Regardless of where the bottleneck occurs, in the end, user experience suffers and that is something that the average TV watcher is not going to understand. They will just assume Apple's TV sucks because they can't watch their favorite show without interruption or picture quality issues. Even if it is the ISP's problem, it affects Apples reputation. I think Apple will have to measure the roll out as they have done with other services simply to maintain quality and then slowly launch additional channels as the infrastructure allows.
Only if it's on-demand streaming or only if they don't have content servers in local areas. The first doesn't seem like Jobs "cracking" the TV market and the latter simply doesn't make any sense unless you neither sell nor support shows in those areas.
If you have local servers and the streaming is NOT on-demand, like like with cable/sat then your "hops" are very limited from the local server to the user. This currently isn't much (or any) of an issue for cable companies today.
Technically downstream can have bandwidth issues on digital coax. I first saw it with the movie Gladiator during an action scene with a lot of extra data. It didn't just affect the channel it was on but also the frequency it was sharing with another channel. This was all last mile stuff.
Touchè. But my watching a TV show here in Vancouver on my cable provider won't affect a users net speed in Florida, which it could if I was watching something on an Internet based service.
If we're talking about all channels being streamed on-demand, but I can only see this being done in the same way cable/sat is currently done, by offering most content at the same time across all distributors. That means that the content is fairly local. Likely using Akamai, as previously noted. Last mile could suffer if too many people are accessing that local server for a specific channel but I wouldn't expect anything close to what can happen from too many people accessing a website at once.
Basically what I'm saying is that the Internet is getting crowded, and is only going to get more so without some infrastructure improvements. I would hate to see an Apple TV suffer because of it. Wanting to be able to to ten things, but only being able to do four because of bandwidth restrictions
Regardless of where the bottleneck occurs, in the end, user experience suffers and that is something that the average TV watcher is not going to understand. They will just assume Apple's TV sucks because they can't watch their favorite show without interruption or picture quality issues. Even if it is the ISP's problem, it affects Apples reputation. I think Apple will have to measure the roll out as they have done with other services simply to maintain quality and then slowly launch additional channels as the infrastructure allows.
I agree.
But Apple do have some options. Most streaming services shove out a bitstream of pre-encoded data with a fixed bitrate.
Either the connection is adequate - leaving users happy. Or the connection is inadequate and we get constant stalling and buffering.
If Apple are doing this the way I expect, they will be building and compressing a custom bitstream for each customer. They can raise and drop the compression ratio according to what the client device is telling them.
If they do this, the Apple TV picture will degrade - but it won't drop out.
Basically what I'm saying is that the Internet is getting crowded, and is only going to get more so without some infrastructure improvements. I would hate to see an Apple TV suffer because of it. Wanting to be able to to ten things, but only being able to do four because of bandwidth restrictions
I'm not talking about the "internet" I'm talking about dedicated servers in close proximity to the users that are pushing one channel/show out at a time at a given time. I'm not picturing Mission Impossible 4 hitting Netflix and 50 million people streaming the show from their servers at 50 million different times. I have heard no argument that makes this viable for Apple to "crack" the TV market. In fact the idea of a all channels streaming on-demand like Netflix makes the whole thing sound even more ludicrous than it was years ago.
Only if it's on-demand streaming or only if they don't have content servers in local areas.
It doesn't really make any difference if it is on-demand or not. When average Americans come home from work, they turn on the TV and many times it stays on until midnight. Regardless of what they are watching, ones and zeros are being pushed down the pipe and the pipe has a limitation, more so as the signal makes its way farther from the server. IP protocol requires more switches, more routers, and has more latency than traditional cable. That is the unfortunate reality of today's Internet. Last mile is the killer. Always has been since the popularization of the Internet. Fiber optics greatly increases the capacity but few homes have that option right now.
$1 per channel per month? I don't see that happening. Remember, you're going for Ã* la carte pricing which means you pay a premium. Lets say right now your cable company has 100,000 customers all subscribing for a particular base channel station. Now you want to pay $1/month for it but it's an unpopular channel and only 5,000 (5%) want to subscribe to it. That means they are only getting $5/month from the cable company. The station was also getting hefty advertisement payments but now that the subscriber rate has dropped advertisers have dropped out completely. The station has gone under because the environment that made it work is no longer supporting it. You've now lost a channel.
I bet people watch plenty of channels and don't realize how much they watch them or that a particular show is on a channel. Can an Ã* la carte method sustain the future of television networks or will umbrella companies have to consolidate shows differently in order to make a buck?
Most of the channels have ads. A channel that can't sustain themselves by their actual audience shouldn't be kept alive by subsidy from people that don't watch it. I think that's a simple bit of market economics that's been subverted by the cable bundling system. If the ads can't sustain it, then what is really going on? Besides, many of the networks you mentioned (I guess I snipped them) already transmit over the air for free.
When I had C-Band satellite, I was paying $20/mo for 20 channels, and I picked at least 10 of them Ã* la carte.
Quote:
Originally Posted by mstone
Are you suggesting that IP network traffic would not be affected by 300 million users simultaneously streaming TV content over the Internet? It is not at all like satellite. When a satellite transmits a signal there is no router, there are no hops, there are no switches, there are no firewalls. The Internet is a completely different scenario. Even a busy day on Wall Street can slow down the Internet everywhere in the country.
Is Apple really going to be a bigger draw than Netflix?
Stock exchanges have been adding their own dedicated backbones.
It doesn't really make any difference if it is on-demand or not. When average Americans come home from work, they turn on the TV and many times it stays on until midnight. Regardless of what they are watching, ones and zeros are being pushed down the pipe and the pipe has a limitation, more so as the signal makes its way farther from the server. IP protocol requires more switches, more routers, and has more latency than traditional cable. That is the unfortunate reality of today's Internet. Last mile is the killer. Always has been since the popularization of the Internet. Fiber optics greatly increases the capacity but few homes have that option right now.
it makes all the difference in the world. Once channel is one data stream until you have to split it for each customer at the last mile. This is how TV streaming works on cable. If you have an independent data stream for each user accessing the initial content because all content is on-demand then you double the data for each user accessing at a given time for the entire stream.
Yes I know but a full IPTV implementatin would choke the entire ISP network between the servers and the households. There are no other ways than multicast and distributed servers from inside the ISP network for large amount of clients.
And even with this, it tooks years and billions of dollars to Bell Canada to come up with a system that works.
And on the netflix side, the traffic is already so bad that its now #1 after peer2peer. Bell announced they will no longer nerf peer2peer downloads and are thinking of choking Netflix trafic if the loads on the network affects internet performance of other clients.
storage and computing power is cheap enough where you can put mini servers closer to the customers. you don't have to store all the content there, but stage the most used content on the local segment
Most of the channels have ads. A channel that can't sustain themselves by their actual audience shouldn't be kept alive by subsidy from people that don't watch it. I think that's a simple bit of market economics that's been subverted by the cable bundling system. If the ads can't sustain it, then what is really going on? Besides, the networks you mentioned (I guess I snipped them) already transmit over the air for free.
Channels have ads because those channels have eyeballs. If there are no longer people watching those channels then the advertisers go away.
Then you have the issue of local ads for local affiliate stations that offer local businesses ads. This space is reserved by the main operators. So how does Apple support that or does all that go away?
PS: What does any of this have to do with dedicated TV HW? WHy would Apple eschew the obvious choice of an AppleTV that can connect to any HDMI-compatible device to display an image in favour of just 3 TV sets? None of this makes no sense and each article seems to ignore every hurdle that would have to be tackled for Apple to "crack" the TV market. All I see are silly wishful dreams of "Wouldn't it be cool if Apple put the AppleTV in a TV?" and "Wouldn't it be great to only get the channels I watch without all the other crap?".
I'm not talking about the "internet" I'm talking about dedicated servers in close proximity to the users that are pushing one channel/show out at a time at a given time. I'm not picturing Mission Impossible 4 hitting Netflix and 50 million people streaming the show from their servers at 50 million different times. I have heard no argument that makes this viable for Apple to "crack" the TV market. In fact the idea of a all channels streaming on-demand like Netflix makes the whole thing sound even more ludicrous than it was years ago.
A good question. Will Apple handle everything out of their data farms? I've heard they are building another one.
The problem is, one you enter into an Internet based TV model, everything you watch becomes a VOD event. It's like everyone with a computer starts watching Netflix, every day, in every household, in every town. That's what an Internet based TV is. It's the ultimate video on demand. Even with a server dedicated to every state, it's going to kill Internet traffic for everyone using that much bandwidth in that area. Then, you have to deal with every country you want to introduce the service to.
It's like going into someone house and turning on a tap. No big deal, until you turn on every tap, and run the shower, flush the toilet. The water pressure drops to nothing. That's what this could be like.
As someone suggested earlier, this is probably something that will need to be rollled out very slowly, to see how it works.
it makes all the difference in the world. Once channel is one data stream until you have to split it for each customer at the last mile. This is how TV streaming works on cable. If you have an independent data stream for each user accessing the initial content because all content is on-demand then you double the data for each user accessing at a given time for the entire stream.
I think we have a communication problem. Assuming there is no limit to Apple's server capacity using whatever technology they have such as NC data center Akamai, etc. The local ISP network is going to require much more capacity than it currently has to provide all the data through IP protocol. It is just math. Neighborhood has 'X' homes and has a network capacity of 'Y'. If each home requires 1000 mbs to stream TV content and everyone is requesting that amount of data simultaneously, it will likely exceed 'Y' with today's bandwidth IP network limitations. I don't think you are taking into consideration all the overhead that DNS, IP, switches, routers, etc. affects IP network traffic even if it is only going a thousand meters down the cul-de-sac.
I think we have a communication problem. Assuming there is no limit to Apple's server capacity using whatever technology they have such as NC data center Akamai, etc. The local ISP network is going to require much more capacity than it currently has to provide all the data through IP protocol. It is just math. Neighborhood has 'X' homes and has a network capacity of 'Y'. If each home requires 1000 mbs to stream TV content and everyone is requesting that amount of data simultaneously, it will likely exceed 'Y' with today's bandwidth IP network limitations. I don't think you are taking into consideration all the overhead that DNS, IP, switches, routers, etc. affects IP network traffic even if it is only going a thousand meters down the cul-de-sac.
On the local end that could happen but how much bandwidth are we talking about since we're only talking one channel per TV that would be streaming and the cable companies can handle that already?
This could be mitigated if Apple got with ISPs to allow last mile broadcasts but they would have to work with them which is something I think they have to do anyway, not work around or against them as has been stated here.
$1 per channel per month? I don't see that happening. Remember, you're going for Ã* la carte pricing which means you pay a premium. Lets say right now your cable company has 100,000 customers all subscribing for a particular base channel station. Now you want to pay $1/month for it but it's an unpopular channel and only 5,000 (5%) want to subscribe to it. That means they are only getting $5/month from the cable company. The station was also getting hefty advertisement payments but now that the subscriber rate has dropped advertisers have dropped out completely. The station has gone under because the environment that made it work is no longer supporting it. You've now lost a channel.
I bet people watch plenty of channels and don't realize how much they watch them or that a particular show is on a channel. Can an Ã* la carte method sustain the future of television networks or will umbrella companies have to consolidate shows differently in order to make a buck? Will we see CBS split NCIS an NCIS: LA — two shows with high ratings — to different channels to maximize the number of stations you rent per month? Will they even go so far as to move a popular show across multiple channels week to week to maximize their return?
What people are wanting is to destroy something that will make it more expensive for customers unless you watch very little TV. Only those that want some select show or two on some cable network could possibly benefit from getting rid of the cable TV model.
Then you have consider how this will affect your cable interest costs. If your cable company loses 1/2 of it 100,000 TV paying customers they still have to pay all the networks the same amount of money. So do they double the costs of their loyal TV viewing customers? Not without losing more TV viewing customers that way. They have to jack the prices on your internet service, they now throttle users speeds, they now put in upper level caps. You're also using more internet than before because that is where your content is coming from. Can you blame them for trying to protect their business? I can't.
Or, the simple solution to all of that is supply and demand. If you lose viewers because they don't want to pay for all the channels they aren't watching, then maybe the cable carrier shouldn't be carrying the channel. You can say cable companies increase their rates to cover costs, I'd say they would decrease their costs to get in line with the new competition...on demand channels.
I think legacy cable carriers would offer an a la carte TV product like Apple supposedly will, or they will cut out all the fluff channels that drive the cost up but no one watches.
On the local end that could happen but how much bandwidth are we talking about since we're only talking one channel per TV that would be streaming and the cable companies can handle that already?
The difference is that the cable companies are not currently sending the data over IP protocol which is much more bandwidth intensive. Some homes will have several TVs on at the same time which might even exceed the bandwidth of their own wifi router in their house. Then multiply that scenario 1000 fold.
A perfect example of this is let's say I have a monster server with gigabit ethernet. I could handle a 100,000 full speed connections on my local network without breaking a sweat. But as soon as I try to deliver that data to 100,000 people on the Internet, the throughput drops by a significant factor because the Internet users cannot receive the data as fast as I can serve it due to their own bandwidth limitations.
I only watch 4 channels -- CNBC and 3 Chinese Channels. It would save me $60 of cable fees if going internet only; or, I will take $4.99 per channel. Still save me $40, if I go by per channels only. Content providers/channel can set their own subscription price and Apple gets a 30% cut!
storage and computing power is cheap enough where you can put mini servers closer to the customers. you don't have to store all the content there, but stage the most used content on the local segment
Do you really think ISP are going to help Apple by doing that? And you cant do this with live feeds, you need the real IPTV architecture for this.
Apple must leave live content distribution to ISP and stay with fixed content streaming. But even that will choke the networks. imo ISP are just going to nerf anything Apple just like they do with peer2peer and kill Apple from the start.
imo Apple is not dumb enough to come up with something like this. A TV needs to be open and allow cable/Tel to deliver either real IPTV or QAM broadcast right into it using apps. It must be deploy with full cable/tel support or its going to fail from the start.
This is one case where a close down ecosystem just wont work.
On the local end that could happen but how much bandwidth are we talking about since we're only talking one channel per TV that would be streaming and the cable companies can handle that already?
This could be mitigated if Apple got with ISPs to allow last mile broadcasts but they would have to work with them which is something I think they have to do anyway, not work around or against them as has been stated here.
The ISPs would never go for that. They would protect their turf to the death.
Just using random numbers, right now the average house probably has five intent capable devices, (not including game boxess like PS3), and three TVs. Currently, the three TVs are using cable to watch TV. We will say they are using 1 mbs to do their thing. That's 3 mbs.
Now, we will say 5 mbs for each computer device. That's 25 mbs. We will add one game console, for another 5 mbs. So, currently, that's a total of 33 mbs of total bandwidth this house uses at maximum consumption, if they are all being used at the same time.
Now, let's go to an all Internet TV model. We still have our computers and PS3 consuming 30 mbs of data. But, now our TVs consume, we will say 7 mbs, because they have now become computers, so that's now a consumption rate of 21 mbs. So now just the Internet TVs alone are now almost consuming the equivalent of the entire household bandwidth! The new household now consumes 51 mbs of bandwidth. This household has almost doubled it's bandwidth by adopting Internet TV. Now, expand that to an entire ISP, and you can see where this is going. Then to an entire country, continent, then, eventually, to the net itself.....
On the local end that could happen but how much bandwidth are we talking about since we're only talking one channel per TV that would be streaming and the cable companies can handle that already?
This could be mitigated if Apple got with ISPs to allow last mile broadcasts but they would have to work with them which is something I think they have to do anyway, not work around or against them as has been stated here.
The way IPTV works it not just 1 channel per TV. They need a compress feed with all channels. This feed is used for PiP in the guide and for the first 8 secs when you choose a channel. When tuning in, you see a broadcast compress feed for the first seconds while in the background the set-top box setup the stream. After a few secs you start receiveng the high quality stream.
In the first month on my Bell Fibe subscription, you could see the transition between the broadcast and the stream feed after exactly 8 secs. Sometimes the stream was not ready even after 8 secs and the channel would then freeze. Thankfully they fixed that now.
Now, let's go to an all Internet TV model. We still have our computers and PS3 consuming 30 mbs of data. But, now our TVs consume, we will say 7 mbs, because they have now become computers, so that's now a consumption rate of 21 mbs. So now just the Internet TVs alone are now almost consuming the equivalent of the entire household bandwidth! The new household now consumes 51 mbs of bandwidth. This household has almost doubled it's bandwidth by adopting Internet TV. Now, expand that to an entire ISP, and you can see where this is going. Then to an entire country, continent, then, eventually, to the net itself.....
I can give real numbers from a real IPTV setup. Bell needs 30 mbps for 3 HD and 4 SD and the live compress broadcast feed. My current capacity is 50 mbps that leaves 20 mbps max for my internet. The Bell modem can go up to 100 mbps but the cables between my house and the fiber optic node cant support more than 50 mbps. Fiber optic cant be more than 1000 feet away.
That 50 mbps is direct, its not shared. Once on the node I enter the fiber network. So I always have 20 mbps no matter what my neighboors are doing. Bell in investing billions of dollars to setup there network for IPTV and so far they only covered Montreal and Toronto. This is only doable in high population area.
I know the article mentioned IP delivery, but what if the data stream came down from a satellite, like Dish or DirecTV? All the channels would be in the stream and the "code" you entered into iTunes would let you watch what you paid for, just like satellite radio or that card in a satellite TV box.
I don't how many people would staple a dish to their houses, but I think this "stream" would be delivered everywhere much more easily. Mux it all together in one place, shoot it into the sky, and bounce it back to receivers.
If Apple depends on the cable company to deliver shows, the cable co's are gonna go nuts and increase rates on data plans.
Comments
Regardless of where the bottleneck occurs, in the end, user experience suffers and that is something that the average TV watcher is not going to understand. They will just assume Apple's TV sucks because they can't watch their favorite show without interruption or picture quality issues. Even if it is the ISP's problem, it affects Apples reputation. I think Apple will have to measure the roll out as they have done with other services simply to maintain quality and then slowly launch additional channels as the infrastructure allows.
Only if it's on-demand streaming or only if they don't have content servers in local areas. The first doesn't seem like Jobs "cracking" the TV market and the latter simply doesn't make any sense unless you neither sell nor support shows in those areas.
If you have local servers and the streaming is NOT on-demand, like like with cable/sat then your "hops" are very limited from the local server to the user. This currently isn't much (or any) of an issue for cable companies today.
Technically downstream can have bandwidth issues on digital coax. I first saw it with the movie Gladiator during an action scene with a lot of extra data. It didn't just affect the channel it was on but also the frequency it was sharing with another channel. This was all last mile stuff.
Touchè. But my watching a TV show here in Vancouver on my cable provider won't affect a users net speed in Florida, which it could if I was watching something on an Internet based service.
If we're talking about all channels being streamed on-demand, but I can only see this being done in the same way cable/sat is currently done, by offering most content at the same time across all distributors. That means that the content is fairly local. Likely using Akamai, as previously noted. Last mile could suffer if too many people are accessing that local server for a specific channel but I wouldn't expect anything close to what can happen from too many people accessing a website at once.
Basically what I'm saying is that the Internet is getting crowded, and is only going to get more so without some infrastructure improvements. I would hate to see an Apple TV suffer because of it. Wanting to be able to to ten things, but only being able to do four because of bandwidth restrictions
Regardless of where the bottleneck occurs, in the end, user experience suffers and that is something that the average TV watcher is not going to understand. They will just assume Apple's TV sucks because they can't watch their favorite show without interruption or picture quality issues. Even if it is the ISP's problem, it affects Apples reputation. I think Apple will have to measure the roll out as they have done with other services simply to maintain quality and then slowly launch additional channels as the infrastructure allows.
I agree.
But Apple do have some options. Most streaming services shove out a bitstream of pre-encoded data with a fixed bitrate.
Either the connection is adequate - leaving users happy. Or the connection is inadequate and we get constant stalling and buffering.
If Apple are doing this the way I expect, they will be building and compressing a custom bitstream for each customer. They can raise and drop the compression ratio according to what the client device is telling them.
If they do this, the Apple TV picture will degrade - but it won't drop out.
C.
Basically what I'm saying is that the Internet is getting crowded, and is only going to get more so without some infrastructure improvements. I would hate to see an Apple TV suffer because of it. Wanting to be able to to ten things, but only being able to do four because of bandwidth restrictions
I'm not talking about the "internet" I'm talking about dedicated servers in close proximity to the users that are pushing one channel/show out at a time at a given time. I'm not picturing Mission Impossible 4 hitting Netflix and 50 million people streaming the show from their servers at 50 million different times. I have heard no argument that makes this viable for Apple to "crack" the TV market. In fact the idea of a all channels streaming on-demand like Netflix makes the whole thing sound even more ludicrous than it was years ago.
Only if it's on-demand streaming or only if they don't have content servers in local areas.
It doesn't really make any difference if it is on-demand or not. When average Americans come home from work, they turn on the TV and many times it stays on until midnight. Regardless of what they are watching, ones and zeros are being pushed down the pipe and the pipe has a limitation, more so as the signal makes its way farther from the server. IP protocol requires more switches, more routers, and has more latency than traditional cable. That is the unfortunate reality of today's Internet. Last mile is the killer. Always has been since the popularization of the Internet. Fiber optics greatly increases the capacity but few homes have that option right now.
$1 per channel per month? I don't see that happening. Remember, you're going for Ã* la carte pricing which means you pay a premium. Lets say right now your cable company has 100,000 customers all subscribing for a particular base channel station. Now you want to pay $1/month for it but it's an unpopular channel and only 5,000 (5%) want to subscribe to it. That means they are only getting $5/month from the cable company. The station was also getting hefty advertisement payments but now that the subscriber rate has dropped advertisers have dropped out completely. The station has gone under because the environment that made it work is no longer supporting it. You've now lost a channel.
I bet people watch plenty of channels and don't realize how much they watch them or that a particular show is on a channel. Can an Ã* la carte method sustain the future of television networks or will umbrella companies have to consolidate shows differently in order to make a buck?
Most of the channels have ads. A channel that can't sustain themselves by their actual audience shouldn't be kept alive by subsidy from people that don't watch it. I think that's a simple bit of market economics that's been subverted by the cable bundling system. If the ads can't sustain it, then what is really going on? Besides, many of the networks you mentioned (I guess I snipped them) already transmit over the air for free.
When I had C-Band satellite, I was paying $20/mo for 20 channels, and I picked at least 10 of them Ã* la carte.
Are you suggesting that IP network traffic would not be affected by 300 million users simultaneously streaming TV content over the Internet? It is not at all like satellite. When a satellite transmits a signal there is no router, there are no hops, there are no switches, there are no firewalls. The Internet is a completely different scenario. Even a busy day on Wall Street can slow down the Internet everywhere in the country.
Is Apple really going to be a bigger draw than Netflix?
Stock exchanges have been adding their own dedicated backbones.
It doesn't really make any difference if it is on-demand or not. When average Americans come home from work, they turn on the TV and many times it stays on until midnight. Regardless of what they are watching, ones and zeros are being pushed down the pipe and the pipe has a limitation, more so as the signal makes its way farther from the server. IP protocol requires more switches, more routers, and has more latency than traditional cable. That is the unfortunate reality of today's Internet. Last mile is the killer. Always has been since the popularization of the Internet. Fiber optics greatly increases the capacity but few homes have that option right now.
it makes all the difference in the world. Once channel is one data stream until you have to split it for each customer at the last mile. This is how TV streaming works on cable. If you have an independent data stream for each user accessing the initial content because all content is on-demand then you double the data for each user accessing at a given time for the entire stream.
Yes I know but a full IPTV implementatin would choke the entire ISP network between the servers and the households. There are no other ways than multicast and distributed servers from inside the ISP network for large amount of clients.
And even with this, it tooks years and billions of dollars to Bell Canada to come up with a system that works.
And on the netflix side, the traffic is already so bad that its now #1 after peer2peer. Bell announced they will no longer nerf peer2peer downloads and are thinking of choking Netflix trafic if the loads on the network affects internet performance of other clients.
storage and computing power is cheap enough where you can put mini servers closer to the customers. you don't have to store all the content there, but stage the most used content on the local segment
Most of the channels have ads. A channel that can't sustain themselves by their actual audience shouldn't be kept alive by subsidy from people that don't watch it. I think that's a simple bit of market economics that's been subverted by the cable bundling system. If the ads can't sustain it, then what is really going on? Besides, the networks you mentioned (I guess I snipped them) already transmit over the air for free.
Channels have ads because those channels have eyeballs. If there are no longer people watching those channels then the advertisers go away.
Then you have the issue of local ads for local affiliate stations that offer local businesses ads. This space is reserved by the main operators. So how does Apple support that or does all that go away?
PS: What does any of this have to do with dedicated TV HW? WHy would Apple eschew the obvious choice of an AppleTV that can connect to any HDMI-compatible device to display an image in favour of just 3 TV sets? None of this makes no sense and each article seems to ignore every hurdle that would have to be tackled for Apple to "crack" the TV market. All I see are silly wishful dreams of "Wouldn't it be cool if Apple put the AppleTV in a TV?" and "Wouldn't it be great to only get the channels I watch without all the other crap?".
I'm not talking about the "internet" I'm talking about dedicated servers in close proximity to the users that are pushing one channel/show out at a time at a given time. I'm not picturing Mission Impossible 4 hitting Netflix and 50 million people streaming the show from their servers at 50 million different times. I have heard no argument that makes this viable for Apple to "crack" the TV market. In fact the idea of a all channels streaming on-demand like Netflix makes the whole thing sound even more ludicrous than it was years ago.
A good question. Will Apple handle everything out of their data farms? I've heard they are building another one.
The problem is, one you enter into an Internet based TV model, everything you watch becomes a VOD event. It's like everyone with a computer starts watching Netflix, every day, in every household, in every town. That's what an Internet based TV is. It's the ultimate video on demand. Even with a server dedicated to every state, it's going to kill Internet traffic for everyone using that much bandwidth in that area. Then, you have to deal with every country you want to introduce the service to.
It's like going into someone house and turning on a tap. No big deal, until you turn on every tap, and run the shower, flush the toilet. The water pressure drops to nothing. That's what this could be like.
As someone suggested earlier, this is probably something that will need to be rollled out very slowly, to see how it works.
it makes all the difference in the world. Once channel is one data stream until you have to split it for each customer at the last mile. This is how TV streaming works on cable. If you have an independent data stream for each user accessing the initial content because all content is on-demand then you double the data for each user accessing at a given time for the entire stream.
I think we have a communication problem. Assuming there is no limit to Apple's server capacity using whatever technology they have such as NC data center Akamai, etc. The local ISP network is going to require much more capacity than it currently has to provide all the data through IP protocol. It is just math. Neighborhood has 'X' homes and has a network capacity of 'Y'. If each home requires 1000 mbs to stream TV content and everyone is requesting that amount of data simultaneously, it will likely exceed 'Y' with today's bandwidth IP network limitations. I don't think you are taking into consideration all the overhead that DNS, IP, switches, routers, etc. affects IP network traffic even if it is only going a thousand meters down the cul-de-sac.
I think we have a communication problem. Assuming there is no limit to Apple's server capacity using whatever technology they have such as NC data center Akamai, etc. The local ISP network is going to require much more capacity than it currently has to provide all the data through IP protocol. It is just math. Neighborhood has 'X' homes and has a network capacity of 'Y'. If each home requires 1000 mbs to stream TV content and everyone is requesting that amount of data simultaneously, it will likely exceed 'Y' with today's bandwidth IP network limitations. I don't think you are taking into consideration all the overhead that DNS, IP, switches, routers, etc. affects IP network traffic even if it is only going a thousand meters down the cul-de-sac.
On the local end that could happen but how much bandwidth are we talking about since we're only talking one channel per TV that would be streaming and the cable companies can handle that already?
This could be mitigated if Apple got with ISPs to allow last mile broadcasts but they would have to work with them which is something I think they have to do anyway, not work around or against them as has been stated here.
$1 per channel per month? I don't see that happening. Remember, you're going for Ã* la carte pricing which means you pay a premium. Lets say right now your cable company has 100,000 customers all subscribing for a particular base channel station. Now you want to pay $1/month for it but it's an unpopular channel and only 5,000 (5%) want to subscribe to it. That means they are only getting $5/month from the cable company. The station was also getting hefty advertisement payments but now that the subscriber rate has dropped advertisers have dropped out completely. The station has gone under because the environment that made it work is no longer supporting it. You've now lost a channel.
I bet people watch plenty of channels and don't realize how much they watch them or that a particular show is on a channel. Can an Ã* la carte method sustain the future of television networks or will umbrella companies have to consolidate shows differently in order to make a buck? Will we see CBS split NCIS an NCIS: LA — two shows with high ratings — to different channels to maximize the number of stations you rent per month? Will they even go so far as to move a popular show across multiple channels week to week to maximize their return?
What people are wanting is to destroy something that will make it more expensive for customers unless you watch very little TV. Only those that want some select show or two on some cable network could possibly benefit from getting rid of the cable TV model.
Then you have consider how this will affect your cable interest costs. If your cable company loses 1/2 of it 100,000 TV paying customers they still have to pay all the networks the same amount of money. So do they double the costs of their loyal TV viewing customers? Not without losing more TV viewing customers that way. They have to jack the prices on your internet service, they now throttle users speeds, they now put in upper level caps. You're also using more internet than before because that is where your content is coming from. Can you blame them for trying to protect their business? I can't.
Or, the simple solution to all of that is supply and demand. If you lose viewers because they don't want to pay for all the channels they aren't watching, then maybe the cable carrier shouldn't be carrying the channel. You can say cable companies increase their rates to cover costs, I'd say they would decrease their costs to get in line with the new competition...on demand channels.
I think legacy cable carriers would offer an a la carte TV product like Apple supposedly will, or they will cut out all the fluff channels that drive the cost up but no one watches.
On the local end that could happen but how much bandwidth are we talking about since we're only talking one channel per TV that would be streaming and the cable companies can handle that already?
The difference is that the cable companies are not currently sending the data over IP protocol which is much more bandwidth intensive. Some homes will have several TVs on at the same time which might even exceed the bandwidth of their own wifi router in their house. Then multiply that scenario 1000 fold.
A perfect example of this is let's say I have a monster server with gigabit ethernet. I could handle a 100,000 full speed connections on my local network without breaking a sweat. But as soon as I try to deliver that data to 100,000 people on the Internet, the throughput drops by a significant factor because the Internet users cannot receive the data as fast as I can serve it due to their own bandwidth limitations.
storage and computing power is cheap enough where you can put mini servers closer to the customers. you don't have to store all the content there, but stage the most used content on the local segment
Do you really think ISP are going to help Apple by doing that? And you cant do this with live feeds, you need the real IPTV architecture for this.
Apple must leave live content distribution to ISP and stay with fixed content streaming. But even that will choke the networks. imo ISP are just going to nerf anything Apple just like they do with peer2peer and kill Apple from the start.
imo Apple is not dumb enough to come up with something like this. A TV needs to be open and allow cable/Tel to deliver either real IPTV or QAM broadcast right into it using apps. It must be deploy with full cable/tel support or its going to fail from the start.
This is one case where a close down ecosystem just wont work.
On the local end that could happen but how much bandwidth are we talking about since we're only talking one channel per TV that would be streaming and the cable companies can handle that already?
This could be mitigated if Apple got with ISPs to allow last mile broadcasts but they would have to work with them which is something I think they have to do anyway, not work around or against them as has been stated here.
The ISPs would never go for that. They would protect their turf to the death.
Just using random numbers, right now the average house probably has five intent capable devices, (not including game boxess like PS3), and three TVs. Currently, the three TVs are using cable to watch TV. We will say they are using 1 mbs to do their thing. That's 3 mbs.
Now, we will say 5 mbs for each computer device. That's 25 mbs. We will add one game console, for another 5 mbs. So, currently, that's a total of 33 mbs of total bandwidth this house uses at maximum consumption, if they are all being used at the same time.
Now, let's go to an all Internet TV model. We still have our computers and PS3 consuming 30 mbs of data. But, now our TVs consume, we will say 7 mbs, because they have now become computers, so that's now a consumption rate of 21 mbs. So now just the Internet TVs alone are now almost consuming the equivalent of the entire household bandwidth! The new household now consumes 51 mbs of bandwidth. This household has almost doubled it's bandwidth by adopting Internet TV. Now, expand that to an entire ISP, and you can see where this is going. Then to an entire country, continent, then, eventually, to the net itself.....
On the local end that could happen but how much bandwidth are we talking about since we're only talking one channel per TV that would be streaming and the cable companies can handle that already?
This could be mitigated if Apple got with ISPs to allow last mile broadcasts but they would have to work with them which is something I think they have to do anyway, not work around or against them as has been stated here.
The way IPTV works it not just 1 channel per TV. They need a compress feed with all channels. This feed is used for PiP in the guide and for the first 8 secs when you choose a channel. When tuning in, you see a broadcast compress feed for the first seconds while in the background the set-top box setup the stream. After a few secs you start receiveng the high quality stream.
In the first month on my Bell Fibe subscription, you could see the transition between the broadcast and the stream feed after exactly 8 secs. Sometimes the stream was not ready even after 8 secs and the channel would then freeze. Thankfully they fixed that now.
Now, let's go to an all Internet TV model. We still have our computers and PS3 consuming 30 mbs of data. But, now our TVs consume, we will say 7 mbs, because they have now become computers, so that's now a consumption rate of 21 mbs. So now just the Internet TVs alone are now almost consuming the equivalent of the entire household bandwidth! The new household now consumes 51 mbs of bandwidth. This household has almost doubled it's bandwidth by adopting Internet TV. Now, expand that to an entire ISP, and you can see where this is going. Then to an entire country, continent, then, eventually, to the net itself.....
I can give real numbers from a real IPTV setup. Bell needs 30 mbps for 3 HD and 4 SD and the live compress broadcast feed. My current capacity is 50 mbps that leaves 20 mbps max for my internet. The Bell modem can go up to 100 mbps but the cables between my house and the fiber optic node cant support more than 50 mbps. Fiber optic cant be more than 1000 feet away.
That 50 mbps is direct, its not shared. Once on the node I enter the fiber network. So I always have 20 mbps no matter what my neighboors are doing. Bell in investing billions of dollars to setup there network for IPTV and so far they only covered Montreal and Toronto. This is only doable in high population area.
I don't how many people would staple a dish to their houses, but I think this "stream" would be delivered everywhere much more easily. Mux it all together in one place, shoot it into the sky, and bounce it back to receivers.
If Apple depends on the cable company to deliver shows, the cable co's are gonna go nuts and increase rates on data plans.
Just a thought.