 espaethDigital PlumberPremium,MVM
join:2001-04-21
Minneapolis, MN
kudos:2
 ·Clear Wireless
| reply to Matt
Re: And why should they lose money so video providers get rich ? said by Matt:The HD content providers are already paying for massive pipes to the internet and if they don't have a direct peering agreement with AT&T or Comcast or Time Warner, someone either pays for the peering or it's a mutually beneficial agreement. I'll say it again, THEY ARE ALREADY GETTING PAID for the traffic. They are trying to double dip and also snuff out any competition to their video business. Hell, getting paid twice while eliminating your competition so you don't have to compete sounds pretty damn good to me. The cost of building out consolidated capacity in a data center or two isn't even remotely close to the cost of building out a DOCSIS plant. You'll spend several orders of magnitude more for DOCSIS delivery to the home, and still end up with less overall bandwidth than is possible at a hosting / content data center.
By stating this I'm not suggesting that content providers should be paying access providers to offset these costs. I do, however, believe that the consumer will (and probably should) be billed a higher level because the delivery method is so ungodly inefficient. There's no way you can take a broadcast medium like cable or satellite TV, convert it to unicast streams where the infrastructure has to scale linearly with every single new viewer, and expect that somehow at full scale this is going to be cheaper.
The consumer costs are going to be higher because of the vast inefficiency of delivery in combination with sharply higher network buildout expenses at the edge. This says nothing of the fact that if people are going to be doing wholesale conversion from broadcast to IPTV, we're only going to get a few percent of the existing viewership converted before we'll run out of edge capacity due to the limits of current broadband access technology. |
|
|
|
| said by espaeth:By stating this I'm not suggesting that content providers should be paying access providers to offset these costs. I do, however, believe that the consumer will (and probably should) be billed a higher level because the delivery method is so ungodly inefficient. There's no way you can take a broadcast medium like cable or satellite TV, convert it to unicast streams where the infrastructure has to scale linearly with every single new viewer, and expect that somehow at full scale this is going to be cheaper.The consumer costs are going to be higher because of the vast inefficiency of delivery in combination with sharply higher network buildout expenses at the edge. This says nothing of the fact that if people are going to be doing wholesale conversion from broadcast to IPTV, we're only going to get a few percent of the existing viewership converted before we'll run out of edge capacity due to the limits of current broadband access technology. Very well put. And it is this unfortunate FACT that many here REFUSE to accept. And the loss of TV viewers will reduce revenues while costs will not drop commensurately. Then, prices will increase drastically for HSI unless caps and overage charges are in place.
--
My BLOG .. .. Internet News .. .. My Web Page
Ask yourself one question: 'Do I feel lucky?' Well, do ya punk? |
|
| reply to espaeth
said by espaeth:There's no way you can take a broadcast medium like cable or satellite TV, convert it to unicast streams where the infrastructure has to scale linearly with every single new viewer, and expect that somehow at full scale this is going to be cheaper. This is a technical problem, that requires a technical solution, not a business solution.
It is time for providers and the network engineering community to seriously sit down and work out the issues with multicasting and getting it up and running. This could address some of the issues with getting HD video on the network.
Additional solutions are working with content providers to put the content close to the edge via the increase usage of caches.
Providers should stick to being IP access providers and do it well. Content providers should stick to being content providers and doing it well. When companies try to branch out and do everything, they loose focus and do everything half-ass.
--
---
Drilling for more oil is akin to giving a methhead the keys to the meth lab. |
|
espaethDigital PlumberPremium,MVM
join:2001-04-21
Minneapolis, MN
kudos:2 Reviews:
·Clear Wireless
| said by NetAdmin1:It is time for providers and the network engineering community to seriously sit down and work out the issues with multicasting and getting it up and running. This could address some of the issues with getting HD video on the network. I believe we call this "reinventing the wheel." After you invest all of the time and money into distributing the content via multicast, you still have the same real time video streams that you had when you started with traditional broadcast options.
said by NetAdmin1:Additional solutions are working with content providers to put the content close to the edge via the increase usage of caches. How about caching the content really close to the edge. Let's have boxes with big-ass hard drives that can capture content from a broadcast stream so the end-user can watch it whenever they want. We can call these things Digital Media Capture Appliances.
On second thought, we might need a new acronym. |
|
| said by espaeth:I believe we call this "reinventing the wheel." After you invest all of the time and money into distributing the content via multicast, you still have the same real time video streams that you had when you started with traditional broadcast options. Not even close to reinventing the wheel.
Multicast deals with part of the problem, live streams such as internet radio and live video.
How about caching the content really close to the edge. Let's have boxes with big-ass hard drives that can capture content from a broadcast stream so the end-user can watch it whenever they want. We can call these things Digital Media Capture Appliances.
On second thought, we might need a new acronym. So, instead of smart-ass answers, how about using that energy toward coming up with an engineering solution? What's your solution?
--
---
Drilling for more oil is akin to giving a methhead the keys to the meth lab. |
|
espaethDigital PlumberPremium,MVM
join:2001-04-21
Minneapolis, MN
kudos:2 Reviews:
·Clear Wireless
| said by NetAdmin1:Not even close to reinventing the wheel. Multicast deals with part of the problem, live streams such as internet radio and live video. We already have a system in place to stream hundreds of live TV feeds; they're called channels.
said by NetAdmin1:So, instead of smart-ass answers, how about using that energy toward coming up with an engineering solution? What's your solution? Build on existing DVR technology with network augmentation, not replacement.
1) Capture all primary content on a DVR that grabs content from satellite, cable, or broadcast media sources.
2) Have a request system such that if a sufficient number of people request retrieval of content, it is streamed down on a channel and populates the hard drive of the DVR. The request system would mostly be used to recall back episodes of shows, missed episodes, etc.
3) For anything that isn't popularly requested, download that content via the network.
Network access can be an incredible tool for improving media consumption, but only if we're smart about it. The key sell of any network access is it provides "any-to-any" connectivity. Technologies like VoIP are a natural fit for network access because it's still based on random point-to-point conversations.
The overwhelming majority of our TV viewing today is based around consistent time schedules: watching live sporting events, live news programs, or regular recurring entertainment programs. These are all "one-to-many" flows that make for an awkward fit in IP networks. Sure, you can make it work, but it's the least efficient method of accomplishing the task. In terms of solving the "I think I'm paying too much for TV, so Internet delivery will be cheaper" problem... for the first few folks it indeed might be cheaper, but it's not a sustainable solution long term. The cost model turns upside down with even minimal loading of such a solution. |
|
Lazlow
join:2006-08-07
Saint Louis, MO | espaeth
"The overwhelming majority of our TV viewing today"
I think that is radically changing. The only reason cable (for lack of a better term) is providing DVRs is that for years people (a small majority admittedly) were using PVRs. Before that they were using VCRs. People want to watch things when they want to watch them, not whenever it is scheduled to show. How that content gets delivered is in question. Until cable offers to play what the customer wants, when they want it, at a reasonable price(say $10 for a season of show X) there will be some form of downloading. Maybe dropping the number of channels(in the current sense) down to say 20 and moving the rest of the channels over to a system similar to pay per view(except by season). Think of it as true ala carte(by the show). If cable cannot accomplish this then I think a large majority will drop cable TV and put that money into a larger capped HSI tier. Spending $100 on HSI(say 5 meg) without cap is going to(if it is not already) suit many peoples lifestyle much better than spending $45 on HSI(capped) and $55 on cable (whatever the actual numbers are) and having to watch only when something is being shown. |
|
espaethDigital PlumberPremium,MVM
join:2001-04-21
Minneapolis, MN
kudos:2 Reviews:
·Clear Wireless
| said by Lazlow:"The overwhelming majority of our TV viewing today" I think that is radically changing. I think the perception of people on this site is radically changing, but I don't think the habits of the average TV viewer is really changing all that much. The population at large is slow to adopt digital delivery.
Take digital music; you can't watch anything on TV without seeing an ipod commercial now. Retailers like Target are stocking all kinds of mp3 players, some for as low as $20. Despite the fact that digital delivery is everywhere for music content (including iTunes for the multitude of ipod owners), 85% of music sales in the US is still in the form of physical compact discs. Source: »online.wsj.com/public/article/SB···526.html
There also hasn't been a recorded major shift in online video consumption according to statistics from ComScore published in September. This report shows:
* 75 percent of the total U.S. Internet audience viewed online video.
* The average online video viewer watched 235 minutes of video.
* 91 million viewers watched 5 billion videos on YouTube.com (54.8 videos per viewer).
* 51.4 million viewers watched 400 million videos on MySpace.com (7.8 videos per viewer).
* The duration of the average online video was 2.9 minutes.
So while the average online user is watching just over 4 hours of video per month, the average traditional television viewer is watching approximately 5 hours per day according to Nielson Media Research. Also, watching a bunch of short (2.9 min average) clips isn't even remotely close to TV viewing habits.
If you read the opinions on this site it sounds like everybody is getting Roku boxes or downloading movies on their Xbox360/PS3 or watching hours upon hours of Hulu. People love to argue that changes in broadband usage policies aren't about infrastructure limits and are simply a way for broadband providers to limit video competition. The problem with this argument is, the stats show that "competition" is barely a rounding error in relation to their core video viewership.
said by Lazlow:Spending $100 on HSI(say 5 meg) without cap is going to(if it is not already) suit many peoples lifestyle much better than spending $45 on HSI(capped) and $55 on cable (whatever the actual numbers are) and having to watch only when something is being shown. I agree the economics could maybe play out on this, but the problems of delivery and the amount of work required to make this happen would need to be solved. Right now I set my Dish DVR to record shows I want to watch (including local news / local sports), and I can watch them with a click of a remote anytime in 1080i on a nice 58" screen from the comfort of my couch. In the past I've had to track down shows that didn't get recorded (due to storms or power outages) from the net, get them converted into something I could stream to my PS3, and watch them that way -- but that's a lot of work. Unless an online delivered solution offers the same quality and convenience of existing DVR/traditional delivery options, it's not going to catch on en-mass as a replacement for standard TV consumption. |
|
Lazlow
join:2006-08-07
Saint Louis, MO | Espaeth
On the digital music front; I think people are still concerned that they will either have to store (long term) it locally or risk the initial provider no longer providing access(which I recall happened earlier this year). So a lot of people bite the bullet and buy the CD, and then rip it to their mp3 player. There is also the issue of most mp3 providers not using a high enough bit rate.
As to the current video consumption; I think you missed the point. Compare all those numbers you provided (for online consumption) to what they were two years ago. Now project that rate of growth forward two years. I think the cable companies have and that is why this change (caps).
The fact that you use a DVR and a PS3 just proves how close you already are. The difference between running a DVR hooked to your big screen and a PC hooked to your big screen is a very small jump. Once you have the PC tied directly to the TV them the format issues drop out of the equation. Most popular (and a lot of not so popular) TV shows are currently available online in Xvid format that is somewhere between 480i and 720p in quality (roughly 350meg per one hour episode). A lot of content is/has been available in 720p(roughly 1.2gig per one hour episode). In the last six months even 1080p(not i) has begun to show up for some media. Personally, while I can tell the difference(on a 40 inch screen) between the typical Xvid format and the 720p stuff, it is not significant enough (for a TV show) to warrant the 3-4 times increase in file size. |
|
| reply to espaeth
said by espaeth:We already have a system in place to stream hundreds of live TV feeds; they're called channels. Yeah, but "channels" don't translate over very well in IP networks. The closest thing you can get to channels with IP, while still using bandwidth efficiently, is to use multicast (why does the site think that word is misspelled).
Even in your example, multicast can reduce network usage. Imagine that a good 10 or 20% of the users on one node or DSLAM want to have the same episode on their DVR. When the scheduled distribution of that show occurs, it would be done via multicast to all of the DVRs. So instead of multiple data streams, you have only one data stream.
--
---
Drilling for more oil is akin to giving a methhead the keys to the meth lab. |
|
espaethDigital PlumberPremium,MVM
join:2001-04-21
Minneapolis, MN
kudos:2 Reviews:
·Clear Wireless
| reply to Lazlow
said by Lazlow:As to the current video consumption; I think you missed the point. Compare all those numbers you provided (for online consumption) to what they were two years ago. Now project that rate of growth forward two years. I think the cable companies have and that is why this change (caps). 63 million unique viewers for YouTube in 2006.Source: »www.comscore.com/press/release.a···ess=1023
91 million uniques in September -- it hasn't even doubled.
said by Lazlow:The fact that you use a DVR and a PS3 just proves how close you already are. The difference between running a DVR hooked to your big screen and a PC hooked to your big screen is a very small jump. It's easy to think of it as a small jump, you're just replacing one box with a video output with another, right? The problem is, the UI for PC video sucks from a couch viewing standpoint. The only widely available commercial remotes are for Microsoft Media Center, you have to fight video card driver issues and ensure that you get a video card that not only supports HD copy protection over HDMI but also does hardware VC1/H264 decoding so it can keep up with 1080p output. Even then, you still need to jerk around with overscan settings to scale the picture to fit your TV because video cards are designed for computer monitors, not the LCD / Plasma displays that you pick up at an electronics retailer. The existing BluRay / HD-DVD software options absolutely blow, be it PowerDVD, Nero Showtime, or WinDVD. There's also the issue that most PCs are damn ugly in the scope of an entertainment center unless you spend a lot of cash for a really nice media center case. You also have the noise factor of the PC unless, again, you spend a lot of cash building a quiet system.
said by Lazlow:Most popular (and a lot of not so popular) TV shows are currently available online in Xvid format that is somewhere between 480i and 720p in quality (roughly 350meg per one hour episode). A lot of content is/has been available in 720p(roughly 1.2gig per one hour episode). In the last six months even 1080p(not i) has begun to show up for some media. Personally, while I can tell the difference(on a 40 inch screen) between the typical Xvid format and the 720p stuff, it is not significant enough (for a TV show) to warrant the 3-4 times increase in file size. That's the thing... I plunked down the cash for a 58" to watch stuff in HD. While there's a lot of content available in highly compressed xvid, or 720p mkv containers, why would I go through all the effort to track down that content when it doesn't look nearly as good as the 1080i feeds I capture I my Dish DVR via the OTA input? I highlighted each show I wanted to watch when I got it, told it to record all new episodes, and it has faithfully done just that for a couple years now. Online video isn't even close to getting to this point yet. |
|
espaethDigital PlumberPremium,MVM
join:2001-04-21
Minneapolis, MN
kudos:2 Reviews:
·Clear Wireless
| reply to NetAdmin1
said by NetAdmin1:Yeah, but "channels" don't translate over very well in IP networks. The closest thing you can get to channels with IP, while still using bandwidth efficiently, is to use multicast (why does the site think that word is misspelled). That just calls out the even bigger problem -- we have more existing infrastructure for channel capacity than we do for IP capacity. I can get a couple hundred channels via Dish or DirectTV, with at least 30-50 viewable HD channels. Multicast is a solution, but still not the solution. Just look at all the issues ATT is having with U-Verse; only being able to watch 2 channels within a house because that's all the bandwidth you have sucks. I run a single household and my Dish DVR still records 3 feeds at times (2 SAT + digital OTA) to get everything I'm interested in watching. Assuming 5mbps per HD feed, that'd still require 15 mbps of IP capacity just to watch those 3 channels.
said by NetAdmin1:]When the scheduled distribution of that show occurs, it would be done via multicast to all of the DVRs. So instead of multiple data streams, you have only one data stream. Why not do the same thing on a traditional channel by playing frequently requested programs in a loop sequence on a "request" channel like they do with PPV? The selection criteria could be: You select a program that is not live, the first thing the appliance does is check the schedule to see if it will be on anytime soon, if it is not in the schedule then it resorts to retrieval over the the network. If enough people select content for retrieval, it can be scheduled to be aired on the request channel and the updated guide information is pushed down to the appliances. With DVRs being networked, this can be implemented on existing technology with very little engineering rework. |
|
Lazlow
join:2006-08-07
Saint Louis, MO | reply to espaeth
Utube
2006: " In July 2006, YouTube served nearly 3 billion video streams worldwide, "
2008: "# 91 million viewers watched 5 billion videos on YouTube.com (54.8 videos per viewer)."
Not quite doubled in two years, and that is just utube. Was Hulu, etc around in 2006?
I guess I use MythTV, a usb remote, and HDMI(1920×1080) output from the PC(no overscan). I put together a very small PC with a 45wattX2 and gigE. My dvd player makes more noise starting up than the PC. Cost me less than $400 and I guess beauty is in the eye of the beholder. All storage is done on my house server in the basement. I run Linux so the video card copyright thing is not an issue. You do have a point that I do not currently know of a service that you can get a seasons worth of content in a one click solution. Blueray is just beyond my budget at this point, so I do not have a player. Take a look at the compression rate of your DVR(yes it does compress too) and compare it to the 5gig/episode available(again I have no idea how these two compare). Personally I have a hard time justifying watching House in Blueray(or even 720p, 1.2gig) vs Xvid(350meg/episode). Now Jeremiah Johnson (or something similar) I could see the point. |
|
