Original URL: http://www.theregister.co.uk/2007/04/14/microsoft_vc-1_codec_analysis/

Microsoft mugged over VC-1 codec patent terms

Self-inflicted victory

By Faultline

Posted in Broadband, 14th April 2007 12:02 GMT

Comment We meet people on the various IPTV, mobile TV and web video circuits who always comment that VC-1, after a flying start, has fallen back and that pretty much these days the codec of choice is either VP6 from On2 Technologies for web video and H.264 for everything else, with no VC-1 in sight.

Just over three weeks ago the MPGE LA issued a final license for the Microsoft-inaugurated VC-1 codec, after forming a group to assess essential patents and to discuss terms for it, back in March 2004, the process taking precisely three years.

The license may in fact only be a license by which Microsoft can pay everyone else fees. There are no less than 16 separate companies that are deemed to have patents that are essential for VC-1 to work, the latest addition to the 15 company list that was first issued last August, being Korean number three handset maker, Pantech Curitel.

Between them they speak for 125 separate patents that are listed in the license, and a total of two of these are allocated to Microsoft. Every other supplier of essential technology including Daewoo, France Telecom, société anonyme (which presumably includes patents from close associates of France Telecom also) Fujitsu, Philips, LG, Matsushita, Mitsubishi, NTT Pantech Curitel, Samsung, Sharp, Sony, Telenor, Toshiba JVC, have more patents at stake that Microsoft (except for two of the above) and many of them have many more.

Effectively Microsoft has been mugged by the attempt to make its VC-1 technology a standard through the Society of Motion Picture and Television Engineers. In so doing it had to reveal how its codec technology worked, and offer a license, and in going to the respected MPEG LA as a patent pool agent, it exposed its technology to all the know how that went into licensing the MPEG 2 and MPEG 4 Level 10 AVC/H.264 codec that has stolen the market.

Usually all of the combatants (essential patent holders) will argue for some basis or other for splitting the royalty stream, and these rules are not public and they are not always the same. But as a general principle the more patents you hold, the bigger your slice of the patent pool pie. Now that’s not absolute, but it can be an indicator.

On that basis every company except Telenor and Sharp will end up getting more money when VC-1 is used, than Microsoft. Of course Microsoft can charge what it likes for any software implementation of its own, but since it has always chosen to “give away” its codec with its media player, this means that it has to cover any licensing bill for patents out of virtually zero direct revenue.

Microsoft's humbling over VC-1 shouldn't be under-estimated. When we last heard a Microsoft official speak about this back in 2004 they said that everyone should use VC-1 because it was charging less than MPEG4/H.264, but that was back when it thought that it owned all of the technology inside the codec.

It's no wonder that the use of VC-1 has fallen by the wayside. Why would Microsoft push it? The more successful it is in getting other people to use it, the more its rivals in the consumer electronics space get paid royalties.

When the patent pool was first put together and the negotiating began, both Sony and Philips were known to be considering taking legal action against Microsoft. They had already been successful in taking Microsoft to the patent courts in the US, over DRM, in support of their acquisition Intertrust, where they won a $440 million settlement. In a way that’s more of less what they’ve won here.

Back in August when the draft license terms were issued, it was made clear that Microsoft was the only company that would have to pay “back payments” prior to 2006 (Microsoft has been bundling the same or similar technology in its PC operating systems for around 10 years).

The patent position for H.264 looks remarkable similar to VC-1, which isn’t too surprising since everyone we’ve spoken to says that the technologies are virtually identical anyway. Common patent holders for both the VC-1 patent pool and the H.264 pool include France Télécom société anonyme, Fujitsu, Philips, Matsushita, Mitsubishi, Samsung, Sharp, Sony, Toshiba, JVC and of course Microsoft.

Terms for building a device with the technology in splits into two formats, if it’s for a specific function in a CE device it costs nothing for the first 100,000 licenses; 20 cents each for up to 5 million licenses; 10 cents for all licenses above 5 million and there is a royalty cap of $5m a year. But if the royalty is for a PC operating system, which can of course be used for multiple applications, the cap is $8m.

If Microsoft is forced to pay back 10 years of back payment for royalties (as was suggested in the August 2006 draft license) then it would have an up front payment of something like $80m, divvied up between all 16 companies, and a continual $8m a year – for something that it was claiming it invented itself.

Content companies also pay license fees, and these also split into three, paid for title by title, where a customer pays for the service or product, and where customer either get encoded content for free, such as broadcast Television or as part of an overall subscription.

For this short videos under 12 minutes bear no royalty and thereafter pay the lower of 2% of the price paid for the video or $0.02 per title. This is for replicators of physical media, and service providers such as cable, satellite, video DSL, internet and mobile, for VOD, Pay per view and electronic downloads. This caps at $4.25 million a year until 2012, and rising after that.

Where remuneration is for other sources, such as advertising the licensee pays either a one-time payment of $2,500 per VC-1 encoder or an annual fee per of $2,500 per broadcast market where a market is at least 100,000 people or $5,000 per year per for a broadcast market up to 500,000 people and $10,000 a year for each broadcast market above 1,000,000 or more television households.

The two patents that were included for Microsoft as being essential for inclusion in the patent pool related to predictive image compression using a single variable length code for both the luminance and chrominance blocks and another for the efficient macroblock header coding.

Coincidentally we’re sure, a few days after the terms were finalized a number of US Broadcasters including NHK, TBS, NTV, TV Asahi, Fuji TV and TV all decided that they would adopt the H.264 encoding for their future programming.

This came after there were hiccups over the H.264 license which has also been in discussion for many months, and this was not thought to have related to the VC-1 patent pooling license finally being settled, although the terms of the license are virtually identical, based on a payment on a $2,500 one time fee for each encoder used.

So after all this time the two technologies, considered by many so similar as to be identical, now cost virtually the same to license as one another, much to the chagrin of Microsoft’s and contradicting its earlier statements. The intervening period of uncertainty has pretty much left VC-1 encoding only scarcely used, except a little on the web, with the exception of its use as an option in both High Definition DVD standards HD DVD and Blu-ray. But this only came about as a snub to the MPEG4/H.264 license, which at the time was thought to be too expensive.

Copyright © 2007, Faultline

Faultline is published by Rethink Research, a London-based publishing and consulting firm. This weekly newsletter is an assessment of the impact of the week's events in the world of digital media. Faultline is where media meets technology. Subscription details here.