About Tom Evslin

Video Profile of Tom Evslin

Follow Tom Evslin on Twitter


Add to Technorati Favorites!
Powered by TypePad
Member since 01/2005


« Time of Day Pricing for Electricity | Main | Will the UN Save the World from Asteroids? »

Television on The Internet – How much Bandwidth is Needed? Where?

There’s no question about it, watching live video over an Internet connection is not something you want to on a dialup connection.  On the other hand, if you receive TV today either over the air (the old way), from a satellite dish, or through a cable, you already have MORE bandwidth coming into your house than you’ll need for watching TV on the Internet.  So why is everyone gonna need a fiber connection to his or her home to be able to participate in the future?  And how come there’s a bandwidth crisis coming?

Short answers: Even though we all do need a lot more bandwidth into our homes than we have on a dialup connection and we’ll soon outgrow basic DSL, we DON’T all need fiber connections.  If we all shift to watching TV on the Internet, the total bandwidth (Internet and other) required INTO our homes will decrease and the load on the Internet backbone and the regional distribution portions of the Internet will be – well – interesting.  It’s a separate subject for a separate post but we most of us need or will soon need lots more bandwidth OUT of our houses for sending stuff.

Just a quick refresher on how TV signal distribution works today.  In most places, there are a number of local channels still broadcast over the air from high towers.  Some of these include high definition programming.  All of these channels are received by your antenna (if you still have one) all the time.  All of these channels fit in the thin wire which runs from the antenna into the back of your TV.  Obviously lots of bandwidth there.

If you have cable, all 200 or so channels including the ones you don’t subscribe to are delivered to your set-top box.  Same thing if your “cable” runs to a satellite dish.  Whatever channel you’ve decided to watch (assuming you’ve subscribed to it) is delivered from the set-top box to the TV.  That (and the vested interest of your satellite or cable provider) is why you “need” one set-top box for each TV capable of watching different channels from the other TVs (unless you’re a TV hacker).

Now let’s fast forward to Internet TV.  You can watch any show you want any time you want although you still may have to pay for some of these.  How many shows are going to be watched at once in your house? Two?  Four?  Not 200, right? So you need LESS bandwidth for incoming shows than you have with cable or satellite today.  The problem – if there’s a problem – is that you need that bandwidth in your Internet connection and it may not be there today.

But wait, you say, my Internet connection already comes into my house on my cable.  And so do all 200 channels.  So I’ve already got enough bandwidth.  You’re right; you do.  It just has to be rearranged a little.  And remember, you’re not even gonna need to bring in 200 channels at once, just the ones you’re actually watching. Maybe you should get a rebate.

Point is you don’t need fiber coming into your house to get all the TV content you can watch – even high def.  Today’s coaxial cable connections have more than enough bandwidth.  Today’s DSL, however, does not although it is possible that successor technologies will be able to deliver enough bandwidth for Internet TV over telephone wires to people VERY close to the telephone office (now you know why, in general, it’s phone companies and not cable companies who are installing fiber).  Fixed wireless Internet access CAN also deliver enough Internet capacity for multiple TV shows and high definition (which needs more bandwidth) although it usually doesn’t today.

The chart below is from a Swedish study.  “Twisted pair” is what is used to deliver today’s DSL.  In all cases, they are talking about what can be achieved with the various delivery technologies – almost always more than is being done today.



Fiber is a great way to bring bandwidth into a house, bit it’s not the only way to meet today’s or even tomorrow’s need for last mile connectivity.  However, fiber to the community is becoming increasingly essential; it’s needed for many more reasons than just allowing TV to migrate to the Internet.  Although we’ve seen that Internet TV decreases the total need for bandwidth coming into the house, it will create a huge new load everywhere but in the last mile.

Think of the cable network as it enters your town but consider just the entertainment on it and forget that it’s also being used for Internet access.  No matter how many homes are connected, there only has to be bandwidth for 200 channels.  Everybody has to watch the same show exactly when it’s delivered (except for those people who use DVRs to record the shows and replay them later). I’m ignoring the pitiful amount of video on demand offered on today’s cable systems. 

But let’s fast forward to Internet TV.  Now any of us can watch whatever we want to watch whenever we want to watch it.  Now how much entertainment bandwidth do we need coming into town?  How many different shows are being watched in the whole community?  Even when two families are watching the same show, unless it’s live, they’re probably not watching it at the same time so many versions of the same show will be delivered.

More bandwidth isn’t the only way to allow more choice; I’ll talk about some others in future posts.  But more bandwidth – in most cases best delivered on fiber – must be provided between communities before full Internet TV is a reality.

Television and the Internet is the first post in this series.


TrackBack URL for this entry:

Listed below are links to weblogs that reference Television on The Internet – How much Bandwidth is Needed? Where?:

» Upstream,Downstream from Connected Home 2 Go
Infonetics is reporting that sales of cable modem termination system (CMTS) gear cleared $1 billion in 2006 for the first time ever. Certainly Motorola’s revenue in this market has grown*, but what’s more telling is the ratio of upstream ports sol... [Read More]


John Miller

Thanks for bringing this subject up for discussion. As a rural Wireless ISP, I certainly have to disagree with your comments about last-mile infrastructure. All ISPs operate on a contention basis. This is why subscriber plans are affordable. This is why we are all losing sleep at night with regard to IPTV. And this is why your service slows during prime usage hours. The number of users exceeds a resonable contention rate. Therefore resources are limited and shared. In my opinion, ISPs will be the hardest hit and that is the reason we do need fiber or coax to the premises. Streaming anything is not compatible with the present business model used by ISPs. The ones with fiber or coax have the excess capacity to make it work.

The cost of Internet from least to most expensive is content, bandwidth, transportation. While my rural perspective may not be exactly the situation that everyone experiences, it serves well to make the point of how expensive transportation is.

Let's look at the cost of Internet bandwidth and transportation. When purchasing 1 - 5 Mbps of Internet bandwidth, it costs about $281.00 to $336.00 per Mbps in our area. Even wholesale Internet bandwidth costs $147.00 - $300.00 per Mbps when it is delivered at a middle-mile providers POP (Point of Presence). If you purchase 100 Mbps of wholesale bandwidth, it goes down to about $50.00 per Mbps. Then it must be transported to the network, through the network and to the end user. Since there is not a lot of fiber available for transport in this area, that means that most often your bandwidth will be delivered using two or more PTP (Point to Point) links. A cheap link to transport the bandwidth can run from $2000.00 to $10,000.00 while a quality link can run $20,000.00. A carrier-grade high-speed licensed microwave link will cost about $40,000.00. Depending on the technology and speed or reliability needed these links can operate from 5 - 30 miles. Now let's add in the cost of a tower on each end. Since some towers may have one or two links and occasionally more, we will just figure one tower per link. Guyed towers are about $125.00 per foot and self-supporting towers can be triple that. Now a link at $30,000.00 and a tower at $25,000.00 gives us $55,000.00. We will estimate a high average of 20 miles per link. By the time you ad in the miscellaneous costs you end up at about $3,000.00 per mile. Then you have the recurring cost of about $100.00 - $500.00 per month for leasing the land the tower is on. Cell phone towers serve more traffic than broadband towers and have driven up the cost of a land lease for a tower. As an alterative you can spend $400.00 - $1,000.00 per month and co-locate on an existing tower saving the cost of tower erection. Unfortunately towers like all other expensive infrastructure follow population. In addition, you are limited as to how much equipment you can hang on any tower. The cost I have been quoted to bury fiber is around $5.50 per foot or about $30,000.00 per mile. For aerial fiber it was about $20,000.00 per mile. Bear in mind that aerial fiber would have monthly recurring rental costs for the poles it is hung on and would probably cost about $200.00 per month/per mile. Even if fiber were available for transport it would not lower the cost much, but it would increase the capacity. In this area it is at least 180 miles to the nearest level one Internet provider (a source of low-cost Internet bandwidth from ~ $6.00 - $12.00 per Mbps in quanity). We are 180 - 300 miles from five level one providers. Between 60% and 70% of the cost of our Internet bandwidth in this area is transportation cost. This is before your last-mile provider (ISP) has to add his transportation costs. Keep in mind, uptime or percentage of availability of your connection is directly related to the cost/quality of the transport medium used to deliver that connection.

The one hurdle that that is the biggest is network capacity. This can only be overcome with technology. Technology like most things in life is a trade-off. We can have somewhat better technology for much more money and much less range. What this means is we can have the capacity to stream video to many but it will raise the price of the Internet service for all on the network. It would easily double the cost of the network in hardware costs alone. Now factor in the bandwidth. If one-third of the customers on the network stream video, the demand for bandwidth will increase by a factor of ten. This does not account for the fact that with 33% of us watching streaming TV during prime time we will exceed the capacity of the Access point by 3.6 times. Another drawback to technology that supports more bandwidth is that range will be greatly decreased and coverage to as much as half the customers in our area will most likely be lost. An Access Point (the radio on the tower that your Subscriber Module connects to) can support from 50 to 150 users depending on the technology. But just 10 customers streaming TV on a 1.5 Mbps connection during prime time can use all the capacity of that access point. This means that the ISP could need as much as ten times the equipment to keep up with demand. So now I hope you are starting to see the value of your present TV service. I can't see a way to design a wireless network that will support 30% - 50% of it's customers watching streaming TV, even one channel per household during three hours of prime time viewing, and be affordable enough to remain in business. I know in our household, various family members watch two or three different channels at once.

I respectfully submit to you my opinion on Internet TV. The customer, content providers and the Internet are ready for Internet TV. Even some of the large providers (the ones that keep leaving us rural customers out in the cold while raking in billions in profit) are ready. It works well in areas where providers like AT&T and Verizon are level one providers with the capacity and have FTTP (Fiber to the Premises) as well. It works well on Comcast and other cable providers with peering agreements and existing infrastructure that readily support streaming video technology. In these examples they are just making more money on the excess capacity of expensive infrastructure that is already in place. However, most networks and especially wireless networks are not ready. Internet TV is most suited for fiber or coax and in my opinion ahead of it's time. When the cost of transportation catches up with it, it will not be such a great deal.

As a final note if you don't believe transportation of streaming video costs, just Google for the dispute between Comcast and Level3. It makes no difference what your opinion is of this dispute, it shows how much traffic they expect and that someone will pay for transportation. When it comes right down to it, there is no wireless technology available within the limits of tower capacity, range and reasonably priced service plans that will really support IPTV in a rural setting. I have carrier-grade PTP links that will transport it to the towers but no viable technolgy to deliver it to the customers.

Dave The TV on the internet guy

I love watching TV on my computer. One of the reasons is that I have a large flat screen monitor which is also a TV. I think that makes a real difference to the picture quality.

The second thing I did, very recently was to upgrade to cable from fiber in my home. Yes, I know it sounds like I'm going the wrong way, but the fiber only gave me 1.5 mbps and my cable shoved me up to 21.5. It made a huge difference.

Also I changed from a "G" level router to the "N' level at a speed of 300. While the change wasn't great on my desktop, it really made a difference on my wirelessly connected laptop.


Good point, but it will cost the cable providers to change the way their networks work. This might take some time and persuasion. Another thing, live TV over the internet is already available on some websites. Quality is below average but very watchable. http://ispsurvey.com


Visit http://www.online-tv-channels.net, here you can watch: Education, Entertainment, Kids & Youth, Movies, Music, News, Politics, Religion, Shopping, Sport and many more online TV Channels.


You can watch movie & TV online on this link :



Sci-Fi channel, Advendure movie, Classics channels, trailers, cartoon ...


http://www.mediaplanetaria.com is one excellent portal with online media. They have thousands of online tv, radio, newspapers, etc


New site to watch free online tv http://tv.stafex.net


Very good post. Internet Television is the next big thing on internet. Recently i have visited wfiTV.com which offers good collection of internet televisions

Jim B

Good post!! However, I would like to question your math. You say last-mile bandwidth requirements would go down, because instead of feeding 200 channels to each house, you only need to provide what is being watched at that time (1 or 2 channels). But a neighborhood cable loop can provide signal to well over 200 homes. Wouldn't that mean 200 channels, delivered simultaneously to 200 homes, would be the same as 1 unique channel or stream delivered to each? (Assuming for the moment that all the different types of streams are about the same bandwidth). And as the number of unique devices within the home goes up, the difference would grow as well.

This is an interesting topic, and I look forward to the convenience of TV on demand.


Great post! I never thought of it in those terms. In that sense the cable companies are perfectly hedged! If business migrates to IPTV no problem - cut back on the number of channels from 1258 to only 106 and the bandwidth appears. Consumer bandwidth costs go up but their cable bill goes down. Im sure i have oversimplified but a very interesting piece.

Mark Tomin

Good point, but it will cost the cable providers to change the way their networks work. This might take some time and persuation. Another thing, live TV over the internet is already available on some websites. Quality is below avarage but very watchable.


Richard Bennett

Interesting post, but misleading in some respects. Internet TV actually combines TV on Demand - the mode you talk about - and Multicast live streams. Putting live TV on the Internet as multicast isn't a huge problem, the BBC did it for the last Olympics and it was fine.

Providing multiple HDTV on-demand streams is the interesting part. You need 20-30 MB/s for HDTV, so the Internet core isn't equipped to handle very many of them using the obsolete end-to-end model, hence we need to discard it and develop some more advanced technologies. Unfortunately, our ability to do that is hampered by the Luddite net neutrality regulations, so we're going to be in deep water if they ever become law.

That being said, it's OK for the Internet to remain a text-based medium; literacy is a good thing, and watching TV makes people dumber. So the problem of delivering TV over the Internet may be better left unsolved.

Rene Churchill

Once Internet TV becomes a possibility to a significant portion of the population, the disruption to the existing TV business model is going to be very interesting to watch.
Why should a show have to pitch it's existence to the major networks in order to survive? Through iTunes or some equivalent, they can sell their content directly to the end consumer. There may well be two versions of each show available, a free/low cost version that contains advertising and a higher cost version that does not.
Why should I, the consumer, have to wait an entire week to see the next episode of a show that's sitting on the studio shelf somewhere? If a show is hot, I'll buy into the whole series and watch them in a weekend binge. From the studio's standpoint, if that show is hot, they can sell episodes as fast as they crank them out. Why wait to schedule them into some specific time block?
However, without a network providing the up-front money for a season's worth of shows, how the production is funded will have to change as well. Once the money starts flowing in a pay-per-view fashion, actors (and their slow to change unions) will need to negotiate their contracts on a percentage basis rather than a flat rate. No more million dollars per episode contracts ala Friends, etc.
Sites like Digg will evolve to review and rate shows that perhaps were never even broadcast. Much like bands, how well a show does will depend on how they can rally and motivate their fan base to introduce their show to other viewers.
Perhaps something similar to a book-publishing model will evolve, where a script is proposed, an advance made and the author gets a percentage off each episode sold. No need for a "book publisher" firm in this model either, anyone with some money to invest could play in the TV show market. Of course this could bring us full circle back to live theater's model, which thrived before TV. A TV show version of "The Producers", que up "Springtime for Hitler"!


I want to be able to subscribe to channels a la carte. This iTunes store as an alternative to cable is getting a bit expensive.

Jay Bowen

Great post!

I want to propose a thought... What if each broadband household in the last mile were pushing out the same stream of the TV show they were currently watching to another viewing household in the same neighborhood, city, state, or world?

Tapping a portion of unused upstream bandwidth each broadband household has available. Think Bittorrent...But for LIVE! Television. (not "Live" as in moving pictures on your PC screen as some marketers have confused it. But "live" as in "as it happens" moving on your PC screen)

Example: If you have four households pushing 150kb streams the receiving household is viewing a 600kb stream or TV program (near broadcast quality).

Now, with this scenario where do you see the bandwidth provider’s position?

Patrizia Broghammer

Good post, as usual ; Mr. Evslin.
And, as usual, Patrizia has something to say.
It's amazing how you always touch subjects I am very fond of and my husband is working at.
How much bandwidth is needed?
On the receiving site: Since, as you say, you are supposed to watch at a channel at a time, and since when you watch is most likely that you do not surf, considering the high and good compression available on the market today, you can have a stream of 1 Mbps steady bandwidth. (which is much more than you need for surfing, because it is STEADY) That is what most broadband connections allow now a days.
Providing that NOT all the ISP users are watching TV at the same time, because 1 Mbps of STEADY conmsuming bandwidth would crash most of network, admitting that if it is not possible right now, it will be possible in the near future (and I am talking about big cities, for small town it will still be a dream for long, at least in Europe...)but then : how much bandwidth is needed on the sending site?
And I am talking about real TV, the full screen at a quality close to the TV now.
I mean looking at TV on a decent monitor size, sitting on a couch, a full movie or a TV program for at least one hour or more of STEADY streaming... that is TV and not Youtube or all the similars, short clips of a size of a cell phone screen, sitting in front of your PC monitor.
Of course you agree, you cannot talk about TV in this case.
Well how much bandwidth on the sending side?
So huge nobody right now can afford, at least the TV we are talking about.
And very likely not tomorrow and not the day after tomorrow.
So what?
You have to create something that gives on one side the chance to see full screen at a decent quality, using less than 1Mbps bandwidth and on the sending side THE SAME.

It is not impossible, you say. It is possible, and we are working on that and hopefully we will be succesful one day soon.
Believe me THAT is the only possible WAY to have real IPTV

Post a comment

If you have a TypeKey or TypePad account, please Sign In.

Now on Kindle!

hackoff.com: An historic murder mystery set in the Internet bubble and rubble

CEO Tom Evslin's insider account of the Internet bubble and its aftermath. "This novel is a surveillance video of the seeds of the current economic collapse."

The Interpreter's Tale

Hacker Dom Montain is in Barcelona in Evslin's Kindle-edition long short story. Why? and why are the pickpockets stealing mobile phones?

Need A Kindle?

Kindle: Amazon's Wireless Reading Device

Not quite as good as a real book IMHO but a lot lighter than a trip worth of books. Also better than a cell phone for mobile web access - and that's free!

Recent Reads - Click title to order from Amazon


  • adlinks
  • adsense