The Unwinnable War in high definition DVD
Remember the old Betamax vs VHS standards war?
Sadly, my family made the wrong choice and I have a box of Beta tapes of my old figure skating competitions languishing about somewhere.
There’s a similar war being waged right now between competing high definition DVD standards — Toshiba/Microsoft/etc’s HD-DVD and Sony’s Blu-ray. Blu-ray and HD-DVDs look and generally act like normal DVDs but they support video resolutions that are about 4x the existing DVD format. The most prominent (and cheapest, if you can find one) players are the Playstation 3 for Blu-ray and Microsoft’s HD-DVD add-on for the Xbox 360.
The impact of the DVD format wars is currently spilling directly into the gaming industry’s latest console generation as Sony designed the Playstation 3 to be a fully fledged Blu-ray DVD player as well as a gaming device. Sony’s strategy in this area was pretty sound – use the PS3 to trojan horse a large installed base of Blu-ray players, giving them a potentially large and early advantage in the high def DVD battle.
The execution of this strategy has not gone so well, however, as the Blu-ray part of the PS3 has been one of the primary reasons the device was delayed several times (the PS3 ended up launching a year after the Xbox 360) and also a reason why Sony has been unable to produce the PS3 in bulk. This has resulted in a huge supply shortage and a very small installed base for the foreseeable future. Given how dependent Sony has become on its gaming division to make up for losses in other divisions, these production problems (and the resulting botched PS3 launch) are likely to be cited as a serious misstep for the company, particularly since - for reasons I’ll go into below - it seems increasingly likely that the high definition format war is unwinnable by either the Blu-ray or the HD-DVD contingent. In essence, Sony has bet their future on a technology that is currently very shaky at best, and which has the potential to pull their gaming division down from the dominant position it currently enjoys. It remains to be seen how well they manage to fix the PS3’s current production problems in the second act of the console’s life, but if the device remains as costly to produce and as difficult to buy as it is now well into 2007, it is easy to see Sony’s PS3 related costs running blood red, turning the go-to profit division of the company into yet another money loser for the company. Paul Kedrosky has a good blog post here about how Sony is losing several hundreds of $ on every unit sold. Microsoft employed a very similar upfront loss strategy during the Xbox 1’s launch (as an attempt to gain market share in a market that most people considered pretty locked up,)losing about $4 billion over the life of the console. The problem for Sony is that they only have about $3b in cash and short term investments to lose (from their 3Q2006 quarterly report,) whereas Microsoft still has about $28b and the profitable Windows and Office divisions to keep growing that pile.
There are, two big problems that lead me to believe nobody will ever actually win the high def DVD war outright:
1) The jump from videotape to DVD was a much bigger value add than the move from DVD to high def DVD. Videotapes were large, bulky and prone to being worn out, due to the magnetic tape technology used. They had to be rewound and had no instant chapter jump features. There was no way to add multiple audio tracks, flexible subtitles or crew commentary features. With high def DVDs, the only true value add is the higher resolution, and while high def does look noticeably better than standard DVD resolution, the difference between a good upscaling DVD player playing a standard DVD and a full high def DVD are not, in my opinion, great enough for the masses to throw out their existing movie collections and start over. The jump from standard DVD to high def DVD formats is more like the move from CD to SuperCDs or DVD-Audio discs than from videotape to DVDs, and those formats never really caught on in the mass market. High definition DVDs will do well with the videophile niche and certainly won’t disappear but they also don’t really have a chance to become mass market for years and years and years when there is no difference between the price of current standard upscaling DVDs (less than $100) and HD-DVD and Blu-ray players. 2) The reason why I don’t think either side will ever technically win this “war” is that downloadable video is likely to take off long before either side can win over the mass market, obsolescing physical media formats altogether. With services like iTunes video store, Direct2Drive and Microsoft’s Xbox 360 video download service and new hardware like Apple’s iTV soon to be on the scene, coupled with really cheap cost per gigabyte harddrives, the (near) future is Tivo-like devices with terabyte+ harddrives and always-on network connections with on-demand video services that stream and download movies in HD resolution. The movie companies seem to be doing a decent job of not making *all* of the same mistakes the record companies made during the MP3 boom, so I believe this future will materialize before either high def DVD standard can declare victory.
I think Microsoft played it smart with their HD-DVD/Video Marketplace strategy. They’re supporting physical media in an opt-in way, i.e. you can get an Xbox360 and then decide later if you want to invest in the HD-DVD drive add-on, vs Sony’s PS3 which costs 600 dollars whether you want the Blu-Ray player or not. I agree with you that digital video on demand will outpace physical media very soon and Microsoft’s Video Marketplace can pick up from there.
Until video on demand becomes the most common form of video distribution, Microsoft can refine and improve their Video Marketplace all the while appealing to the videophile niche with their HD-DVD add-on.
Downloadable high def video distribution will be held back by available bandwidth to the home. An HD movie is about 25-30 Gig. No one will be quickly downloading files that big over the current infrastructure.
If I can’t get the movie “right now”, if the download is an overnight affair (as it is with BT with current broadband access), the netflix of the world will be just fine.
i think you are exactly right susan. sony and toshiba, and everyone who lined up behind them, have waged a bitter war over what is likely to be a short-lived interim distribution system.
I disagree with Erik here. There is already high def video distribution to the home… it’s called cable & satellite. A Tivo Series 3 with 1TB storage is doable today (see my blog post here, it’s one of the most popular). With that, you can set up a pretty long wish list and acquire a pretty large HD library. It’s not quite on demand, but it’s “good enough ™”.
The only way I see the HD DVD market taking off is if someone cracks the copy protection freely, and the PC companies embrace “ripping” of DVDs the way they embraced the “ripping” of CDs. People inherently prefer to get a physical copy, but they want the ability to have the content anywhere once they pay for it.
Not saying that the MPAA wouldn’t be aggressive here legally, like the RIAA, and with the DMCA in effect, it’s an upward legal battle. But the genie is out of the bottle for broadcast HDTV, like it was for unprotected audio CDs. If they don’t match the offering with HD DVDs, people will turn to the path of least resistence, which right now is a DVR-based video acquisition system.
Hi Erik,
25-30 gigs is an overly large estimate of the size of an HD movie. HD-DVDs can only hold 30 gigabytes maximum and that is using dual layers (single layer discs can only hold 15 gigabytes) and that size is meant to allow for lots of extra features as is the norm with current DVDs.
Microsoft already has an HD movie download service available for the Xbox 360 via Xbox Live, its HD movies average about 6 gigabytes per movie and more importantly than the size estimate, the service supports streaming, so you can start watching the movie a couple of minutes after the download starts.. the movie is saved as downloaded to disk and as long as the user has an uninterrupted but fairly generic broadband connection, there is no overnighting involved.
Granted, Microsoft’s service got off to a rough start for the first few days it was in operation, but that’s pretty standard for any network service where massive scalability is required, but this is all very technically feasible and is even being done right now, though on a smaller scale than it will be in the future.
In regards to xbox live… H.264 is between 7 and 8 Mbps in full HD, that is NOT a generic broadband connection. Verizon DSL is 768K (can be up to 1.2 Mbps), cable modems are (can be) faster (1-4 Mbps) but they are contention based, their speed depends on the number of subscribers in the cable node. In other words, if you’re the only one in your neighborhood trying to stream HD, you’re fine (if you fill your buffers for a while), if you get a dozen people in your neighborhood trying to do it, you all grind to a halt.
Finally, calling 720p (which are the 6 gig xbox files) an HD resolution is a tad deceptive.
So the qustion is, if you need to pay your cable company an extra $25-50 a month to boost your cable modem from 1-4Mbps, and you need to pay for the movies ON TOP of that, why are you not paying netflix $20/month?
As for the tivo solution… If all you want to watch is timeshifted broadcast television, it’s great. Some of us have more varied tastes.
Back in the day (late 1980s early 1990s) when I was working on interactive television and VOD systems, we used to say the fastest way to get 45 MEG of data cross country was a syquest drive cartridge and a fedex envelope. The numbers are bigger now, but it’s going to be a while before the cable companies are going to spend the money to improve their infrastructure to reliably get 7-8 Mbps to the home.
Eric, I think what Adam was getting at is that if companies like Comcast can already deliver HD content through current broadband infrastructure (whether its television or OnDemand videos) then there is really no reason why those same companies couldn’t refactor these services into something more akin to iTunes video store or Xbox360 Video Marketplace.
Furthermore, digital distribution offers far more customizability with respect to consumer choice. Erik may not think 720p qualifies as HD, but the vast majority of HDTV’s in the world today can only handle 720p and with digital distrubtion it is now up to the consumer to decide if they want to pay 3 dollars to rent a standard-def video, 4 for 720p or 5 for the glorious 1080p. I chose these prices arbitrarily but such high-grained tiered pricing just isn’t feasable with physical media.
The same customizability allows customers to decide which extra features they are willing to pay for (or wait to download) and can help appease some of the bandwidth issues surrounding HD content delivery. As Susan pointed out, much of the 30GB in HD-DVD will be used for extra features that many people just aren’t interested in. So, if they don’t want it maybe they won’t have to pay for it and they could certainly be left out to decrease the download time.
“there is really no reason why those same companies couldn’t refactor these services into something more akin to iTunes video store or Xbox360 Video Marketplace.”
Yes there is. The real limitation right now for the growth of streaming, on-demand HD is the HFC network (hybrid fiber-coax) architecture of most cable plants. It comes down to node size and how much bandwidth is available within the node.
Remember it’s a contention based system, all the homes within the node share the same bandwidth to and from the node. While it is comparitively cheap to increase the bandwidth from the headend to the node, it is REALLY expensive to increase the bandwidth within the node. All the homes passed within the node share that commodity.
If the loop within the node is 100 Mbps it doesn’t take many people trying to stream HD at 5Mbps to bring the network to a crawl. Node sizes range from about 250 homes passed to around 2000 homes passed. Average node size is about 750 homes passed.
Hi Erik -
What you seem to be talking about is “full HD”, by which I assume you mean 1920×1080 content, which may be a big deal for videophiles but not so much for most people with HDTVs. Because the vast majority of HDTVs sold thus far can only display 720p content anyway (because the native resolution of the display is only 720 or 768 vertical pixels) and the TV just down-converts 1080p or 1080i signals to 720, wasting any bandwidth spent on delivering a 1080 resolution movie to the user.
You seem to be a big proponent of Netflix, which is a fine solution but a long way away from on demand. I like Netflix myself, but if you can wait a week for DVDs to arrive from Netflix, it seems like you can probably wait a few minutes for enough of an HD movie to stream to be watchable.
And lastly, Microsoft’s solution was brought up simply to show that an early version of the technology exists right now. Mainstream adoption of HD-DVD/Blu-ray formats is 4-5 years out, minimum, IMO and I’m positive that full and scalable HD VOD/streaming services will be there by then. Of course this is just a prediction so there is no way to prove it, but I can’t see how it won’t happen unless the MPAA types try to step in the way like the music guys did with MP3, but all signs (iTunes, etc, though not HD yet) point to them not making the same mistakes.
A lot of the bandwidth problems can be solved by the fact that harddrive space is ridiculously cheap now and getting cheaper by the minute. The on-demand service could learn from TiVo or Valve’s Steam (which is for videogames, but is still relevant as they are serving multiple gigabyte streaming game downloads) and automatically background download movies you’re likely to be interested in using an encrypted format before you ever even buy them, as soon as they are available on the server, so they are there or mostly there when you finally get around to selecting to watch them.
I’m not being clear.
Let’s assume we have a compression breakthrough and you can stream HD using 1Mbps of bandwidth.
The total bandwidth available within an RG59 HFC node is about ~30 Mbps.
All the homes within that node SHARE the same 30 Mbps.
A node is 500 homes.
Within those 500 homes, the node loop can support only 30 concurrent 1 Mbps HD streams. The other 470 homes are SOL. The system works now because most users are using very little bandwidth at any given time. The contention rate is very low.
Until you increase the bandwidth WITHIN the node, the system can not scale. It will work now, when the market penetration of xbox live is so low, but if you’re talking about mainstream adoption, you will not be streaming on demand HD over IP until the MSO’s upgrade and make MUCH smaller nodes ($$$) or preferably fiber to the home ($$$$$).
Here’s a not-so-technical description of how an HFC cable system operates…
http://en.wikipedia.org/wiki/Hybrid_Fibre_Coaxial
I think the article pretty much nails it. There is no doubt that VOD is here today. Over time those services will only improve, as will the selection. Who wins this is anyones guess, but the whole market is moving in this direction. In the meantime, Sony and the HD-DVD would rather wage a winless war, then to either agree on a standard or agree to support both standards. The result is that consumers are being trained to accept inferior video quality of DVDs and VOD over the higher quality content that is HD-DVD. The longer the battle wages on, the better VOD gets and the more used to it consumers become. The end result is that the studios lose the fat $15 prices they can charge for a movie and consumers only pay $3 or ad support for their movies. If the jump to HD content was so mind blowing maybe one could argue that this battle would be worth fighting, but it’s really just a marketing ploy designed to sell more DVDs and DVD players. The jump from cable to HDTV cable makes a pretty big difference, but the jump from DVD to HD-DVD is actually pretty weak. Let them keep fighthing their silly war, one day they’ll just wake up and realize how much money they lost because they were too greedy.
There is no war. By the time there are enough titles to make the average consumer curious, players will simply support both formats and the whole thing will no matter a la DVD +/-RW.
Wow. Quite a thread here. Touchy subject, I guess…
Just to clarify part of my point, since it seems lost here:
I think the entire “on demand” piece of the home delivery of video may be over-rated. The cable/satellite companies are already delivering an incredible amount of HD video over many channels today. It’s like a message bus, and with a reasonably intelligent home server with extensive storage, by the time you want your video “on demand”, it could already be collected on your home server for no additional cost.
A couple of years ago, I drastically upgraded the storage on my DirecTivo. I had always thought I had ample storage, but when you get your storage to the point where you can hold 100-200 movies, your usage pattern changes. Simple wishlists for broad categories of content like, “all science fiction movies” or “all movies with Will Farrell” can easily be created, and the movies stored indefinitely.
As a result, it’s fairly rare that I want to “sit down and watch a movie” and have to download anything… typically it was already recorded, weeks ago, off some channel at 2am.
I largely agree with Adam. Storage gets cheaper way way faster than bandwidth. Storage was about 1$/meg in 1995. Ten years later terabyte drives are under $500, in 5 years 10-20 TB of storage in the house is a perfectly reasonable assumption. In 10 years 100 TB is reasonable.
Now how do you fill those drives? Bandwidth isn’t growing nearly as fast as storage.
Why not sell them full? You can fit a large percentage of enduring culture on 100 TB of storage.
I’ve been thinking a lot about the infinite storage issue. Storage is becoming free. Bandwidth is not. I’m not sure what I think yet, but the equation includes:
Speed of propogation of brand new media to remote servers
Shape of the value vs time graph of the media
Amount of bandwidth required to get the media to remote storage
For example, no one should ever stream Casablanca or the Godfather, no one should ever stream anything from the Beatles or Stones catalog. They both have essentially infinite shelf life and fairly steady value over time. The value graph is flat and high.
But if you have a blob of media (say from American Idol), and it takes 3 days to propogate to infinite storage tivo on the edge of the network, and the shelf life of said media is a week until the next show, then perhaps streaming makes sense. The value graph of the AI content spikes immediately, then trails off to nearly zero very quickly.
The hard part is the graph is not the same for everyone. The ideal situation would be a hybrid. The network knows what I like and that alters the delivery order to match my usage patterns. So if my niece is an American Idol fan, AI shows up on her iPod right away, her value v. time graph peaks early, very high and trails off quickly. But me, it can show up 3 weeks (or months) later and it won’t really matter, my value v time graph is essentially flat (and very low).
[...] The Unwinnable War in high definition DVD « Susan Wu – Venture Capital i watched the Lord of the Rings animated film several hundred times as a kid – why? well, yes, it was good, but also b/c it was one of maybe a dozen Betamax titles at the video store – format wars suck for consumers (tags: DVD Wu formats VHS Betamax) [...]
You have great insights here. You might be interested in the articles found on Optical Storage .org
It’s another ugly war, alright, but it’s just like these companies to not back down. Hardware-wise it’s too early to tell, right?
[...] The Unwinnable War in high definition DVD « Susan Wu – Venture Capital Share and Enjoy:These icons link to social bookmarking sites where readers can share and discover new web pages. [...]
In Australia we have ADSL2+ which is up to 24mbps. I’ve only got an ADSL 1 modem so i regularly get 800kiloBYTES a second which is 8-ish mbps…that’s miles higher then the on-demand video service i use to rent and download movies. (www.reeltime.tv)
If they did 720p @ 7 then i’d be streaming HD through Mediacenter and straight into the Xbox. it’s almost here…