Hacker News new | comments | show | ask | jobs | submit login
Why Intel's New IPTV Service Will Do What Google, Apple, and Microsoft Can't (forbes.com)
31 points by kellyhclay 1784 days ago | hide | past | web | 35 comments | favorite

I don't understand how the writer supposes Intel can acquire unbundled content and sell it per-channel or even per-show.

Yes, that's the consumer's dream, to be able to subscribe to and pay for only one's preferred channels (or even shows), but that well-known desire has been consistently opposed by the cable and satellite systems. And content producers have, I believe, always cooperated with the cable and satellite providers in this. It's one more item for negotiation: "we'll charge less for our material if you'll bundle us with a popular group of channels", etc. Hulu, iTunes, et al only get to rebroadcast single shows only as reruns, so providing a trickle of secondary income but never threatening the main cash cow.

So it seems highly unlikely that even Intel could acquire the right to broadcast first-run material in competition with other systems. Also the cable and sat providers have plenty of pricing elasticity acquired from years of semi-monopoly, and could easily compete on price if they wanted to.

This is what I was wondering, too. It's basically the critical step in being able to make TV a viable product for any* company (cable companies included).

The only thing I noticed the writer saying along these lines was that Intel offered billions in long-term commitment to the industry while Apple/Google only offered tens/hundreds of millions. I am not sure I buy it; both Google and Apple know what that market would be worth if tapped (20-30% * enormous number) so I don't think it was a lack of money at the negotiating table.

Rather I think the TV is going to be stuck in neutral for a long time due to the complexity of making that work. In music and mobile phones, the N in the N-party negotiations for acquiring a critical mass was much smaller than in the TV business. Unlike the music and mobile industries, there are many more exclusive and long-term contracts that make it hard to maneuver around. The TV problem is certainly not a technology, bandwidth, or market-size (available devices) problem as of the last 2-3 years.

Is it possible Intel cracked this nut? Yes. Likely? No.

The only thing I can think of is that they're going to sell TV below cost to get people to switch, then try to renegotiate later.

Xbox and amazon both offer show subscriptions, why is it hard to believe that Intel can't get the same deal?

Channel by Channel is the only "new" thing, and while it might be hard to believe Scripps would unbundle all their channels, it would not be hard to believe that BBC, and Nick, and Disney would be VERY happy to have their channels not muddied with the offerings of "lesser" channels

Looking on Amazon I see you are correct in part, you can order a season pass for the current season of, e.g., CSI NY.

However, "TV episodes are delivered as soon as possible after the episode is aired by the network (usually the following day)" (Amazon's about-TV-passes page). That's not at all the same as watching it in its "proper" time slot on CBS via DirecTv or Comcast. If Intel offered show subs only on the basis of a 24-hr delay they'd be at a big disadvantage.

They could make up some of the difference by offering a really smart dvr-type system with a clean interface (pay for all the Tivo patent licenses and go beyond them), so their set top box knows what you want and gets it automatically with no fuss -- but they'd still be selling yesterday's TV today. And the cable and dish providers can play the smart dvr card as well, with current content.

And in any case, there are times when you want this minute's content this minute: news, commentary, sports. When Intel gets a contract to resell ESPN or Fox Sports Net, live, I'll be a believer.

There is no check sufficiently large to convince a $130B industry to smash their business model into bits. And the second they unbundle, half of their content pipeline evaporates overnight.

Cable companies like bundles more than Content providers do. And Producers would like to cut much of the "pipeline" out of the picture.

Shows like Firefly would have done much better in a world where consumers paid directly for the content, than a world where the networks determined air dates, and messed with schedules.

Shows like Firefly don't get made in a world where consumers pay directly for the content.

Not yet..

I do agree such a world would provide clearer signals. But while I admit that I am speaking well out of my depth, I don't see much evidence that people currently value television above 'Free' or 'Unlimited After Monthly Subscription'.

Most people don't value anything unless they pay for it, and they don't pay for it unless they absolutely have to. That's an Iron Law of Humans.

Sure dude, and where are you going to get the development budget for a show like Firefly? Kickstarter?

A market like the one between startups and VCs?

I'm pretty doubtful. Intel have been doing in-out dance for TV solutions for a few years. They don't have consumer product experience or media market experience. Would be stunned if this happened on a substantial scale.

Also, what clout do they have with the content providers? Why would somebody release control to them rather than Apple, Microsoft or Google...which have larger, more-established viewer bases.

Well money talks and Intel could afford to buy content but I just don't see the business case for them to outbid the existing channels and cable companies for exclusive rights for premium content.

And Apple can afford to buy content creators outright. Without any mention of deals having been made with movie studios etc. the whole article is nothing more than a pipe-dream.

What planet do you live on? Even with $100B in the bank, you are going to buy A (1) content creator out right, and not the biggest. Lucas Films which has a few dozen properties, and a few hundred hours of content, went for $4b. It is a VERY small content provider. Disney would cost in the neighborhood of $250B

Viacom just bought Bellator (MMA / UFC content producer) for $225M. So apple could afford a few producers of a single show.

Your numbers are way out.

Disney has a market cap of $87B and that gets you Disney, Disney Theme Parks, ABC, ESPN, The Disney Channel, A+E, Lucasfilm, Industrial Light and Magic, and a hell of a lot more to boot.

News Corp has a market cap of $57B and that gets you 20th Century Fox, Fox, and a long list of international broadcasters and producers.

Viacom has a market cap of $26B.

Any of those gives you pretty much all the content you can shake a stick at in both TV and Film, as well as several broadcasting networks both in the US and globaly and a long list of other assets that would likely be packed up and sold off. Even accounting for a leveraged buy-out premium, Apple could buy either Disney, or News Corp and Viacom and instantly be one of the worlds biggest media companies.

Market cap, an "price to buy" are not the same. Anyone who has been through a purchase can tell you that. Why do you think Lucas was so expensive?

Market cap is the price of the outstanding stock, and doesn't reflect the cost of the assets, or the will of the primary holders to keep their shares.

This article doesn't say anything about where the content will come from. Possibly Intel hasn't got that resolved, unless they will front for the content providers as an independent initiative (you bring the hardware, we bring the content).

Apple is just toying around the market. It's obvious for everyone that: a. People would love having a huge Apple display in their livingroom, playing nicely with their mobile stuff and Apple-ized life b. When Apple goes into this market, it will go with an entire TV set and not just a set top box; this TV will not be bound to the low margins of mass market TVs, either

As for Google, they are the long term winner. You WILL need to feature Google services if you want your product to be more than just a dumb streamer. Now they are doing their little fiber experiment which, in my eyes, is just a precursor to a massive rollout of infrastructure fast enough to support high quality IPTV.

I'm sure we will see Youtube stuff coming to our livingroom in few years.

In my eyes, Intel got nothing to bring to the table here. The problem was never hardware, that exists for many years now. The problem is content first, software second, hardware in a very distant third.

There is some hand-wavy mention towards the end of Intel going to "Hollywood" with a bunch of money. Whatever and whomever that means.

The only way I see it happening is if the content providers decide to do a preemptive move and base their own STBs on top of some Intel platform. But it will not be an Intel STB nor will it be marketed as such.

Intel has been willing to work with Hollywood in the past. There was a lot of controversy over their inclusion of "DRM" hardware in their Sandy Bridge Chips.



I'd love for more competition in this space. If real IPTV could finally happen it would be great. I use a HTPC with Windows Media Center as my main TV, and I'm tired of dealing with the issues. Cablecard is a pain, Comcast is a pain, commercials are a pain. I just want to watch the few shows I care about.

The only thing stopping me from cutting cable and just watching shows on iTunes is live sports.

one cheap price is more interesting for me, and if the content is available on netflix, that would seal the deal for me

All I want is to replicate the experience I can currently get from uTorrent. ie - high res content downloaded [not streamed and thus vulnerable to buffering] to watch at my leisure around the time of initial cable viewing.

The price for doing this legally should probably be similar per episode to what I pay for a song on iTunes.

I'm happy to pay for a netflix sub, but it doesn't do what I want. I hate buying boxed sets (though I sometimes do) because playing DvDs on my PC is worse than having a quality .avi file neatly filed and quickly accessed.

In fact (I've said this here before) my ideal world is where Steam holds all of my purchased series and I can manage it all from there. It remembers stuff I no longer want clogging up my HD to be re-downloaded, and downloads the latest episodes of stuff I've bought automatically.

What I want will never happen.

Eugh, more abuse of the term IPTV. IPTV is a standardised term requiring a managed QoS network (i.e not the Internet) so you can match the QoS provided by managed RF spectrum or dedicated cable frequencies.

IPTV does not require mangaged QoS -Brandon Wirtz Google me I know more about IPTV than you do. (or just about anyone on the planet)

If you knew more about IPTV than I did then you'd know that there are internationally standardised definitions for what "IPTV" is designated as compared to "Internet TV".

QoS of the network is not a requirement for IPTV. IPTV is a definition for Broadcast like experience to a Television display.

Several "IPTV Named Standards" work over Multicast and require "IP Traffic Quality Metrics" this is not "QoS"

PS I authored many of those standards... (I was also on the committee for BD, h264, VC1, and a few other SMPTE standards)

Now you're just splitting hairs about the definition of QoS, which again is defined in standards and fits in with the standardised definition of IPTV. I'm not going to quote it all because the chances are there is another standard with slightly different wording that can be construed to mean something else.

It doesn't change the fact that the article is describing an "over the top" network where content is delivered over the public internet which isn't suitable for a broadcast-like television experience (start torrenting on another machine and you'll see why). It also ignores the fact that the content will almost certainly be delivered in low-bitrate/poorly encoded and half-framerate (for full frame/field rate material) like every OTT service out there. Try watching sports at 25/30fps on a television and your headache will tell you that it's not a television experience.

You are horribly ill informed. Xbox Live, Hulu, all of those are at bit rates and quality roughly what a Sports offering would be doing, and work just fine. I'm not splitting hairs, you just have never worked with a CDN, or done any real work with a real budget. The tech and pipes to do this has been around for 7+ years.

First of all that’s a lovely ad hominem based on absolutely zero knowledge of the projects I have worked on. I have led teams for Pay-TV deployments on satellite and OTA with budgets orders of magnitude larger than any live streaming package and deployed hundreds of television channels. Of course you make the same claim again which I’ve had time after time from engineers like Microsoft, Google etc who think web is the be-all and end-all. I use the following technical arguments each time to bring these guys back to earth.

First of all you totally ignore the point I made about high bandwidth usage affecting the viewing experience. Do you really want to sit there on Patch Tuesday waiting for the video on your TV to buffer because your PC in the other room is downloading updates? At best you’ll see huge quality fluctuations as the adaptive streaming tries to react. At worst you’ll miss the goal waiting for the stream to buffer. A managed network gives you a similar broadcast experience to licensed RF spectrum.

All the web services offering “HD” are based around a simple concept – you can get acceptable quality at web bitrates when you’re delivering p24 content with buffer times of around 5-10 seconds (the usual chunk length in an adaptive streaming system), which gives you a huge VBV to work with and a huge keyframe interval. Contrast this with broadcast and its ~0.5-2 second channel change times. Good luck convincing users to wait 5-10 seconds to change the channel (channel-surfing is part of a television experience). The fact is to deliver p50/60 content for sports you need significantly higher bitrates than web bitrates – with current bitrates you’re just watching blocks all day because of the necessary smaller keyframe intervals and VBV. It is nowhere near to a sports offering that people are paying the equivalent of a cable subscription for - even more so on a larger screen. That doesn’t even take into consideration the poor quality of web encoders.

Overall latency is nowhere near television-like. Web streams are 30s to 1 minute delayed compared to television streams thanks to CDN buffering and adaptive streaming chunks. Different viewers in a household end up getting streams that are hugely out of phase (demonstrated during the Olympics where a race on one stream hadn’t even started when it had finished on another). Your neighbours have already finished celebrating the goal by the time you see it. The technology to deliver web streams to relatively small numbers of users has been around but once you get to large numbers the capacity is not there – it’s one reason adaptive streaming is useful; once your ISP runs out of capacity you can be adaptively streamed to a low bitrate but still get a picture. In my city (Bristol, UK) on cable, YouTube interconnects are congested and only 240p works as result.

It’s no surprise live streaming OTT is provided as an add-on to Pay-TV – operators know it isn’t good enough yet to charge for as a standalone package. Some operators are going to use streaming to deliver their channels because of problems with multicast but they’ll be able to use DiffServ to maintain a decent experience.

I think the battle is lost on that one. When working in these areas I tried to distinguish between "Managed Network IPTV" and "Internet TV" when the difference mattered and tried to educate those discussing these issues. "IPTV" by itself is ambiguous and it can normally be understood by context.

All the perceived problems that Intel 'solves' are all legal and copyright issues. Tunnel technology just works. Hell, there's even an autodownloader for TPB on one of the open source media players, so this stuff isn't theoretical.

I'm gueasing this is some sort of hackeneyed scheme to get DRM embedded in more of our devices.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | DMCA | Apply to YC | Contact