g r o t t o 1 1

Peeve Farm
Breeding peeves for show, not just to keep as pets
Brian Tiemann
Silicon Valley-based purveyor of a confusing mixture of Apple punditry and political bile.

btman at grotto11 dot com

Read These Too:

InstaPundit
Steven Den Beste
James Lileks
Little Green Footballs
As the Apple Turns
Entropicana
Cold Fury
Capitalist Lion
Red Letter Day
Eric S. Raymond
Tal G in Jerusalem
Aziz Poonawalla
Corsair the Rational Pirate
.clue
Ravishing Light
Rosenblog
Cartago Delenda Est




Book Plugs:



Buy 'em and I get
money. I think.
BSD Mall


Amazon Honor System Click Here to Pay Learn More



  4/9/2007 -  4/12/2007
  4/2/2007 -   4/8/2007
 3/26/2007 -   4/1/2007
 3/19/2007 -  3/25/2007
 3/12/2007 -  3/18/2007
  3/5/2007 -  3/11/2007
 2/26/2007 -   3/4/2007
 2/19/2007 -  2/25/2007
 2/12/2007 -  2/18/2007
  2/5/2007 -  2/11/2007
 1/29/2007 -   2/4/2007
 1/22/2007 -  1/28/2007
 1/15/2007 -  1/21/2007
  1/8/2007 -  1/14/2007
  1/1/2007 -   1/7/2007
12/25/2006 - 12/31/2006
12/18/2006 - 12/24/2006
12/11/2006 - 12/17/2006
 12/4/2006 - 12/10/2006
11/27/2006 -  12/3/2006
11/20/2006 - 11/26/2006
11/13/2006 - 11/19/2006
 11/6/2006 - 11/12/2006
10/30/2006 -  11/5/2006
10/23/2006 - 10/29/2006
10/16/2006 - 10/22/2006
 10/9/2006 - 10/15/2006
 10/2/2006 -  10/8/2006
 9/25/2006 -  10/1/2006
 9/18/2006 -  9/24/2006
 9/11/2006 -  9/17/2006
  9/4/2006 -  9/10/2006
 8/28/2006 -   9/3/2006
 8/21/2006 -  8/27/2006
 8/14/2006 -  8/20/2006
  8/7/2006 -  8/13/2006
 7/31/2006 -   8/6/2006
 7/24/2006 -  7/30/2006
 7/17/2006 -  7/23/2006
 7/10/2006 -  7/16/2006
  7/3/2006 -   7/9/2006
 6/26/2006 -   7/2/2006
 6/19/2006 -  6/25/2006
 6/12/2006 -  6/18/2006
  6/5/2006 -  6/11/2006
 5/29/2006 -   6/4/2006
 5/22/2006 -  5/28/2006
 5/15/2006 -  5/21/2006
  5/8/2006 -  5/14/2006
  5/1/2006 -   5/7/2006
 4/24/2006 -  4/30/2006
 4/17/2006 -  4/23/2006
 4/10/2006 -  4/16/2006
  4/3/2006 -   4/9/2006
 3/27/2006 -   4/2/2006
 3/20/2006 -  3/26/2006
 3/13/2006 -  3/19/2006
  3/6/2006 -  3/12/2006
 2/27/2006 -   3/5/2006
 2/20/2006 -  2/26/2006
 2/13/2006 -  2/19/2006
  2/6/2006 -  2/12/2006
 1/30/2006 -   2/5/2006
 1/23/2006 -  1/29/2006
 1/16/2006 -  1/22/2006
  1/9/2006 -  1/15/2006
  1/2/2006 -   1/8/2006
12/26/2005 -   1/1/2005
12/19/2005 - 12/25/2005
12/12/2005 - 12/18/2005
 12/5/2005 - 12/11/2005
11/28/2005 -  12/4/2005
11/21/2005 - 11/27/2005
11/14/2005 - 11/20/2005
 11/7/2005 - 11/13/2005
10/31/2005 -  11/6/2005
10/24/2005 - 10/30/2005
10/17/2005 - 10/23/2005
10/10/2005 - 10/16/2005
 10/3/2005 -  10/9/2005
 9/26/2005 -  10/2/2005
 9/19/2005 -  9/25/2005
 9/12/2005 -  9/18/2005
  9/5/2005 -  9/11/2005
 8/29/2005 -   9/4/2005
 8/22/2005 -  8/28/2005
 8/15/2005 -  8/21/2005
  8/8/2005 -  8/14/2005
  8/1/2005 -   8/7/2005
 7/25/2005 -  7/31/2005
 7/18/2005 -  7/24/2005
 7/11/2005 -  7/17/2005
  7/4/2005 -  7/10/2005
 6/27/2005 -   7/3/2005
 6/20/2005 -  6/26/2005
 6/13/2005 -  6/19/2005
  6/6/2005 -  6/12/2005
 5/30/2005 -   6/5/2005
 5/23/2005 -  5/29/2005
 5/16/2005 -  5/22/2005
  5/9/2005 -  5/15/2005
  5/2/2005 -   5/8/2005
 4/25/2005 -   5/1/2005
 4/18/2005 -  4/24/2005
 4/11/2005 -  4/17/2005
  4/4/2005 -  4/10/2005
 3/28/2005 -   4/3/2005
 3/21/2005 -  3/27/2005
 3/14/2005 -  3/20/2005
  3/7/2005 -  3/13/2005
 2/28/2005 -   3/6/2005
 2/21/2005 -  2/27/2005
 2/14/2005 -  2/20/2005
  2/7/2005 -  2/13/2005
 1/31/2005 -   2/6/2005
 1/24/2005 -  1/30/2005
 1/17/2005 -  1/23/2005
 1/10/2005 -  1/16/2005
  1/3/2005 -   1/9/2005
12/27/2004 -   1/2/2004
12/20/2004 - 12/26/2004
12/13/2004 - 12/19/2004
 12/6/2004 - 12/12/2004
11/29/2004 -  12/5/2004
11/22/2004 - 11/28/2004
11/15/2004 - 11/21/2004
 11/8/2004 - 11/14/2004
 11/1/2004 -  11/7/2004
10/25/2004 - 10/31/2004
10/18/2004 - 10/24/2004
10/11/2004 - 10/17/2004
 10/4/2004 - 10/10/2004
 9/27/2004 -  10/3/2004
 9/20/2004 -  9/26/2004
 9/13/2004 -  9/19/2004
  9/6/2004 -  9/12/2004
 8/30/2004 -   9/5/2004
 8/23/2004 -  8/29/2004
 8/16/2004 -  8/22/2004
  8/9/2004 -  8/15/2004
  8/2/2004 -   8/8/2004
 7/26/2004 -   8/1/2004
 7/19/2004 -  7/25/2004
 7/12/2004 -  7/18/2004
  7/5/2004 -  7/11/2004
 6/28/2004 -   7/4/2004
 6/21/2004 -  6/27/2004
 6/14/2004 -  6/20/2004
  6/7/2004 -  6/13/2004
 5/31/2004 -   6/6/2004
 5/24/2004 -  5/30/2004
 5/17/2004 -  5/23/2004
 5/10/2004 -  5/16/2004
  5/3/2004 -   5/9/2004
 4/26/2004 -   5/2/2004
 4/19/2004 -  4/25/2004
 4/12/2004 -  4/18/2004
  4/5/2004 -  4/11/2004
 3/29/2004 -   4/4/2004
 3/22/2004 -  3/28/2004
 3/15/2004 -  3/21/2004
  3/8/2004 -  3/14/2004
  3/1/2004 -   3/7/2004
 2/23/2004 -  2/29/2004
 2/16/2004 -  2/22/2004
  2/9/2004 -  2/15/2004
  2/2/2004 -   2/8/2004
 1/26/2004 -   2/1/2004
 1/19/2004 -  1/25/2004
 1/12/2004 -  1/18/2004
  1/5/2004 -  1/11/2004
12/29/2003 -   1/4/2004
12/22/2003 - 12/28/2003
12/15/2003 - 12/21/2003
 12/8/2003 - 12/14/2003
 12/1/2003 -  12/7/2003
11/24/2003 - 11/30/2003
11/17/2003 - 11/23/2003
11/10/2003 - 11/16/2003
 11/3/2003 -  11/9/2003
10/27/2003 -  11/2/2003
10/20/2003 - 10/26/2003
10/13/2003 - 10/19/2003
 10/6/2003 - 10/12/2003
 9/29/2003 -  10/5/2003
 9/22/2003 -  9/28/2003
 9/15/2003 -  9/21/2003
  9/8/2003 -  9/14/2003
  9/1/2003 -   9/7/2003
 8/25/2003 -  8/31/2003
 8/18/2003 -  8/24/2003
 8/11/2003 -  8/17/2003
  8/4/2003 -  8/10/2003
 7/28/2003 -   8/3/2003
 7/21/2003 -  7/27/2003
 7/14/2003 -  7/20/2003
  7/7/2003 -  7/13/2003
 6/30/2003 -   7/6/2003
 6/23/2003 -  6/29/2003
 6/16/2003 -  6/22/2003
  6/9/2003 -  6/15/2003
  6/2/2003 -   6/8/2003
 5/26/2003 -   6/1/2003
 5/19/2003 -  5/25/2003
 5/12/2003 -  5/18/2003
  5/5/2003 -  5/11/2003
 4/28/2003 -   5/4/2003
 4/21/2003 -  4/27/2003
 4/14/2003 -  4/20/2003
  4/7/2003 -  4/13/2003
 3/31/2003 -   4/6/2003
 3/24/2003 -  3/30/2003
 3/17/2003 -  3/23/2003
 3/10/2003 -  3/16/2003
  3/3/2003 -   3/9/2003
 2/24/2003 -   3/2/2003
 2/17/2003 -  2/23/2003
 2/10/2003 -  2/16/2003
  2/3/2003 -   2/9/2003
 1/27/2003 -   2/2/2003
 1/20/2003 -  1/26/2003
 1/13/2003 -  1/19/2003
  1/6/2003 -  1/12/2003
12/30/2002 -   1/5/2003
12/23/2002 - 12/29/2002
12/16/2002 - 12/22/2002
 12/9/2002 - 12/15/2002
 12/2/2002 -  12/8/2002
11/25/2002 -  12/1/2002
11/18/2002 - 11/24/2002
11/11/2002 - 11/17/2002
 11/4/2002 - 11/10/2002
10/28/2002 -  11/3/2002
10/21/2002 - 10/27/2002
10/14/2002 - 10/20/2002
 10/7/2002 - 10/13/2002
 9/30/2002 -  10/6/2002
 9/23/2002 -  9/29/2002
 9/16/2002 -  9/22/2002
  9/9/2002 -  9/15/2002
  9/2/2002 -   9/8/2002
 8/26/2002 -   9/1/2002
 8/19/2002 -  8/25/2002
 8/12/2002 -  8/18/2002
  8/5/2002 -  8/11/2002
 7/29/2002 -   8/4/2002
 7/22/2002 -  7/28/2002
 7/15/2002 -  7/21/2002
  7/8/2002 -  7/14/2002
  7/1/2002 -   7/7/2002
 6/24/2002 -  6/30/2002
 6/17/2002 -  6/23/2002
 6/10/2002 -  6/16/2002
  6/3/2002 -   6/9/2002
 5/27/2002 -   6/2/2002
 5/20/2002 -  5/26/2002
 5/13/2002 -  5/19/2002
  5/6/2002 -  5/12/2002
 4/29/2002 -   5/5/2002
 4/22/2002 -  4/28/2002
 4/15/2002 -  4/21/2002
  4/8/2002 -  4/14/2002
  4/1/2002 -   4/7/2002
 3/25/2002 -  3/31/2002
 3/18/2002 -  3/24/2002
 3/11/2002 -  3/17/2002
  3/4/2002 -  3/10/2002
 2/25/2002 -   3/3/2002
 2/18/2002 -  2/24/2002
 2/11/2002 -  2/17/2002
  2/4/2002 -  2/10/2002
 1/28/2002 -   2/3/2002
 1/21/2002 -  1/27/2002
 1/14/2002 -  1/20/2002
  1/7/2002 -  1/13/2002
12/31/2001 -   1/6/2002
12/24/2001 - 12/30/2001
12/17/2001 - 12/23/2001
Tuesday, April 10, 2007
11:14 - Lock washer
http://daringfireball.net/2007/04/some_facts_about_aac

(top)
John Gruber has a nice piece a-sploding a few myths about the AAC format that seem to be gaining currency.

AAC is not “unique” to Apple. It’s not even controlled or invented by Apple, or any other single company. It is an ISO standard that was invented by engineers at Dolby, working with companies like Fraunhofer, Sony, AT&T, and Nokia. Licensing is controlled by Via. For up to 400,000 units per year, AAC playback costs $1.00 per unit; for more than 400,000 units per year, the price drops to $0.74 per unit.

In terms of licensing costs, patents, and openness, AAC is very much comparable to MP3. MP3 does have the advantage of near-ubiquitous support in consumer electronics and software; AAC has the advantage of slightly better audio quality at the same encoding bitrate. Additionally, MP3 requires a royalty fee of 2 percent for “electronic music distribution”, AAC requires no royalty fee for distribution.

Gruber's purpose in writing the article is to put paid to FUD about AAC in general; he doesn't, however, really address the question of DRM'd AAC (using FairPlay), which is an Apple-specific implementation. I feel I ought to point out, in case it looks like an omission, that the reason why he isn't addressing this is clear if you've read his previous missives on the subject: the people who bark about FairPlay being a "lock-in" feature for Apple and the iPod always seem to miss the fact that you're perfectly free to rip your CDs to MP3 or unprotected AAC and put those on your iPod; nothing in the iTunes/iPod universe requires you to ever, ever touch FairPlay if you don't want to.

He makes this point:

The ideal scenario would be for a genuinely open and free file format such as Ogg Vorbis to supplant MP3 as the de facto world standard. No patents, no licensing fees, a documented file format, open source libraries for encoding and decoding. That doesn’t seem to be in the cards, however. In the real world, major corporations only seem comfortable with multimedia formats backed by other large corporations.

Which does make sense, notwithstanding our usual snarkiness about Ogg Vorbis; the idea of it is sound, it just isn't going to catch on if it hasn't already. And Apple isn't going to bother with it if nobody uses it. Still, as recent DRM-free developments have highlighted, and knowing that even Apple has to pay license fees to use AAC, you can bet that they'd standardize on Ogg Vorbis—or an equivalent—in a heartbeat if it were to become popular outside the hardcore geek community.

But until that happens, they're likely to want to stick with a format that's backed by an independent licensing body with whom Apple can negotiate and remediate. That's just how business is done—it's the same reason why enterprises still roll out Windows mail and web servers even though they could do the same stuff for "free" with Linux. It's not because they're stupid; it's because they know why I put quotes around "free" in the previous sentence.

Another case in point, which came up in e-mail following from comments on a previous post, is Apple's chosen video format for importing into iTunes. Why MPEG-4 instead of DivX? They don't appear to be much different in playback quality to most users' eyes, and DivX seems a lot more widespread. iTunes won't import videos in formats that aren't natively supported by QuickTime, even if you have a plugin installed; but leaving aside that limitation, why not standardize on DivX? Why create a new format?

Well...

http://en.wikipedia.org/wiki/Divx
http://en.wikipedia.org/wiki/Mpeg-4

Just reading those two links ought to tell you all you need to know about that particular decision. Worth pointing out in the process is that a) Apple did not invent MPEG-4; and b) MPEG-4 predates and is a basis for DivX. But the real lesson is that whereas MPEG-4 is a series of structured standards developed by independent bodies and corporations such as IBM to whom patent and license fees must be paid, DivX is more like a backyard rocket made of chicken wire and aluminum siding and stolen North Korean engines. Its history is full of reverse-engineering, spyware scandals, rogue competitors, and a general intangible feeling that using this format makes one a DVD pirate. True, you'll find DivX certification logos on all kinds of DVD players these days; but looking at the prospects of the two formats and who's backing each, it seems clear that real MPEG-4 is the way the wind is blowing, and Apple would do better to support it—even if it's no more "theirs" than DivX is, and even if Apple is falling afoul of some of the MPEG-4 authorities for not paying enough license fees.

(Another thing is that DivX apparently doesn't make provision for DRM.)

Apple isn't making up the rules as it goes along, is the bottom line. As Gruber notes:

Let’s imagine for one paragraph that Microsoft’s and Apple’s digital music positions were flipped: that it was Microsoft that shipped the world-changing Zune in 2001, that they had sold 100 million Zunes to date, and that Microsoft’s online music store had 85 percent market share for legal downloads — all of them protected by Microsoft’s proprietary DRM. Can you imagine, in this scenario, Steve Ballmer or Bill Gates publishing an open letter like Jobs’s “Thoughts on Music”? Can you imagine Microsoft volunteering to switch from DRM-protected songs to an unprotected industry standard file format?

Me neither.

. . .

Apple is not the “Microsoft of digital music”, and everyone ought to stop trying to view their actions as though they were. Alas, that’s too much to hope for, and so in the meantime, now that Apple has proven its commitment to DRM-free music downloads, keep your eye out for anti-AAC propaganda from those pushing an anti-iTunes or anti-Apple agenda.

Which isn't to say they haven't done things that can be criticized. But there's no need to accuse them of doing what they've been so careful to avoid being accused of.

UPDATE: Geez, can we maybe not go overboard? Criminy.

UPDATE: A followup:

A bunch of readers emailed with a suggestion I hadn’t considered before: that the confusion over whether AAC is an “Apple format” is in some measure a byproduct of the format’s acronym, and that many people assume that one of the A’s in “AAC” stands for “Apple”. (It stands for “Advanced Audio Coding”.) If it were called, say, “MP4” instead, it might be more clear that it’s the successor to MP3.

M'yep.

UPDATE: Joe Bezdek, one of the co-founders of DivX, Inc., speaks out in the comments.

Back to Top

24 comments

1. Stephen Rider - 13:51 Tue 4/10/2007 ( web )

"Can you imagine Microsoft volunteering to switch from DRM-protected songs to an unprotected industry standard file format?"

That whole paragraph is worth the price of admission. The correct answer is "The sun will go nova long before Microsoft voluntarily gives up control of any format, 'standard' or platform."

2. bob - 23:36 Tue 4/10/2007 ( email )

Squeezebox also supports AAC http://www.slimdevices.com/

3. Andrew Cory - 12:39 Wed 4/11/2007 ( email | web )

I know next to nothing about Ogg Vorbis, but I don't see what's stopping Apple from implementing it on iTunes, the iPod, and making it the default "rip" option. And if they did that, how long do you think until other companies started doing the same?

4. Brian Tiemann - 12:49 Wed 4/11/2007 ( email | web )

One potential reason is that Ogg Vorbis just might not be as trouble-free from a legal standpoint as it seems. See Gruber's second post on this subject (the second UPDATE I added) for details.

5. smackfu - 15:04 Wed 4/11/2007 ( email )

I don't really care how much of a hack Divx is. It's what the good stuff is released in so, so its what I need. Not supporting it may be easier for Apple but it loses a sale.

6. Brian Tiemann - 15:12 Wed 4/11/2007 ( email | web )

Uh... "released" in?

7. tf - 16:01 Wed 4/11/2007 ( email )

smackfu raises the significant issue. It is well known that DivX "IS" a massive violation of mpeg patents. The reason its existence is allowed is because mpegla wants to avoid messy legal battles, and DivX gets nearly zero revenue off it anyway. The same is true of Xvid and Ogg formats more so: no one is going to ever use them for commercial gain so there is no legal battle. If anyone tries to do so, they are going to be "approached" quite quickly.

More important is the fact that Windows formats are also rip-offs of mpeg standards (VC-1 only has 2 patents that belong to Microsoft, the rest belong to other members of mpeg... Microsoft actually developed more of H.264 than they did WMV9) yet they continue to distribute their formats at cheaper licensing fees than mpeg, they limit their liability to their own patents and license fees paid to them, and had been (but I believe no longer) requiring the granting of patent licenses held by any licensees. With the finalization of VC-1 as a standard, and mpegla agreeing to license terms and license holders in recent months, this should finally become an interesting topic soon. Now, THAT is a legal mess!

8. tf - 17:40 Wed 4/11/2007 ( email )

Sorry, not smackfu (obviously). I meant Brian's comment.

9. Paul D - 21:44 Wed 4/11/2007 ( email )

OGG is a decent format, and I suspect the malarkey about unforeseen patent trouble is just that, malarkey.

I do seem to recall some discussions of incompatibility. OGG decoding supposedly requires floating-point arithmetic, and iPod processors to date have no floating-point unit.

10. Joe Bezdek - 10:35 Thu 4/12/2007 ( email | web )

Brian, I'm one of the co-founders of DivX, Inc. and I'd be happy to answer any questions you have. Unfortunately, the Wikipedia article about DivX you reference is inaccurate in several key respects. This is something we are aware of, but for several reasons have to date been unsuccessful in cleaning up. (Your post is the latest evidence that we need to try harder. :-))

When considering the specific issue of DivX vs. MPEG-4, I think the way to think of it is this: MPEG-4, despite its status as a "standard," is only a de jure standard. There are multiple ways to create an implementation of the standard, and there is nothing that guarantees all of those implementations will be interoperable. What DivX (Inc.) has done is create a de facto standard based on MPEG-4. We've not only created a popular MPEG-4 implementation and distributed it widely, we've also built the organization necessary to test and certify third-party products that use that implementation. Notably, there exists no "MPEG-4 certification" organization that can ensure your MPEG-4 product will play all (or even a majority of) MPEG-4 content. (For years, DivX was the _only_ de facto MPEG-4 standard. Now, with the iPod with video, there's arguably another.)

In the comments, tf writes "It is well known that DivX 'IS' a massive violation of mpeg patents." If this is "well known," then it's another example of inaccurate conventional wisdom about DivX. In fact, we (DivX, Inc.) are a licensee of MPEGLA, have been for many years, and pay all MPEG-4 licensing fees in full. (This, by the way, is more than can be said for other MPEG-4 implementations such as Xvid.)

As of September 2006, DivX, Inc. is a public company traded on the NASDAQ (NASDAQ: DIVX), so I encourage anyone who is interested in DivX and the company to read our public documents on file at the SEC (sec.gov).

Cheers. Joe

11. JulesLt - 10:42 Thu 4/12/2007 ( email )

>I don't really care how much of a hack Divx is. It's what the good stuff is released in >so, so its what I need.
So, download the codec and install it into Quicktime. There's even a link to it from Apple's Quicktime components page. The issue is more over the legal liability of distributing it (something that also effects a lot of open source software that uses the Lame encoder or ffmpeg), and what formats Apple want to officially promote.

The thing that's also worth mentioning is that AAC and H.264 are both MPEG supported (and MPEG support tends to lead to ISO support) - that makes them genuine industry standards, rather than 'industry standards'. Apple have always been closely involved with the MPEG group (i.e. Quicktime's .mov container being an MPEG supported format over .avi, etc).

Finally, I'd say the other thing is that people barking about Fairplay also tend to miss the fact that the average number of downloads per iPod is still about 30 per customer, or $25. Which isn't a huge financial lock-in. Even the lock-out of WMA isn't an issue - customers aren't crying out to buy WMA tracks for iPods, because they're not crying out to buy on-line music.

12. Wes Felter - 13:19 Thu 4/12/2007 ( email | web )

Joe Bezdek: You could have built your company around the superior MP4 container, but you didn't. You chose AVI. MP4 isn't necessarily better because it is ISO-approved, but the converse is also true: AVI isn't necessarily better just because DivX Inc. is hipper than ISO. So why not build your de facto standard to be compatible with the de jure standard? Why fork the format?

BTW, don't M4IF and ISMA certify MP4 implementations? Granted, they seem to be doing a poor job.

13. Brian Tiemann - 13:32 Thu 4/12/2007 ( email | web )

"Why fork the format?"

Because it's the Linux way to fork whenever your messiah complex has a disagreement with someone else's messiah complex?

14. danieleran - 13:40 Thu 4/12/2007 ( email | web )

MPEG is part of the ISO.

MPEG-4 isn't a standard, but rather a suite of standards. It includes the MPEG-4 part 2 video codec, which DivX is an implementation of, as well as the AAC MPEG-4 part 3 audio codec, which Apple uses in iTunes. It also includes the much newer MPEG-4 part 10 video codec, aka H.264, which Apple is using; various open implementations also exist, including x264.

MPEG-4 standards are designed specifically for use over networks and mobile devices, so they scale better than the previous ISO standards for video and audio, which were delivered in MPEG-2. DivX is based on minor advances made upon MPEG-2 and carried forward into MPEG-4 (aka H.263), not the newest and much more effecient technology available in MPEG-4: also known as H.264

In other words, people who "compare" MPEG-4 to DivX need to be better informed. DivX is a proprietary implementation of an open standard. Microsoft's own VC-1 is also a proprietary implementation based upon MPEG-4 part 2, but non-standard and incompatible with everything else.

We need open standards for interoperability. We don't need to settle on "the most refined version" of an old outdated codec. We need to keep pushing the state of the art, in a compatible, open, and interoperable way.

Apple is pushing ISO standards, Microsoft is against them.

http://www.roughlydrafted.com/RD/RDM.Tech.Q1.07/6FFE4614-E1AB-45C7-8611-19D06ECCAD9F.html

15. Joe Bezdek - 14:35 Thu 4/12/2007 ( email | web )

@Wes F: Wes, DivX's use of AVI is due primarily to historical reasons. When DivX first hit the scene in 1999, the MP4 container format wasn't yet ready for prime-time. AVI was used not so much because it was judged to be better, but rather because it worked, and was widely supported in various tools -- including commercial and open source applications.

In 2002, when we began to develop the DivX Certified program and enter the consumer electronics market in a big way, we considered the MP4 container format and ran into a practical limitation. (Disclaimer: I'm not one of our codec engineers, so please forgive me if the details of the next statement are inaccurate.) That practical limitation was this: the MP4 container format had more verbose header information and that information was stored at the end of the file. This meant that CE devices needed more memory to read the header information and play back the file properly. This memory issue was more pronounced with DivX than with other formats because a higher percentage of DivX content was longer form (i.e., 2 hour+ movies) rather than short video snippets.

The AVI file format, due to its structure, required less memory to do the same type of play back. We were trying to design a certification program that would support a broad cross-section of the CE devices available on the market, and if we switched to the MP4 format, the memory issue would have limited the ability of our CE partners to support it. So despite the limitations of AVI (which we understand as well as anyone), we stuck with it. (You could argue this is a "lowest common denominator" approach, though we'd describe it as an "open" approach. Be practical. Use the tools that work.)

Times have changed since 2002, of course, and more CE devices are no doubt capable of parsing the MP4 container format for large (2 hour+) videos. Apple certainly seems to have solved it with the iPod. But in this case, Apple's approach of owning the whole widget gives them the advantage of control that we don't have when trying to support a wide variety of products from many different manufacturers.

A couple more points: first, I forgot to mention this in my original comment, but I wanted to correct the original post and note that DivX does in fact have a provision for DRM. We have since 2001, and indeed our DRM scheme is fully supported in virtually all of the 60 million+ DivX Certified products available today. I'm personally ambivalent about the value of DRM, but it's worth noting that we have it. (Another place where Wikipedia is not authoritative. :-))

Second, @danieleran: we support the idea of standards (MPEG, ISO, or otherwise) as much as anyone. But it's worth repeating the point I made in my earlier comment: these are all de jure standards. Until there exists a popular and widespread implementation of these formats, no de facto standard exists. This is a need we recognized early on and have spent the last several years trying to meet. Whether the "de facto" version of MPEG-4 is "DivX" or "iPod/Apple TV" doesn't really matter. DivX's MPEG-4 implementation is no less "standard" or "proprietary" than Apple's MPEG-4 implementation. Both are just widespread implementations of a de jure MPEG standard. Indeed, arguably DivX is less proprietary than Apple, as we make our decode and encode libraries freely available on our web site (which Apple does as well) and license support for our implementation widely to third parties (which Apple does not).

Cheers. Joe

16. tf - 18:56 Thu 4/12/2007 ( email )

Mr. Bezdak, I appreciate your comments. It is your last statement that gets to my point:

"Whether the "de facto" version of MPEG-4 is "DivX" or "iPod/Apple TV" doesn't really matter. DivX's MPEG-4 implementation is no less "standard" or "proprietary" than Apple's MPEG-4 implementation. Both are just widespread implementations of a de jure MPEG standard. Indeed, arguably DivX is less proprietary than Apple, as we make our decode and encode libraries freely available on our web site (which Apple does as well) and license support for our implementation widely to third parties (which Apple does not)."

Firstly, this makes your tech proprietary: a compliant mpeg-4 codec does not inherently have the ability to play a DivX file. A compliant mpeg-4 codec will play any true mpeg-4 file.

And I know you are a licensee (though you joined after taking the draft technology for your own use.) However, it is that you license your tech, but do you or your licensees pay the equivalent encode, decode, and distribution fees due to mpegla? From what I've been told, no. Certainly none of the P2P users who upload DivX files, which is the sole reason people want your support in the first place, are paying distribution fees to mpegla.

Anyway, if I am incorrect, I'd appreciate the correction. Your participation is appreciated likewise.

17. Watts - 20:24 Thu 4/12/2007 ( email )

tf: Why would DivX be responsible for paying licensee fees of content other people encode/decode using their software? Because DivX is the format of choice for video pirates? It's absurd to suggest the latter implies the former. MP3 is the format of choice for audio pirates, but that isn't Thomson Media's fault, and those who make MP3 encoders/decoders (like, say, Apple?) don't have much to do with that, either.

As for "true mpeg-4 files," AVI and MOV are both essentially "container" formats, and Joe already explained why AVI was chosen. Is the official MP4 container format a "better" format? Probably, at least to the degree that it's, well, official. (The file format is, after all, also a de jure standard.)

On technical grounds I agree with you, but it seems like you're basically trying to argue against DivX on the grounds that it's beloved by pirates, to which I think an honest answer is: yeah, so? If AppleTV does well enough, we'll just start seeing more movies ripped to the official MP4 standard. Apple not including DivX support in it from the start isn't Apple taking a moral stand against DivX, it's Apple not being particularly interested in paying licensing fees when they have their own container format already, thanks.

18. Joe Bezdek - 21:30 Thu 4/12/2007 ( email | web )

@tf. A few replies to your last comment.

You write: "a compliant mpeg-4 codec does not inherently have the ability to play a DivX file. A compliant mpeg-4 codec will play any true mpeg-4 file." And there's the rub. While it's true that a generic MPEG-4 codec does not have full compatibility with all DivX files (though there is some measure of compatibility), it's _not_ true that a generic MPEG-4 codec can magically play all MPEG-4 files. There's no such thing as a "true mpeg-4 file," at least in a theoretical sense. There is no a priori (mechanical) test of whether a file is or is not a compliant MPEG-4 file. The only way you know is by writing software to parse the container and decode the content _in_full_. And if you've done that, the software you've written can be called an "MPEG-4 codec" (or at least a "MPEG-4 decoder"). But your measure of full "MPEG-4 compatibility" is only as good as the breadth of test content against which you test your codec.

That said, certainly Apple adheres more closely to the de jure MPEG-4 standard than DivX does. The reason for this is partly historical as I noted, but it's also philosophical. We believe the true "standards" are de facto standards (e.g., like it or not, MP3 for audio, to reference the original topic of this blog post), so we've taken a very practical approach -- we've tried to use what works at the highest quality and lowest cost to ensure the widest adoption in an attempt to create a de facto standard that is useful.

You also ask: "And I know you are a licensee....However, it is that you license your tech, but do you or your licensees pay the equivalent encode, decode, and distribution fees due to mpegla? From what I've been told, no." Unfortunately, I can't reveal the confidential details of our licenses with specific companies, but I'll note that the largest CE companies in the world, including Philips, Samsung and Sony, are DivX licensees. I think it's safe to assume the MPEGLA would have come calling if it wasn''t being paid its license fees.

Cheers. Joe

19. tf - 12:20 Fri 4/13/2007 ( email )

"While it's true that a generic MPEG-4 codec does not have full compatibility with all DivX files (though there is some measure of compatibility), it's _not_ true that a generic MPEG-4 codec can magically play all MPEG-4 files."

I understand this, but you are stretching the argument extremely thin: yes, there is a theoretical possibility that an mpeg-4 encoder (which poorly implements the spec) may not play an mpeg-4 file (which is encoded with an encoder that poorly implements the spec).

But this is not equal to: DivX is just as compliant or more so when an mpeg-4 decoder inherently will not decode your avi containers.

I certainly disagree with this statement: "DivX's MPEG-4 implementation is no less "standard" or "proprietary" than Apple's MPEG-4 implementation." That's just simply untrue.

"I think it's safe to assume the MPEGLA would have come calling if it wasn''t being paid its license fees."

I don't think it is: mpegla has already shown that they are not threatened by "challenges" the their spec and licensing that are marginal and do not result in much gain or profits for their "competition."

Watts said "As for "true mpeg-4 files," AVI and MOV are both essentially "container" formats, and Joe already explained why AVI was chosen." So what? I am not comparing AVI to MOV. I am addressing DivX in general versus mpeg-4 (any and all of its parts).

"Why would DivX be responsible for paying licensee fees of content other people encode/decode using their software?"

Apparently, Joe thinks its a good enough question to respond to.

"MP3 is the format of choice for audio pirates, but that isn't Thomson Media's fault, and those who make MP3 encoders/decoders (like, say, Apple?) don't have much to do with that, either."

Thompson is doing the licensing for mp3. Not licensing a non-standard, proprietary implementation of a spec which it must also license. Paying the licensing fees to Thompson is the responsibility of anyone distributing mp3 content though as it should be with anyopne using technology that uses mpeg-4 patents.

"On technical grounds I agree with you, but it seems like you're basically trying to argue against DivX on the grounds that it's beloved by pirates, to which I think an honest answer is: yeah, so?"

I think you are misunderstooding my position. I am not anti-DivX on moral grounds because of pirating. I am not pro-Apple in this matter. I am anti_Bezdek's comments that DivX is just as standard as mpeg-4, that it is not a splintering of the standard, that is unemcubered with legal issues. Those claims I take serious issue with.

20. Henriok - 16:36 Fri 4/13/2007 ( email )

I respect the merits of de facto and de jure standards but the best thing would certainly be a combination of both, wouldn't it? DivX has the power to do this. Apple is trying to push a de jure standard to a de facto standard, but it DivX is arguing that their product already is a de facto standard but separate from a de jure, then why not just change the inferior AVI container to the superior MPEG4 container and create a complete standard, de facto AND de jure? As Joe Bezdek claims, there are no technical reasons anymore not to. It seems to me that DivX is being stubborn just for the sake of it, and is fragmenting the market just for spite.

21. Joe Bezdek - 00:05 Sat 4/14/2007 ( email | web )

Hi guys. I'll post once more here to provide a couple more responses. In the future, I'll be happy to respond to questions via email (or in another public forum, if you send me the link), but I don't want to comment more on DivX and MPEG-4 here on Brian's blog when his post was primarily about MP3 and AAC.

tf writes: "yes, there is a theoretical possibility that an mpeg-4 encoder (which poorly implements the spec) may not play an mpeg-4 file (which is encoded with an encoder that poorly implements the spec)." You'd be surprised. Our experience in certifying thousands of third-party product models has shown us that poorly encoded content (i.e., content that doesn't adhere to every letter of the specification) is more widespread than you would think. That said, we have to ensure compatibility with a great deal of legacy DivX content of dubious encoding quality, whereas Apple does not. So over time, yes, things are improving and the overall quality of "MPEG-4" content is improving vv. the specification.

tf also writes: "I think you are misunderstooding my position. I am not anti-DivX on moral grounds because of pirating. I am not pro-Apple in this matter. I am anti_Bezdek's comments that DivX is just as standard as mpeg-4..." Remember, when I say "standard" I'm arguing for de facto standards. So I mean to say that DivX is just as much of a de facto standard than is "generic" MPEG-4 (indeed, I'd argue it's more so).

"...that it is not a splintering of the standard..."

I don't mean to argue that we are adhering to the letter of the MPEG-4 standard. We are not. As I mentioned in an earlier comment, we try to choose what works and can be widely supported. I would, however, argue that we aren't "splintering" the standard. If we were calling our format "MPEG-4" (or maybe "4-GEPM" ;-)), then I'd agree with you. But we're not. We've created our own brand name for our format.

"...that is unemcubered with legal issues."

On this one I can only refer back to my earlier comment: "we (DivX, Inc.) are a licensee of MPEGLA, have been for many years, and pay all MPEG-4 licensing fees in full." You'll have to take me at my word that we're square with MPEGLA. As for others who might appear with patent claims on MPEG-4, well, we and everyone else (including Apple) will have to deal with them as they come.

Henriok writes: "I respect the merits of de facto and de jure standards but the best thing would certainly be a combination of both, wouldn't it? DivX has the power to do this....why not just change the inferior AVI container to the superior MPEG4 container and create a complete standard, de facto AND de jure? As Joe Bezdek claims, there are no technical reasons anymore not to. It seems to me that DivX is being stubborn just for the sake of it, and is fragmenting the market just for spite."

Well, we're not being stubborn or spiteful. For the sake of argument, let's assume there isn't a technical reason that the MP4 container can't be supported now. (I'm not saying that's the case. Apple only has to support the container format on two models of iPods it designs. We would have to support it on thousands of models of different CE products from dozens of different manufacturers.) Even if that were the case, AVI is working well for us now and is widely supported. So the MP4 container format would need to offer a user benefit beyond that offered by AVI that is enough to justify the cost of supporting it. At this point, I'm not sure it does. You could argue there is some benefit to supporting MP4 because it would make us fully compliant with the de jure standard, but philosophically, this reason alone is not enough for us. We take a practical approach. We don't support something just because some piece of paper tells us to -- we support it because it works well in a widespread way and offers a real user benefit.

I've enjoyed the back-and-forth. Hit me up via email if you'd like to continue the discussion.

Cheers. Joe

22. tf - 03:31 Sat 4/14/2007 ( email )

Thanks, Joe. I'll probably take you up on your email offer at some point, but your "clarifications" seem to solidify my points in several ways:

1. "Our experience in certifying thousands of third-party product models has shown us that poorly encoded content (i.e., content that doesn't adhere to every letter of the specification) is more widespread than you would think. That said, we have to ensure compatibility with a great deal of legacy DivX content of dubious encoding quality, whereas Apple does not."

Strange, later on you claim you aren't acting as if your format is mpeg-4, but here you claim it is. (More on this later.) Why would I be surprised that there is tons of poorly encoded DivX? I'm not. But this has nothing to do with non-compliant mpeg-4. As you said, DivX is completely different.

2. "Remember, when I say "standard" I'm arguing for de facto standards."

I know. But you appear to be trying to conflate your format with mpeg-4 compliance. Yes, you are 100% using mpeg-4 patents, but you are not complying with mpeg-4 in a standard way. Moreover, I think claiming you are even a de facto standard is specious.

3. "(indeed, I'd argue it's more so). "

And I would laugh at that. You have some adoption in DVD players and some other electronics. Commercially, there is almost zero content. Mpeg-4 is supported broadly across many industries and from top to bottom, from production to consumption.

4. "I don't mean to argue that we are adhering to the letter of the MPEG-4 standard. We are not. "

I know you aren't. But you seem to suggest that poor implementations with DivX has something to do with mpeg-4 when it does not. Why do you conflat compliance when it serves you to criticize mpeg-4 or to market DivX?

5. "I would, however, argue that we aren't "splintering" the standard. If we were calling our format "MPEG-4" (or maybe "4-GEPM" ;-)), then I'd agree with you. But we're not. We've created our own brand name for our format. "

You barely have a format; you recycle 98% mpeg-4 patents and 2% Microsoft tech and you slap your name on it. That is splintering. All you are accomplishing is monetizing and splintering a market for your own gain on the backs of others' patents. Even if you are in compliance with licensing those patents. Just giving it a new name does nothing. Or rather, it defines splintering... well, non compatibility is what TRULY defines your splintering.

6. "You'll have to take me at my word that we're square with MPEGLA. As for others who might appear with patent claims on MPEG-4,"

That I will have to do. I know you are a licensee; I also know you were using their tech before you were licensing it. (I'm sure that issue is resolved with mpegla.) However, you know my questions are about how your licensees are complying with mpegla: are you doing the complying or are they? How is this handled in the contracts? You've already said you will not share this.

"I'm not saying that's the case. Apple only has to support the container format on two models of iPods it designs."

This is hogwash. They also have to do it for their computers. And they are not even doing it: there are thousands and thousands of chips which support mpeg-4 being produced by hundreds of companies. Every DVD player needs to support it. Whether you are focusing on mpeg-4 part or H.264 or whatever, widespread implementation of mpeg-2 was readily achievable... Acting as if developing a reliable implementation of a format is some sort of voodoo is simply FUD.

7. "So the MP4 container format would need to offer a user benefit beyond that offered by AVI that is enough to justify the cost of supporting it."

It would offer compatibility and wopuld make you irrelevent. The question is: what benefit does your format bring at all? It is not 1998 anymore.

"You could argue there is some benefit to supporting MP4 because it would make us fully compliant with the de jure standard, but philosophically, this reason alone is not enough for us. We take a practical approach. "

Not philosophically, financially. If you adopted mpeg-4, the best you could do is become a Fraunhofer providing a "best-of-breed" codec; you wouldn't be creating your own gate into the market.


Again, thank you, Joe. I mean know offense. I understand your motive is to have a sustainable business. I don't fault that. I think your comments tend to hide your main motivations, but still... I greatly appreciate your comments.

(I am working on a similar story, but addressing VC-1, not you guys, that maybe you could help me with... I'll reach you via email.)







23. tf - 03:45 Sat 4/14/2007 ( email )

Joe, I realize that when I wrote: "You barely have a format; you recycle 98% mpeg-4 patents and 2% Microsoft tech and you slap your name on it" that's a bit hyperbolic and facetious: I don't mean it 100% literally. I know you've created your own media container format (again, who knows how derivative it is of .mov/mpeg-4 or .avi/.asf) and that you have done additional improvement of the codec as well as receiving some open source support prior to fully privatizing and commercializing your efforts.

(I just didn't want you to need to correct my statement: I'm aware of your history. I also stand by the intent and thrust of my statement as well.)

24. Anonymous - 16:45 Sat 4/14/2007 ( email )

I just want to correct an assertion above-- from what I've seen, Xvid is the choice of video pirates (TV and movies), not DivX. I watch pretty much all of my TV in Xvid format, and download the occasional movie as well, and unless there's some well of DivX-formatted video out there i'm not aware of, the mainstream sources of pirated video (generally bittorrent and USENET, some IRC sources) are Xvid, with occasional MPEG2 format (for easy DVD or SVCD authoring, I believe is the intention-- those tend to be for, say, Star Trek completists, back when Enterprise was airing) releases.

Not that Apple, MS or Creative need the endorsement of video pirates or pirate-video users such as myself, but I'd sooner buy a Zune or the lovely Creative Zen Vision:W (or the M is nice as well) since they'll play Xvid, a codec "standard" which hasn't been tampered with since late last year, than an video iPod. True that one would want to convert the video for size and possibly shape to put on a portable player, but from what I've seen, the video encoders for DivX, Xvid, WMV, and even RealVideo are free. Quicktime .MOV? Not free.

The video release groups, the pirates in question, by the way, aren't great at evolving standards. TV Shows are encoded either to 350MB or 700MB for hour-long shows (british shows, being a geniune hour instead of a 42-minute-hour, tend to get the latter), a CD standard-- Likewise movies tend to be two 700MB files labelled CD1 and CD2. But who burns to CDs anymore?. Drop the 350MB file to 345MB, and you could fit 13 episodes (typical season length for a cable show such as The Sopranoes or Rescue Me) on a DVD instead of 12. It is common enough, though, to get "HR" releases, which has half the resolution of the original digital broadcast, but the bitrates are again calculated by the size of a CD. HR tends to have AC3 sound, the DVD standard(?) which probably isn't too compatible with any of the portable video players. And then they now release full-size HDTV releases, but I don't even think the small video players could begin to cope, and most likely don't support the container (MKV, Matroska Video).


© Brian Tiemann