also widely recognized that it is not a good measure. I have done my own image compression and got better PSNR than jpeg per bit, and yet it looked far worse.

So I'm not really sure where you getting the idea that is even in the same category as audio.

--
If information wants to be free, why does my internet connection cost so much?
Re:News from OGG Theora, too! by Frenchman113 · 2008-09-20 08:53 · Score: 1

Yes, let me know when "Thoera" has b-frames. You know, like those things in MPEG-1.
Re:News from OGG Theora, too! by benwaggoner · 2008-09-20 10:29 · Score: 1

Except that Theora not only start being behind the then-emerging H.264 and VC-1, but its implementations are launch were quite a bit weaker than even Xvid of the era, and have essentially stagnated.
And since then, it's fallen even further behind; the implementations of standardized codecs has been improving a lot more each year than Theora, as have proprietary codecs like the later entries in On2's VPx series (Theora was forked from VP3; On2 just announced VP8).
There's been some interesting work in the last year or so, but very much research, not development. The new screen shots showing improvement a few months ago didn't even include rate control, and weren't close to being usable in production.
In the same six months, x264 has made a variety of dramatic improvements that are already usable in the real world.
So, question: why so little progress on Theora for so long? For all the focus of the industry on "we must have a non-patent encumbered codec!" it seems like the open source efforts have all gone into xvid and x264. Anyone know what was blocking progress on Theora for so long?

--

My video compression blog
Re:News from OGG Theora, too! by Anonymous Coward · 2008-09-20 15:45 · Score: 0

In sound the idea of masking works really well. That is if there is a loud sound at a particular frequency we tend to not be able to hear sounds that are a low in pitch and a bit quieter (IIRC). Its effectively masked. The other big advantage is also the linear nature of sound.
But the human visual system is a *lot* more complicated. IIRC about 1/3 or our brains are used for visual perception. Currently we use PSNR (Peek signal to noise ratio) as a measure. But this has been shown many times to be a very poor indication of what we perceive. One example is blocking. Blocking cause straight lines to form in the image and our brains lock on to them far more quickly that other artifacts.
Next is the colour and the 2d nature of a image. Then add that the eyes do a bunch of preprocessing on motion perception and its getting quite difficult. Finally we have the method of comparison. Which often involves comparing still images from the video stream. Yet if thats a high motion scene the codec might be better off encoding these frames with low quality because we can't perceive the quality loss combined with fast motion.
Lets also not forget how many people think youtube is good quality or at worse, good enough!
So you want a psychovisual compression scheme, like what MP3 has done for audio, get rid of the crap we can't see?
Re:News from OGG Theora, too! by delt0r · 2008-09-21 06:18 · Score: 1

Dev time. There are more people working on xvid etc than theora. Folks seem to be more interested in helping out the patent encumbered formats by providing good encoders and decoders. While the bulk (if not all) of Theora work is done by only a few people. And thats a lot of work even if you get to do it full time.

A patent free codec is still good as long as its pretty close to say mpeg4. It would end up in a lot of games etc, and it keeps everyone else playing nice because there are alternatives. It not going to replace mpeg4/h.264 anymore than mp3 was replaced with vorbis, unless MPEG-LA gets stroppy with the OS encoders and decoders and the fees for produced content.

--
If information wants to be free, why does my internet connection cost so much?
Re:News from OGG Theora, too! by evilviper · 2008-09-21 07:56 · Score: 1

but it does have a few advantages over MPEG-4 ASP: the loop filter, adaptive block sizes, and multiple reference frames, putting it closer to H.264 than MPEG-4 ASP.
AFAIK, the deblocker in VP3/Theora is NOT an in-loop filter, but just a simple post-processor, with just a little more smarts due to being codec-specific.
VP3/Theora also has some significant inherent DISADVANTAGES over MPEG-4 ASP, such as lacking B-frames (which Xvid uses to great effect). And yet, while VP3/Theora isn't competitive with H.264 or MPEG-4 ASP, it is a monster on CPU usage (worse than H.264 in most cases), which makes playback on modest hardware impossible (where the likes of Divx have no problems).
But, IMHO, codec design issues are relatively minor where quality is concerned. H.264/AVC is the first significant change to video coding since MPEG-1. MPEG-2 and MPEG-4 ASP merely change coding around slightly, and offer nominal inherent improvements, yet most people don't believe that MPEG-1/MPEG-2 is competitive with Divx/Xvid/FMP4 (though they technically are). And all the neat tricks that the likes of H.264 do still can only help on low-bitrate material, as there are inherent limitations at high bitrates that even the ancient MPEG-2 can approach. Yet, VP3/Theora do quite poorly at high bitrates as well, encoding tricks or no,
Theora could have been a very good video codec, but some 6 years of development time have been completely squandered, with no design nor quality improvements to speak of. Honestly, developers would do just as well to work on tweaking the VP3.2 codec as Theora's code base, and get a somewhat larger installed base of decoders that will be able to play the video (Theora offers backwards compatibility).

With these features, it's really a pretty strong showing from Xiph,
All the features you have described are present in On2's VP3.2 code base, which they donated to the public. Xiph.org didn't have anything to do with developing them.

--
Slashdot gets worse every day... Pipedot: News for nerds, without the corporate slant
Re:News from OGG Theora, too! by evilviper · 2008-09-21 08:14 · Score: 1

Also you don't design codecs for 0.5% of the population that can hear the difference, but for the 90% that can't and the other 9.5% that don't care.
In fact you do design for the 0.5%. Testing of codecs specifically uses expert listeners, with an in-depth double-blind testing setup, hidden anchors, and the like. Of course you aren't going to please everyone, all the time, but codec developers certainly do try their best to do so.

--
Slashdot gets worse every day... Pipedot: News for nerds, without the corporate slant
Re:News from OGG Theora, too! by evilviper · 2008-09-21 08:23 · Score: 1

I think the fact that there are people who *can* tell the difference between a 256kbs MP3 and CD-audio and those who *can't* perhaps shows that there's no easy way to map quality of audio onto something that matches human perception.
In fact you're wrong.
MP3 was never designed to be indistinguishable from CD-audio. It was designed to sound GOOD at even lower bitrates (64kbps mono). MP3 has several limitations that prevent it from doing so. It's immensely ironic that people are now cranking up the bitrates on MP3s to try and get it to do so.
In fact it's possible to get far superior sound, that next to nobody can distinguish from the original, at lower bitrates, using better audio codecs, like MUSEPACK, or the even older MPEG-1 Layer II (yes, MP3's "little brother").
Now, the models of the human auditory system AREN'T theoretically perfect by any stretch of the imagination. They are defined through practical testing rather than some proven law of perception. However, it is exhaustively testing, and very well defined, thanks in large part to Bell Labs' work back in the 80s. You can find ou all the details you could possibly want about this by looking through the old 1980s research papers by J. D. Johnston about "Perceptual Entropy".

--
Slashdot gets worse every day... Pipedot: News for nerds, without the corporate slant
Re:News from OGG Theora, too! by Anonymous Coward · 2008-09-21 21:01 · Score: 0

AFAIK, the deblocker in VP3/Theora is NOT an in-loop filter, but just a simple post-processor, with just a little more smarts due to being codec-specific.
Sorry but it IS a in loop blocking filter. I have read the documentation more than once.
Have you ever heard of H.263? H.264 is not new in either in history or ideas. Its also not that different. The fact that patents are issued now (as in 2008) that cover some encoding stuff and even some parts of the codec is plain stupid.
The only reason everyone here uses H.264 and mpeg4 is the lacks enforcement of the licensing requirements. I think they are repeating the mp3 sham, wait until its a entrenched format then strike. Did you know you are suppose to pay fees (or will soon) just to produce *content* in these formats. ie put a movie up on your web page.
Oh and even if you pay your fees, there is no protection from parties that may have a patent on h264 from suing you.
This is why we need patent free codecs.
Re:News from OGG Theora, too! by evilviper · 2008-09-22 04:48 · Score: 1

Sorry but it IS a in loop blocking filter.
My mistake. It is rather simplistic, but it is in fact an in-loop deblocking filter.

I have read the documentation more than once.
Interesting you should say that, because I just looked through the specs and I see NO indication at all that Theora can use "adaptive block sizes" as claimed in the parent post. In fact it repeatedly says Theora uses fixed 8x8 blocks, just like MPEG-1/2/4ASP.

Have you ever heard of H.263? H.264 is not new in either in history or ideas.
*Ahem*. I did not claim it was significantly improved, or theoretically revolutionary.
The point I was making are that previous codecs have done just fine without dynamic block sizes, multiple reference frames, deblocking, etc. And that encoding improvements happen not from new features, but from newer/better encoders.

Did you know you are suppose to pay fees (or will soon) just to produce *content* in these formats. ie put a movie up on your web page.
Yes, I know the licensing terms in-detail, since I do quite a bit of work in the industry.

Oh and even if you pay your fees, there is no protection from parties that may have a patent on h264 from suing you.
There is never any guarantee that you won't be sued by anyone in the world, for some ridiculous reason. That's the nature of the legal system. Using Theora isn't going to prevent you from being sued, either, so claiming that as a benefit is a red herring.

This is why we need patent free codecs.
We have them. H.261 and MPEG-1 are patent-free. MPEG-1 is rather efficient at low-bitrate encoding (on par with MPEG-4 ASP) and likely the most widely supported video format in the world, by both hardware and software. I'd put it up against Theora any day, as well.
So why don't you use it? Why wait for Dirac or Theora?

--
Slashdot gets worse every day... Pipedot: News for nerds, without the corporate slant

Re:Performance? Benefits? by Tab+is+on+Slashdot · 2008-09-20 06:55 · Score: 5, Informative

How does it stack up to other codecs?

As I say below, unfortunately the quality is lacking compared to modern codecs like H.264 and even (dare I say) VC-1. Apparently that's just the nature of using wavelets. While they give a very natural style of compression on still images (JPEG-2000, etc), they do not translate well to moving sequences because, unlike all other current codecs, the image is not broken up into blocks that can then be tracked and diff'd in time. Still, it'll be interesting to follow Dirac, if only because they're taking a radical new approach with only Michael Niedermayer's Snow as a peer.

Re:Performance? Benefits? by delt0r · 2008-09-20 07:11 · Score: 4, Informative

As I state below. Most of codecs performance has to do with the encoder. At 1.0.0 its too early to tell if the format/codec design is limited.

However a great codec without a good encoder is no good at all. But its early days yet considering h.264 has been around for 5+ years.

--
If information wants to be free, why does my internet connection cost so much?

Re:Performance? Benefits? by Anonymous Coward · 2008-09-20 07:18 · Score: 0

Apparently that's just the nature of using wavelets.

Dirac doesn't use wavelets.

Re:Performance? Benefits? by Tab+is+on+Slashdot · 2008-09-20 07:28 · Score: 5, Informative

Dirac employs wavelet compression, instead of the discrete cosine transforms used in most older codecs (such as H.264/MPEG-4 AVC or SMPTE's VC-1). Dirac is one of several projects attempting to apply wavelets to video compression. Others include Rududu [2], Snow and Tarkin. Wavelet compression has already proven its viability in the JPEG 2000 compression standard for photographic images.

Yes it does :|

Where is the windows codec. by NuclearFirestorm · 2008-09-20 07:51 · Score: 0, Troll

The new VLC can decode dirac files, but there is no way to encode in windows. So this wonderful release doesn't do 99% of the world much good.

Re:Where is the windows codec. by Anonymous Coward · 2008-09-20 08:10 · Score: 1, Insightful

Windows is not 99% of the world. It is, depending on who you ask, between 88% and 95% of all desktop computer users. Of course, most of those don't give a damn about encoding Dirac files; the proportion of Windows users among highly IT literate, technically minded people is somewhat lower.
Re:Where is the windows codec. by turgid · 2008-09-20 08:28 · Score: 1

Hmmm... "Killer application" drives adoption of platform.
Most people on Windows only consume, not produce.
Flower chase the sunshine.

--
Stick Men
Re:Where is the windows codec. by figleaf · 2008-09-20 08:29 · Score: 2, Interesting

I downloaded the code from sourceforge and compiled the code using Visual Studio 2008.
Looks like the encoder is distributed in source format only. I could not locate any pre-built binaries.
I am having trouble figuring out what the command-line parameters mean from the README supplied in the source tarball.
This certainly needs better documentation for non technical users.
The samples certainly look impressive. I will try to compare it against my current favorite encoder -- x264 -- over the weekend.
Re:Where is the windows codec. by calmofthestorm · 2008-09-20 08:37 · Score: 1

It's development software; new innovative software tends to be one platform, generally whatever the devs are most comfortable with. I don't know a thing about windows software engineering since early xp, but if I make something cool enough for linux someone else will probably be interested in porting it.
Video processing is commonly done on non-windows platforms because then you don't need a per-computer license for each node in the render farm, and FOSS software can follow an edl as easily as the proprietary stuff that created it. Maybe that's what bbc is using?
Just a guess

--
93rd rule of Slashdot: No matter how obvious my sarcasm is, my comment will be taken seriously by someone.
Re:Where is the windows codec. by figleaf · 2008-09-20 11:09 · Score: 2, Interesting

Took me a while to figure I needed to use YUV input.
Unfortunately, it looks like Dirac is no match for x264. :(
Even VC-1 beats Dirac.

Re:Performance? Benefits? by Ant+P. · 2008-09-20 08:05 · Score: 2, Funny

The codec is new, give it a few months.

Early DVDs looked like shitty 90% compressed jpegs too, you know.

Re:Performance? Benefits? by Tab+is+on+Slashdot · 2008-09-20 08:30 · Score: 2, Interesting

Update: I've been told by the devs that Dirac is optimized for HD live action, wheres my tests have thus far involved SD animated content, so, YMMV. I'll have to try some live action sources next.

Re:Performance? Benefits? by calmofthestorm · 2008-09-20 08:32 · Score: 1

Exactly, Matroska is great and all from a freedom standpoint, but technically it's far behind the encumbered ones.

At least we have ogg for audio, it seems like nothing can beat it in terms of quality/bitrate:-)

--
93rd rule of Slashdot: No matter how obvious my sarcasm is, my comment will be taken seriously by someone.

Re:Performance? Benefits? by i_liek_turtles · 2008-09-20 08:41 · Score: 1

Matroska is a container, not a codec.

Re:Performance? Benefits? by Tweenk · 2008-09-20 08:42 · Score: 2, Interesting

Matroska is not a codec. It is a container format, and it beats any closed-source competitions hands own on features (e.g. as far as I know it is the only format that supports embedding custom TrueType fonts for subtitles).

The best video encoding combo right now is:
- Matroska as the container
- H.264 for video
- Ogg Vorbis for audio
- ASS for subtitles

--
Those who would give up liberty to obtain working drivers, deserve neither liberty nor working drivers.

Re:Performance? Benefits? by calmofthestorm · 2008-09-20 08:52 · Score: 1

Yeah I realized that after I wrote this:-) I meant Theora. Oopsie.

--
93rd rule of Slashdot: No matter how obvious my sarcasm is, my comment will be taken seriously by someone.

Re:Performance? Benefits? by Anonymous Coward · 2008-09-20 10:17 · Score: 1, Interesting

Also, try to encode with the newer versions of Theora. It has gotten much improvements in the last year - quality problems were never in the decoding as some will have you believe, but that the encoder pretty much sucked.

Not sure what you expect and I'm no video buff... but it sure looks a LOT better.

(You may still be right, of course. I've just found that 99% of all who state anything about anythings quality usually have formed their opinion once, maybe years ago, and then keep on repeating it).

Free as in beer or speech? by benwaggoner · 2008-09-20 10:23 · Score: 1

This came up in yesterday's discussion of the Canonical codec pack.

Standardized codecs, like VC-1 and H.264, have full open specifications and typically even reference source code implementations that can be reused in a variety of ways.

However, they also require patent fees depending on use and jurisdiction.

The issue of free software has always been asserted to be about "speech, not beer" but it seems like there's an assumption that it has to be free as in "speech AND beer." I'm sure all kind of arguments can be made that it should be that way, but everything I've read recently sort of begs that question by conflating the issues of closed source and patent licensing. When to me they look like pretty orthogonal issues; all of the reasons why people say they like open source are still delivered if the source is open, irrespective of whether a patent fee is paid. And MEPG-LA makes patent administration and payment pretty straightforward under RAND terms.

Did I just miss some prior discussions?

--

My video compression blog

Re:Free as in beer or speech? by delt0r · 2008-09-20 11:06 · Score: 1

You are right and wrong. If I want to keep MPEG-LA happy, i need to pay quite a bit of money if there are more than 100K downloads IIRC. For a open source project that is *not* selling the code it is a lot of money. Most OS projects are not done with money, but with time.

How many copies of mplayer do you think are out there? Its a lot more than 100K copies.

But here is the real rub. Even if you pay the fees, they give *no* guarantee what so ever that you are not infringing some other patent.

Also these non discriminatory terms are crap. If the fee is high enough, it discriminates plenty. Add to that the bulk of these patents are utter rubbish. I would say there are not with the spirit of anything free, freedom, beer, or otherwise.

--
If information wants to be free, why does my internet connection cost so much?
Re:Free as in beer or speech? by benwaggoner · 2008-09-20 11:56 · Score: 1

You are right and wrong. If I want to keep MPEG-LA happy, i need to pay quite a bit of money if there are more than 100K downloads IIRC. For a open source project that is *not* selling the code it is a lot of money. Most OS projects are not done with money, but with time.
But if you're just distributing the source, you wouldn't need to pay the patent. If it's a real product, or if users are compiling it into one, is doesn't seem to be an infringement on "freedom" in the classic RMS definition to wind up paying a fee. You still have full control over the code and technology used.

But here is the real rub. Even if you pay the fees, they give *no* guarantee what so ever that you are not infringing some other patent.
Theoretically true, although that hasn't happened much in practice, at least in this space. And Theora and Dirac are in the same legal position, and don't have the market effet of lots of companies looking for patents to assert to get a share of the MPEG-LA revenue.

Also these non discriminatory terms are crap. If the fee is high enough, it discriminates plenty. Add to that the bulk of these patents are utter rubbish. I would say there are not with the spirit of anything free, freedom, beer, or otherwise.
Have you actually all of the H.264 or VC-1 patents? If so, you've got more patience than I do!
Also, it's not ND, but RAND - REASONABLE and non-discrimintory. If anything H.264 and VC-1 are much more reasonable than MPEG-2, since they're a lot cheaper per unit and have an annual cap for high-volume products.
Anyway, feel free to hate software patents, but it seems to me like that's a different issue than open source.

--

My video compression blog
Re:Free as in beer or speech? by delt0r · 2008-09-20 20:48 · Score: 2, Interesting

But if you're just distributing the source..
So now I can't also distribute binary but my freedom is not affected? I don't think I would want to test source=ok, binarys=!ok as far as patent law is concerned with my wallet. Economic harm is all thats needed if software patents are valid.

Theoretically true, although that hasn't happened much in practice, at least in this space.
So you pay a crap load of money and only don't get sued much? Thats a raw deal. There has been at least one case I know of with mpeg4 | h.264, and thats a lot more than what both theora and dirac have had to deal with. Add the fact that theora is based on VP3 with a active company now with VP7. That they are active with this "IP".

and don't have the market effect of lots of companies looking for patents to assert to get a share of the MPEG-LA revenue.
And yet these codecs are the only ones that have any history of problems.

Have you actually all of the H.264 or VC-1 patents?
Not all, but most of them. I even got "advice" and we did decide that most could be overturned with prior art, patenting math, and obviousness. But the cost and most importantly *time* that this would take... Its not that there is one, its that there 10+ or more for each company in the pool.

If anything H.264 and VC-1 are much more reasonable than MPEG-2...
This is true. Its a lot better. And yet still discriminates against OS ideals of freedom and free. Yes you lose freedom, no matter how slice it.

There are patents that are free to license for free GPL type products. These do not reduce freedom. The patents we are talking about do. Software patents in general do. Our freedom is reduced if they apply to code we write.

Another problem is what you are expected to sign up to when you get a license. This license itself is also restrictive. So if you donate that +3 million per annum cap to the mplayer/ffmpeg whatever group so they can release legal worldwide free codecs, I think you will find that the license will prevent this. After all what would all the other licensees think.... If the license dose not prevent this it would not take long to do so, as mpeg-la reduce themselfs to a single licensee (everyone else can just use mplayer/ffmpeg code even in hardware players).

--
If information wants to be free, why does my internet connection cost so much?

Re:Performance? Benefits? by Dwedit · 2008-09-20 10:27 · Score: 2, Interesting

How does Vorbis really compare against AAC? Besides the whole royalty/patent free issue, does Vorbis really beat out AAC? (Ignoring royalty/patent issues here because you also mentioned H264)

hooray for dirac by Dimble+ThriceFoon · 2008-09-20 10:29 · Score: 1

been waiting a long time.

top gear = rockin' :) by timothy · 2008-09-20 10:30 · Score: 1

I've seen only one episode (vsiting a friend who watches TV more than I do, and who gets more channels), but it made me want to see many more. (The one I saw was about zooming and crunching through East Africa in beaters purchased in-country -- pretty impressive how they all held up, actually, though all suffered pretty badly.)

Thanks for your support, TV licensees of Britain!

timothy

--
jrnl: http://tinyurl.com/c2l8yr / foes: http://tinyurl.com/ckjno5

Re:top gear = rockin' :) by caluml · 2008-09-20 11:32 · Score: 2, Funny

Thanks for your support, TV licensees of Britain!
Uh, that's OK. Just send me some money, and I'll make sure it gets passed on to the right pub... I mean people.

--
Get your own free personal location tracker

Oh snap! by Anonymous Coward · 2008-09-20 10:33 · Score: 0

Geeks are getting uppity now!

wavelet codec by pikine · 2008-09-20 10:55 · Score: 1

What about Dirac being a wavelet based codec that has inter-frame motion compensation? Wavelet is superior to DCT-based codec, like mpeg-1, 2, and h.264. Dirac's inter-frame encoding is also something that motion JPEG 2000 doesn't have.

--
I once had a signature.

Re:Performance? Benefits? by non-poster · 2008-09-20 11:20 · Score: 1

How will you use it?

If it's really important to you, don't guess; do your own test. The Hydrogen Audio folks know something about this.

http://wiki.hydrogenaudio.org/index.php?title=Choosing_the_best_codec

Re:Performance? Benefits? by Anonymous Coward · 2008-09-20 14:18 · Score: 0

OGG is far behind AAC in terms of quality/bitrate. And from what I hear Dirac is only comparable to good old MPEG-4 ASP (divx/xvid), not to AVC (H.264).

Dirac? Schrodinger? wha...? by aqk · 2008-09-20 19:47 · Score: 2, Funny

Hey, I KNEW those Atomic Physics courses I took way back in University would come in handy!

Who needs to compress time, when all you hafta do is compress the video.
If I took a video of my cat, and then compressed it with this new codec, would the cat be...
Umm.. never mind...!
.
.

--

.
- aqk
F U /. illiterates.

Re:Performance? Benefits? by Anonymous Coward · 2008-09-20 20:19 · Score: 0

Vorbis is far behind AAC in terms of quality/bitrate.

That's just not true at all.

Re:Performance? Benefits? by seeker_1us · 2008-09-20 21:14 · Score: 1

Right now Ogg/Vorbis is the overall BEST quality lossy compression codec. Check out the wikipedia page.

Yay! More codecs to download! by rangergordon · 2008-09-20 22:13 · Score: 1

Now I can feel just as good about using open-source video compression as I do about buying eggs from free-range chickens!

The porn-viewing experience just gets better and better.

Re:Performance? Benefits? by pezezin · 2008-09-20 23:30 · Score: 1

While they give a very natural style of compression on still images (JPEG-2000, etc), they do not translate well to moving sequences because, unlike all other current codecs, the image is not broken up into blocks that can then be tracked and diff'd in time.

You seem to be confusing motion estimation/compensation with residual coding. Dirac does break the image into blocks, using overlapped block motion compensation. However, the residual image is coded as a whole, thanks to wavelets. This should greatly reduce blocking artifacts.

Vorbis is *audio*, THEORA is video... by Anonymous Coward · 2008-09-21 10:23 · Score: 0

dude...

http://en.wikipedia.org/wiki/Theora

"Previous evaluations have found VP3[3] and Theora[4] [5] substantially lacking compared to contemporary video codecs.

Efforts to improve performance:

Sources close to Xiph.org have stated that the performance characteristics of the current Theora reference implementation are mostly dominated by implementation issues inherited from the original VP3 code base. An internal document exists, which lists the known implementation problems and gives an example of how improving one aspect of the encoder can lead to visibly improved quality.[6] Current work on Theora is focused on an experimental version, which targets correcting aspects of the encoder which were identified in that paper as being suboptimal. This experimental version is supposed to replace the current encoder in a future Theora release."

Re:Performance? Benefits? by Tweenk · 2008-09-22 13:05 · Score: 1

According to Wikipedia aoTuV beta 4 Vorbis encoder beats WMA, MP3 and AAC at 64-128 kbit/s.

--
Those who would give up liberty to obtain working drivers, deserve neither liberty nor working drivers.

127 comments