Wimax?

bk

useruser wrote:

I did bother to read it, I just disagree with you, that's all! Don't be so touchy.
I think I could summarise your position as "Wireless broadband is unusually successful in Ireland because of lack of availability of DSL." My position is that wireless broadband is unusually successful in Ireland because DSL is over-priced and when the price inevitably drops wireless operators will be unable to compete head to head with wireline products.

The reason I'm being "touchy" is because that isn't what I said. What I'm saying is that because Eircom delayed the rollout of DSL for so long, it allowed wireless tech to mature (LOS) and therefore it is in a better position to compete.

useruser wrote:

Hmmm. HSDPA? 1xEVDO? Mobile Wimax doesn't exist yet, UMTS will provide 300-400kbps today, where are the multi-megabit mobile applications? There are already flat-rate HSDPA packages available from Cingular.

The companies who don't have license for this tech like the cablecos and telcos, are the ones who are interested in Wimax to compete with HSDPA, etc. They see mobile BB as an important extension to their current triple play operations.

useruser wrote:

As for the mobile killer app? The cellcos have been asking this question for a while now and the idiots still haven't realised that the answer is simply: "the internet".

You just supply the pipe, your customers will come up with the killer apps.

useruser wrote:

Fixed wireless is not far less expensive to deploy than DSL (unless you are including the cost of running the copper!). Obviously there are subscribers that cannot be reached by DSL in which case wireless may be the answer.

Yes in rural areas, people are so far from exchanges and the copper is in such a bad condition and there are so many dacs, that the only way to get DSL to many people will by laying lots of new copper.

useruser wrote:

Nonsense, I can think of 4 companies off the top of my head with 1m+ (combined) IPTV subscribers (PCCW, Free, Neuf, Fastweb). Why do you think it's not possible?

It is possible it just doesn't scale. A standad definition transmission uing MPEG 2 uses 4Mbit/s. With ADSL2+ you are lucky to get 12Mbit/s. What happens f the person wants two streams (8Mbit/s) or what about 3? I've got 3 digital boxes and 3 analogue connections in my house. That would be impossible with IPTV over DSL.

And what about HD, it will use about 10Mbit/s using MPEG4, this will barely work on ADSL2+.

Magnet have started to do IPTV over ADSL2+ here in Ireland and the first reports say that it is pretty bad if you want more then one stream. It is a toy compared to cable, fibre or sat (for broadcasting).

useruser wrote:

I bet all the wireless operators just can't wait to get into this lucrative market - there must be tens of thousands of euro in it.

Another great market, that will pay for the CEOs' lunches.

I can think of 3 friends, all of whom live within 3 miles of O'Connell St and still can't get DSL. I wouldn't hold my breadth, FWA is going to be filling the gap for a long time to come.

useruser wrote:

And all of those people who just have to have broadband when they're on the go - they must number in the thousands! Must draw up a business plan, there's money to be made!

Yes there is, I honestly believe that 10 years from now we will all be walking around with at least one device (game console, MP3 player, mobile, UMPC, etc.) that is constantly on the internet. We will need mobile BB for that. Antime, anywhere will be the new mantra.

useruser wrote:

So, once the DSL prices come down (9.99 for naked DSL in France at the moment) and the services come along (TV, VoD, VoIP) where does that leave the wireless operator?

I wouldn't hold my breadth for either, between Comreg and Eircom we willnever get that pricing and as I said above TV over DSL just won't scale.

useruser wrote:

Or, more likely, most people will have one fixed triple-play provider and there will be a small market for mobile high-speed broadband.

Triple play, maybe on cable and fibre, but not DSL. DSL can't do TV well and most people are dumping landlines for mobils anyway.

Most people will have wireline BB. But there will also be a relatively large and profitable market for mobile BB and a secondary market for FWA as in fill were DSL doesn't serve (a large area).

watty

useruser wrote:

I don't understand this comment, why is DVB "better" for broadcast?

15km at 10GHz is really pushing it in my experience - heavy fog may knock you out - I would have said 11kms.

200miles at 430MHz - just goes to show what could be done.

A man of mystery! Go on, tell us more!

DVB is actually an IP broadcast format. It can carry non AV multicast IP data as a broadcast stream. Strict vanilla TCP/IP has no mechanisim for multiple broadcast streams that is compatible with normal routers, firewalls etc.

I'm 12.5km from a Metro base. I lose about 3dB in heavy fog or torrential rain. Still plenty of signal margin.

As distance increases the main issue is getting LOS. A twice the width aerial would have 4 times the gain = twice the distance. However various issues do limit 10GHz terrestrially to about 15km. Atmospheric conditions on a very clear day with no wind can cause longer microwave paths to fade out almost completely.

A Satellite at 50W or 42E can be received in Ireland (about 45,000km away), it would be interesting to calculate how much of the path is in the atmosphere. I do get more rain fade on BBC World @13E than on Metro.

BK also clearly shows that indeed IPTV (and by implication realtime VOD) is a toy.

3G's data is about 340k. About twice ISDN or one tenth of typical broadband. It is also more expensive per megabyte than GPRS. "EDGE" over GSM system is almost the same speed as GSM.

There is an overhead always. To get 5Mbps video (BBC typically) you need at least 6Mbps 1:1 (uncontended) IPTV. HD TV for one channel needs about 15Mbps 1:1 (uncontended) BB. Oddly DVB has a much higher overhead. It is used for FEC (Forward Error Correction). Video on BB as IPTV tends to be UDP as there is not time to resent lost packets via TCP/IP. Internet has no defined QOS, so with UDP there will be lost video packets. With DVB there is Data FRAME based FEC, so depending on FEC (1/2 most overhead 7/8 least overhead) "lost packets" can be recovered. No picture pixelation/breakup.

Cable TV could use IPTV , but uses DVB for A/V as this is better quality/ more efficent / more compatible.

Satellite, even for two way Internet (!) uses a data stream inside a DVB frame. TCP/IP packet acknowledgement is not done over the medium, but at each end.

Rather than IPTV over cable, the present method of separate DVB and DOCSIS spectrum makes more sense. DOCSIS ins a sense packages the BB and makes it more like a Digital TV (DVBc) signal.

There is no reason why a hybrid scheme could not also be developed for copper using DVBc and adsl2 rather than all of the spectrum for vdsl.

useruser

bk wrote:

The reason I'm being "touchy" is because that isn't what I said. What I'm saying is that because Eircom delayed the rollout of DSL for so long, it allowed wireless tech to mature (LOS) and therefore it is in a better position to compete.

Doesn't sound any different than "Wireless broadband is unusually successful in Ireland because of lack of availability of DSL" to me. You can't beat the fact that fixed wireless is very expensive in comparison to DSL and will not be able to compete on price when LLU finally brings prices down.

As for the mobile killer app? The cellcos have been asking this question for a while now and the idiots still haven't realised that the answer is simply: "the internet".
You just supply the pipe, your customers will come up with the killer apps.

Ah yes, "build it and they will come" not sure I believe that one. Again, it's not that I don't believe that some people want this, I just question the logic of an operator putting 100s of millions into it.

Yes in rural areas, people are so far from exchanges and the copper is in such a bad condition and there are so many dacs, that the only way to get DSL to many people will by laying lots of new copper.

This is a small, and difficult market to cover, can't see the business case myself. If you want to serve these users with FWA it will cost a fortune, base station occupancy will be too low.

It is possible it just doesn't scale. A standad definition transmission uing MPEG 2 uses 4Mbit/s. With ADSL2+ you are lucky to get 12Mbit/s. What happens f the person wants two streams (8Mbit/s) or what about 3? I've got 3 digital boxes and 3 analogue connections in my house. That would be impossible with IPTV over DSL.

I just pointed out 1m+ users of IPTV that I could name off the top of my head, I don't know what the world wide figure is. Do you not consider 1 million users "scale?" Guess you're hard to please!

And what about HD, it will use about 10Mbit/s using MPEG4, this will barely work on ADSL2+.

Interesting point, I will do a bit of research on HDTV over DSL. Do you know how many HD users are forecast for the next 5 years in Europe?

Magnet have started to do IPTV over ADSL2+ here in Ireland and the first reports say that it is pretty bad if you want more then one stream. It is a toy compared to cable, fibre or sat (for broadcasting).

So, one small Irish player have made a mess of their IPTV deployment? Hardly a damning endightment of the whole technology? What about the 1 million users above?

I can think of 3 friends, all of whom live within 3 miles of O'Connell St and still can't get DSL. I wouldn't hold my breadth, FWA is going to be filling the gap for a long time to come.

This huge sample set has convinced me! Seriously, I'm sure I can think of 3 people who can't get FWA either - I'm not sure that either statistic is telling.

Yes there is, I honestly believe that 10 years from now we will all be walking around with at least one device (game console, MP3 player, mobile, UMPC, etc.) that is constantly on the internet. We will need mobile BB for that. Antime, anywhere will be the new mantra.

I suspect you are correct, however I don't think that Wimax (or its descendants, will be connecting these devices.)

I wouldn't hold my breadth for either, between Comreg and Eircom we willnever get that pricing and as I said above TV over DSL just won't scale.

I disagree, we are behind the rest of Europe in broadband but I am hopeful that we will get there in the end - I am sure IOFFL are similarly hopeful, why bother otherwise?

Triple play, maybe on cable and fibre, but not DSL. DSL can't do TV well and most people are dumping landlines for mobils anyway.

1 million users of IPTV today! I think you are entirely wrong with the notion that "most" people are dumping landlines. Care to provide some statistics?

Most people will have wireline BB. But there will also be a relatively large and profitable market for mobile BB and a secondary market for FWA as in fill were DSL doesn't serve (a large area).

I'd add to this - "these wireless markets will be small with tiny ARPUs. Not particularly compelling business"

bk

1 million users is not significant or scale.

There are over a billion users of sat and close to a billion users of cable TV.

Magnet hasn't messed up because they are bad, but because the tech just doesn't deliver over DSL. All the other players you mentioned in IPTV have had the exact same problems and non of them has even thought of challenging HD yet. I've been watching this area with interest for a while now. It does (sort of) work if you just want one SD stream, that is where the 1 million customers come from, but it doesn't scale to multiple streams or HD, so it will always be doomed compared to cable, fibre or sat. I suppose a bit like what you are saying about wireless

It is interesting to note that Verizon, the company with the largest DSL subscriber base in the world has decided to skip ADSL2+ and VDSL and IPTV as they can't deliver and is instead going straight to Fibre To The Home to deliver TV services (ironically they aren't even using IPTV over this!).

Here is a fantastic article that explains the whole TV over BB market, well worth reading: http://www.dslreports.com/shownews/65298

As for HD penetration in Europe in 5 years, I don't have any figures, but it is expected to be widely available and very much mainstream in 5 years, with the majority of high value pay tv customers on HD. It took far less then 5 years for HD to take over big time in the US.

I disagree with you on wireless business, but I don't think I can convince you, so lets agree to disagree.

bk

watty wrote:

Cable TV could use IPTV , but uses DVB for A/V as this is better quality/ more efficent / more compatible.

Actually a very interesting idea that has been floated in the catv world is a mixed network.

At the moment you have over 100 channels using up a lot of bandwidth (even in digital). However well over 90% of all the TV viewed is only on the top 20 channels (terrestials, etc.). All the other channels are rarely watched.

The idea is to deliver the top 20 channels using traditional DVB, while you transparently deliver the remaining niche channels via IPTV.

The advantage is that you potentially free up a lot of bandwidth to deliver more channels, HD channels and faster BB speeds, while it wouldn't use that much extra bandwidth as 90% of the time people are watching the top 20 channels on DVB. Very clever.

watty

Except if even a few people watch the less used channels at the same time using DVB is more efficent than IPTV.

The ideal number of viewers er IPTV is one, the ideal number viewers for DVBx is anything greater than one.

A users PVR now can easily have the same number of different titles as a video library now, so you use broadcast to send them slower than even one IPTV channel and then any number of users can watch any number of movies at the same time.

So how do you apply this to TV Stations. Are some programs watched by nobody? Or would be watched by nobody if you could order your own custom TV station for the month ahead, and programs that got insufficent vote 3 weeks in advance would be "deleted".

In US they argue that IPTV would replace Cable TV where you get a bunch of channels and all the programs wheter you want them or not. Well, rigerous mathematics rather than wishfull thinking has always proved that real IPTV only works if few people watch at all, and is not scaleable. But a "faked" VOD works very well, which is why on 160G Sky+ half the storage is "reserved".

I can see broadcasting stations being unhappy with the "roll your own station" concept, using real VOD or the better "faked" version. Rather than improving quality it might make niche programs disappear as companies become more conservative an pander to an even lower common denominator.

We live in interesting times, but some operators could save themselves and custome grief if they did basic arithmetic.

useruser

watty wrote:

DVB is actually an IP broadcast format. It can carry non AV multicast IP data as a broadcast stream. Strict vanilla TCP/IP has no mechanisim for multiple broadcast streams that is compatible with normal routers, firewalls etc.

Umm, Multicast? Seems to be working OK for 1m users? I can see the advantages of DVB but say you're a telco who doesn't want to get into the broadcast business?

BK also clearly shows that indeed IPTV (and by implication realtime VOD) is a toy.

I don't think he's demonstrated that at all? 1 million users demonstrates that IPTV is realistic and scalable.

3G's data is about 340k. About twice ISDN or one tenth of typical broadband. It is also more expensive per megabyte than GPRS. "EDGE" over GSM system is almost the same speed as GSM.

But this is more than adequate for most mobile needs today? So you're not going to be able to watch SD TV on the move today but so what?

Rather than IPTV over cable, the present method of separate DVB and DOCSIS spectrum makes more sense. DOCSIS ins a sense packages the BB and makes it more like a Digital TV (DVBc) signal.
There is no reason why a hybrid scheme could not also be developed for copper using DVBc and adsl2 rather than all of the spectrum for vdsl.

Is there anyone working on such a scheme?

useruser

bk wrote:

1 million users is not significant or scale.

This is a nonsensical statement. On what planet is 1 million users not an indication of a successful technology? Clearly there is momentum behind IPTV and it can be deployed today.

There are over a billion users of sat and close to a billion users of cable TV.

So what? There are many more analogue lines than DSL enabled lines in the world, does that mean dial-up is superior?

Magnet hasn't messed up because they are bad, but because the tech just doesn't deliver over DSL. All the other players you mentioned in IPTV have had the exact same problems and non of them has even thought of challenging HD yet. I've been watching this area with interest for a while now. It does (sort of) work if you just want one SD stream, that is where the 1 million customers come from, but it doesn't scale to multiple streams or HD, so it will always be doomed compared to cable, fibre or sat. I suppose a bit like what you are saying about wireless

This is absolute rubbish, IPTV doesn't just "sort of" work for SDTV over DSL, it demonstrably works! 1 million users are watching TV today over DSL. It will be interesting to see how they handle HDTV and I think maybe Watty's ideas around "fake" VoD could come into play there.

It is interesting to note that Verizon, the company with the largest DSL subscriber base in the world has decided to skip ADSL2+ and VDSL and IPTV as they can't deliver and is instead going straight to Fibre To The Home to deliver TV services (ironically they aren't even using IPTV over this!).

Verizon are deploying ADSL2+, they have stated that their primary video offering will be fibre. How many other Telcos can afford to do this? My guess is that they will deliver TV over copper too.

Here is a fantastic article that explains the whole TV over BB market, well worth reading: http://www.dslreports.com/shownews/65298

From the article:

BBR: IPTV in general, do you see it as a serious competitor to Satellite and cable?

DB: Single channel, not HD IPTV is working well, with a million customers around the world and tens of millions coming in the next few years. Multichannel, HD, to several sets turns out to be much harder and takes more bandwidth, which is why it's coming slower. But $30B in planned investment is coming, and almost surely by 2006-2010, millions will be buying fancy TV programming from telcos.

He reckons $30B will be invested in the technology? Maybe there's something to it after all?

Dave Burstein is well worth reading - www.dslprime.com, he's a strong proponent of IPTV by the way.

I disagree with you on wireless business, but I don't think I can convince you, so lets agree to disagree.

Well, we definitely disagree alright, I note that no-one is suggesting that we will see a fixed wireless broadband TV product. Could that be because it is not a realistic possibility?

watty

Have you watched IPTV compared with a Satellite receiver. Or tried to run IPTV with two different TVs?

Lots of people are seduced by the idea of IPTV and pouring money into it, but the sums and the quality don't add up to a scaleable service. BB is typically 24:1 to 50:1 contention. If everyone used it for IPTV you would need to plan around 1:1 contention. VERY expensive.

useruser

watty wrote:

Have you watched IPTV compared with a Satellite receiver. Or tried to run IPTV with two different TVs?

Lots of people are seduced by the idea of IPTV and pouring money into it, but the sums and the quality don't add up to a scaleable service. BB is typically 24:1 to 50:1 contention. If everyone used it for IPTV you would need to plan around 1:1 contention. VERY expensive.

Yes, I have seen IPTV and it looked fine to me - I can only watch a single TV at a time (I realise that there is this limitation but I don't think it's a big deal). Again I come back to the fact that there are already 1 million IPTV subscribers, that alone indicates that it is viable.

What is this guff about 1:1 contention? The broadcast video streams are multicast out to IP DSLAMs - easy peasy, no need for 1:1. VoD obviously needs 1:1 but what are the ratios required? 20:1 or less?

bk

DB: Single channel, not HD IPTV is working well, with a million customers around the world and tens of millions coming in the next few years. Multichannel, HD, to several sets turns out to be much harder and takes more bandwidth, which is why it's coming slower.

You conveniently ignored this part. See he says that one stream SD can be done, but multi stream SD or HD is very hard, exactly what I've been saying.

Most people want more then one stream, watch one show while recording another or have a second/third TV in bedroom/kitchen. This is very normal in the pay tv world which tends to target households with more disposable cash. IPTV over DSL simply can't compete with cable, fibre or sat in this area and you can completely forget about HD over DSL which most people in the US now demand and over the next 4 years will become the norm in Europe.

Just because you wish it doesn't make it so, the maths simply don't add up:

MPEG2 SD: 4Mbit/s plus extra for tcp/ip overhead
MPEG4 HD: 11Mbit/s plus extra for tcp/ip overhead

ADSL2+ average user speed in the real world about 12Mbit/s

Square peg, round hole.

But $30B in planned investment is coming, and almost surely by 2006-2010, millions will be buying fancy TV programming from telcos.

Remember Verizon are using fibre, must of those millions will be over fibre. IPTV over fibre can work, but ironically Verizon aren't actually using IPTV over fibre, they are using traditional DVB tech over fibre.

useruser

bk wrote:

You conveniently ignored this part. See he says that one stream SD can be done, but multi stream SD or HD is very hard, exactly what I've been saying.

I conveniently ignored it by quoting it! Hmm. So 2+ SD streams and HD is difficult, so what? - 1 million users seem to be OK with that?

Most people want more then one stream, watch one show while recording another or have a second/third TV in bedroom/kitchen. This is very normal in the pay tv world which tends to target households with more disposable cash. IPTV over DSL simply can't compete with cable, fibre or sat in this area and you can completely forget about HD over DSL which most people in the US now demand and over the next 4 years will become the norm in Europe.

I doubt it, I don't want more than one stream, my guess is that "most" people couldn't give a toss as long as it's cheap.

IPTV over DSL is already competing with cable, fibre and satellite - why do you find that so hard to believe?

Just because you wish it doesn't make it so, the maths simply don't add up:
MPEG2 SD: 4Mbit/s plus extra for tcp/ip overhead
MPEG4 HD: 11Mbit/s plus extra for tcp/ip overhead
ADSL2+ average user speed in the real world about 12Mbit/s

Square peg, round hole.

So, what you are saying is that IPTV over DSL is not possible and there are in fact NOT 1 million such subscribers today? What is your beef with this? Why not accept that it is a workable alternative to other TV technologies?

Remember Verizon are using fibre, must of those millions will be over fibre. IPTV over fibre can work, but ironically Verizon aren't actually using IPTV over fibre, they are using traditional DVB tech over fibre.

The article says that $30B will go into IPTV, it does not mention DVB.

Again I must ask - why are you arguing the point so strongly? IPTV is a working technology, it gives telcos a foothold in TV that they otherwise would not have and as such is to be welcomed - the more competition the better. The fact that there are more efficient methods of delivering TV to the end-user is neither here nor there - are you seriously suggesting that Telcos should not try to take full advantage of the copper loop available to them?

I'm finding it harder and harder to take your points seriously, tell me again that 1 million users does not represent "scale!"

bk

useruser wrote:

I conveniently ignored it by quoting it! Hmm. So 2+ SD streams and HD is difficult, so what? - 1 million users seem to be OK with that?

I doubt it, I don't want more than one stream, my guess is that "most" people couldn't give a toss as long as it's cheap.

I'm sorry, in that case you simply do not understand the dynamics of the pay tv industry. Watty and I are trying to explain it to you, but you just don't want to listen.

There are two types of TV customers, people who want TV for free or cheap and those who are willing to pay for extra quality.

People who want TV for free or cheap will be served by DTT or maybe FTA sat. IPTV will not be suitable for these types of customers as it is relatively expensive to run.

People who buy pay tv typically want more channels, ability to record one channel while watching another, multiple TV's and high picture and sound quality (HD over the next 5 years).

IPTV over DSL simply can't offer any of these services that are expected by the pay tv market.

useruser wrote:

IPTV over DSL is already competing with cable, fibre and satellite - why do you find that so hard to believe?

So, what you are saying is that IPTV over DSL is not possible and there are in fact NOT 1 million such subscribers today? What is your beef with this? Why not accept that it is a workable alternative to other TV technologies?

I'm saying that only one SD feed is usually possible and HD is not possible, can't you do the maths for yourself?

useruser wrote:

The article says that $30B will go into IPTV, it does not mention DVB.

Again I must ask - why are you arguing the point so strongly? IPTV is a working technology, it gives telcos a foothold in TV that they otherwise would not have and as such is to be welcomed - the more competition the better. The fact that there are more efficient methods of delivering TV to the end-user is neither here nor there - are you seriously suggesting that Telcos should not try to take full advantage of the copper loop available to them?

I'm finding it harder and harder to take your points seriously, tell me again that 1 million users does not represent "scale!"

Seriously can you not read?

When I say scale, what I mean is that it can't deliver multiple feeds to each house. It can't scale for one customers needs.

You continuously refuse to ignore the factual figures that I've put forward about the bandwidth use of IPTV, I was actually being rather generous, it is actually far worse.

Cable TV uses statistical multiplexing, these means that it only averages about 4Mbit/s, if the TV show has lots of fast moving footage, it can burst upto 7Mbit/s. This is very hard to do on DSL and is why the PQ of IPTV over DSL is far worse then CATV (which is inferior to sat).

BTW, There are more then 1 million FWA users in the world, does that mean I was right all along about FWA and that it does have a future? Please make up your mind.

1 million IPTV customers makes up less then 1% of the cable or satellite market, it is barely a niche market.

I'm sure there is a market for people who only want one SD stream, with a picture and sound quality inferior to cable, fibre or sat, for almost the same price as cable or sat with much greater capabilities. But as you said about FWA, it is a fairly niche market.

Blaster99

People are seemingly signing up for Magnet's IPTV service, so there is presumably a market. A lot of people don't get much excitement from broadband on its own and I guess Magnet figures that adding TV to the mix makes it a bit more appealing to the average consumer.

I don't really get the whole fascination with finding new ways of delivering TV. Surely the whole TV distribution thing has been solved a long time ago? It doesn't strike me as a problem worth solving again. But if it is worth solving, why aren't the wireless guys getting into it? They can broadcast it surely, just like a cable company can. Perhaps there isn't enough bandwidth available or is there a licence issue?

useruser

bk wrote:

I'm sorry, in that case you simply do not understand the dynamics of the pay tv industry. Watty and I are trying to explain it to you, but you just don't want to listen
There are two types of TV customers, people who want TV for free or cheap and those who are willing to pay for extra quality.
People who want TV for free or cheap will be served by DTT or maybe FTA sat. IPTV will not be suitable for these types of customers as it is relatively expensive to run.
People who buy pay tv typically want more channels, ability to record one channel while watching another, multiple TV's and high picture and sound quality (HD over the next 5 years).

All interesting points.

IPTV over DSL simply can't offer any of these services that are expected by the pay tv market.

And yet, there are 4 companies with 1 million IPTV paying customers between them, all in developed markets with cable and satellite readily available - these are not insignificant numbers.

I'm saying that only one SD feed is usually possible and HD is not possible, can't you do the maths for yourself?

I agree, but 1 million subscribers don't see this as a problem, I don't either.

Seriously can you not read?

When I say scale, what I mean is that it can't deliver multiple feeds to each house. It can't scale for one customers needs.

I can read, what you actually wrote was (and I quote) "1 million users is not significant or scale." Care to rephrase that?

You continuously refuse to ignore the factual figures that I've put forward about the bandwidth use of IPTV, I was actually being rather generous, it is actually far worse.
Cable TV uses statistical multiplexing, these means that it only averages about 4Mbit/s, if the TV show has lots of fast moving footage, it can burst upto 7Mbit/s. This is very hard to do on DSL and is why the PQ of IPTV over DSL is far worse then CATV (which is inferior to sat).

Golly, I didn't realise how awful it must be for those 250,000 Fastweb TV viewers, not only do they have to put up with Italian TV but the bandwidth used can be up to 7mb! They must be turning off in their droves! Why do their subscriptions increase every year? Beats me.

BTW, There are more then 1 million FWA users in the world, does that mean I was right all along about FWA and that it does have a future? Please make up your mind.

Yes of course you are right, FWA has a good future in underdeveloped markets (Africa especially) and infill for low density suburban areas. I expect to also see it used for government subsidised rural access. I do not expect to see FWA widely used in developed urban markets five years from now. What has this got to do with IPTV?

1 million IPTV customers makes up less then 1% of the cable or satellite market, it is barely a niche market.
I'm sure there is a market for people who only want one SD stream, with a picture and sound quality inferior to cable, fibre or sat, for almost the same price as cable or sat with much greater capabilities. But as you said about FWA, it is a fairly niche market.

I think I have made some cogent arguments as to why FWA will disappear from urban markets. I don't see what this has to do with triple-play DSL vs broadcast TV technologies.

Here are a few simple points (again):

1. IPTV works - there are 1m+ users (across only 4 companies) paying for the service today.
2. IPTV over DSL allows telcos to provide a triple-play service today without the massive investment required for PON/FIOS.
3. Broadcast TV rots your teeth, causes famines and was Hitler's favourite past-time when he wasn't busy invading Russia (He couldn't get the History channel so was doomed to repeats - or something).

Suffice to say, it looks to me as if IPTVoDSL is here to stay. You have presented evidence to show that it has competition but the fact remains that it is an excellent opportunity for Telcos to enter the TV market and squeeze some more ARPU out of the copper loop.

cgarvey

Can we stick to the constructive argument without the silly snide remarks (please)?

watty

Blaster99 wrote:

People are seemingly signing up for Magnet's IPTV service, so there is presumably a market. A lot of people don't get much excitement from broadband on its own and I guess Magnet figures that adding TV to the mix makes it a bit more appealing to the average consumer.

I don't really get the whole fascination with finding new ways of delivering TV. Surely the whole TV distribution thing has been solved a long time ago? It doesn't strike me as a problem worth solving again. But if it is worth solving, why aren't the wireless guys getting into it? They can broadcast it surely, just like a cable company can. Perhaps there isn't enough bandwidth available or is there a licence issue?

IBB ripwave etc can't handle broadcast TV. But Digiweb have already publically said Metro will have a TV service, maybe HDTV.

IMO there can be a licence issue too.

I wonder how many people with Magnet will get TV aerials or dish on Balcony. Many have no choice in apartments with dish forbidden and no cable TV wiring.

There is less competition than people think. Only in Rural area where choice is MMDS or Satellite or 2 to 4 Aerial channels

useruser

Blaster99 wrote:

People are seemingly signing up for Magnet's IPTV service, so there is presumably a market. A lot of people don't get much excitement from broadband on its own and I guess Magnet figures that adding TV to the mix makes it a bit more appealing to the average consumer.

I don't really get the whole fascination with finding new ways of delivering TV. Surely the whole TV distribution thing has been solved a long time ago? It doesn't strike me as a problem worth solving again. But if it is worth solving, why aren't the wireless guys getting into it? They can broadcast it surely, just like a cable company can. Perhaps there isn't enough bandwidth available or is there a licence issue?

I think it is likely that straight broadband will be an unusual product within the next few years in developed markets. All operators are gearing up to provide at least broadband and voice and most are also considering TV. I suspect those who don't/can't will disappear. How will they compete when the actual broadband service rental gets down to the €10/month level? It's not about creating more excitement for the end-user, it's about creating more revenue for the operator.

The main point about IPTV over DSL is that it allows telcos to provide TV over their copper loops without the (huge) expense of putting in fibre. Correct me if I'm wrong Watty but I don't think that incumbents would usually have access to broadcast spectrum even if they wanted to go down that route.

bk

useruser wrote:

Suffice to say, it looks to me as if IPTVoDSL is here to stay. You have presented evidence to show that it has competition but the fact remains that it is an excellent opportunity for Telcos to enter the TV market and squeeze some more ARPU out of the copper loop.

Oh, that is all true. I'm sure the telcos are desperate to get a slice of the TV market, they see it as a way to compete against cable who are entering the telco and BB market.

All I'm trying to say is that:

1) It is a pretty bad solution for broadcast TV, which is already a pretty mature market.
2) IPTVoDSL will never offer what most pay tv customers want (multiple streams and HD).
3) It will remain a small niche market because of the above problems.

BTW Half your 1 million customers are on PCCW in Hong Kong. Typically they run fibre to each apartment block and then distribute using copper around the building. This makes the lenghts very short and it means they can deliver much greater bandwidth and therefore multiple SD streams and HD.

Most of Fastwebs TV customers are on FTTH.

This article is also interesting:
http://www.isp-planet.com/research/2005/iptv.html

"It means a lot of people can't get multi-channel TV other than over DSL," Bosnell says. "That's the demand side reason for it."

It's also technologically and economically more feasible to build an IPTV network and service in big cities in Europe and places like Hong Kong where population density is very high and one high-bandwidth fiber connection to a high-rise will let you light up a significant number of potential subscribers.

Blaster99

So long as people are happy to pay €20 a month for NTL's diabolical analogue service, I'd say there is plenty of scope for a 30 channel IPTV setup along the lines of Magnet. It's not scalable but you're dealing with a market that NTL/CableLink has been milking for a very long time.

useruser

bk wrote:

Oh, that is all true. I'm sure the telcos are desperate to get a slice of the TV market, they see it as a way to compete against cable who are entering the telco and BB market.
All I'm trying to say is that:
1) It is a pretty bad solution for broadcast TV, which is already a pretty mature market.
2) IPTVoDSL will never offer what most pay tv customers want (multiple streams and HD).
3) It will remain a small niche market because of the above problems.

So we agree (nearly), let's look at those 3 points:

1. I say it's just "not as good." it's only "bad" if you need multiple SD streams or HD - if it was cheap I'd be happy.
2. Seems very unlikely given the investment going into it that this will remain the case.
3. As you point out PCCW have 500k subs in Hong Kong - that is a lot more than a niche already. (Population is 7m or so - what % of households does 500k subscribers represent?)

BTW Half your 1 million customers are on PCCW in Hong Kong. Typically they run fibre to each apartment block and then distribute using copper around the building. This makes the lenghts very short and it means they can deliver much greater bandwidth and therefore multiple SD streams and HD.

Most of Fastwebs TV customers are on FTTH.

This article is also interesting:
http://www.isp-planet.com/research/2005/iptv.html

So, there you have it - PCCW actually already deliver multiple SD streams and HD TV over DSL (albeit over very short loop lengths) - Just goes to show what is possible. Others are experimenting with multiple loops and new DSL tech (DSM for example) is extending reach and bandwidth all the time. Looks to me as if IPTV works pretty well.

bk

useruser wrote:

So we agree (nearly), let's look at those 3 points:

1. I say it's just "not as good." it's only "bad" if you need multiple SD streams or HD - if it was cheap I'd be happy.

The picture quality of a single SD feed also tends to be inferior to sat or cable and there isn't any reason why it would be cheaper then sat or cable. I don't understand where you get that idea from.

useruser wrote:

2. Seems very unlikely given the investment going into it that this will remain the case.

You can invest all you want, physics is physics and it can't be overcome. An unshielded twisted pair only has about 1MHz of radio frequency, you just can't change that magically.

useruser wrote:

3. As you point out PCCW have 500k subs in Hong Kong - that is a lot more than a niche already. (Population is 7m or so - what % of households does 500k subscribers represent?)

So, there you have it - PCCW actually already deliver multiple SD streams and HD TV over DSL (albeit over very short loop lengths) - Just goes to show what is possible. Others are experimenting with multiple loops and new DSL tech (DSM for example) is extending reach and bandwidth all the time. Looks to me as if IPTV works pretty well.

It isn't usually DSL, they normally use fibre to the apartment block and then distribute using Cat5 (networking cable) or COAX. They rarely use DSL, so it isn't really an example of IPTV over DSL.

Honk Kong has one of the highest population densities in the world, with almost everyone leaving in apartment blocks over 20 storeys. This makes this sort of setup viable. It isn't viable in most places (except maybe Tokyo and New York).

As I've said, IPTV is possible over fibre, coax or even VDSL, but all of these require massive capital investments and basically running fibre to the curb. Your dream of just using the existing copper and adding TV is not realistic, specially in most Western European countries with relatively long cable runs, pair gains and bad quality copper.

BT has recognised this problem, their solution is that they are going to sell Freeview boxes with a DSL modem builtin. You will use Freeview for the broadcast TV and the DSL for BB and interaction.

Patrickof

High quality TV via satellite is available everywhere today. Most people don't care how the signal gets to their TV sets. So unless any IP based service is cheaper (and would need to be significantly cheaper) or much better, then there's no need for people to change.

Also, why clog already congested networks with extra traffic thats being quite successfully delivered via satellite or terrestrial transmission today.

If it ain't broke.......

watty

Satellite has capacity for about 2000 HD channels on one orbital slot. I can pick up twenty different satellites. I have about 1200 free TV and 1100 Free radio and currently 6 satellite receivers running off only two dishes.

Blaster99

I'm gonna ask a potentially stupid question that I've been wondering about for a while. In a terrestrial context, frequencies are managed, allocated, and sold. How does satellite work? Do they blast away at whatever frequencies they feel like or is there international frequency space that one can buy from? Or is it not an issue even?

watty

It is an issue. The Terrestrial and Satellite is all decided internationally in ITU conferences etc. Terrestrial and Satellite can interfer with each other.

Though As you go above 20MHz the likelyhood of long distance drops rapidly, there are borders and even 1GHz can go 1000km or more sometimes. It is all co-ordinated from about 10kHz to 1000Ghz roughly. www.comeg.ie to see what we use it all for in Ireland.

Satellites have footprints and orbital slots as well as frequencies. Astra had to pay millions in compensation when they pinched slot/frequencies allocated to Eutelsat for Central Europe. That is why Sky platform now comes from an Eutelsat bird as well as the 3 Astra birds. Eutelsat's Eurobird also has a Central Europe beam.

International agreement decides the satellite stuff and National Coverments get told what the terrestrial is for. LW, MW, SW particular the allocatoins are not up to Comreg but Europe or even world wide allocation (though Comreg licences those agreed frequencies to local broadcaster). VHF/UHF/Microwave Comreg licences the individual allocations, but has to laiase particularly with Ofcom and a lessser Extent France and Spain (VHF to Spain and France fiarly regular)

cgarvey

Patrickof wrote:

High quality TV via satellite is available everywhere today.

Not it's not! It's not available in lots of Dublin (rented accomodation, planning laws, apartment blocks with management companies).

.cg

watty

That should be illegal. At least the residents should be allowed to club together and have a single dish on roof comunal system.

Ah well some advantages of living in Countryside, no Adsl but lots of dishes

useruser

How did I get involved in a debate about the merits of IPTV?!

bk wrote:

The picture quality of a single SD feed also tends to be inferior to sat or cable and there isn't any reason why it would be cheaper then sat or cable. I don't understand where you get that idea from.

OK, it's perhaps not quite as good as sat or cable but so what? If it's priced right and included with my voice and broadband package that's fine by me. I am that user who is happy with the basic NTL analogue service - I just don't watch that much TV.

You can invest all you want, physics is physics and it can't be overcome. An unshielded twisted pair only has about 1MHz of radio frequency, you just can't change that magically.

This is just silly, who's talking about overcoming physics? Shannon's law still has plenty to give! If we had been discussing this 15 years ago would you have been telling me that DSL was impossible?

It isn't usually DSL, they normally use fibre to the apartment block and then distribute using Cat5 (networking cable) or COAX. They rarely use DSL, so it isn't really an example of IPTV over DSL.
Honk Kong has one of the highest population densities in the world, with almost everyone leaving in apartment blocks over 20 storeys. This makes this sort of setup viable. It isn't viable in most places (except maybe Tokyo and New York).

Both Neuf and Free use DSL for the vast majority of their subscribers - it is absolutely possible to do this. I notice also that PCCW is leading research into HDTV over DSL - I assume you would counsel them to save their money?

As I've said, IPTV is possible over fibre, coax or even VDSL, but all of these require massive capital investments and basically running fibre to the curb. Your dream of just using the existing copper and adding TV is not realistic, specially in most Western European countries with relatively long cable runs, pair gains and bad quality copper.

So, at least you'll give me VDSL then! We're getting somewhere.

BT has recognised this problem, their solution is that they are going to sell Freeview boxes with a DSL modem builtin. You will use Freeview for the broadcast TV and the DSL for BB and interaction.

http://www.theregister.co.uk/2005/09/28/bt_iptv/ BT to launch IPTV by summer 2006, VoD included apparently, they claim that the viewer will be able to switch between freeview and BB channels on the same box. Sounds to me like they are going to do IPTV over DSL - what's your read on it? (Don't just say that it's not possible!)

useruser

watty wrote:

IBB ripwave etc can't handle broadcast TV. But Digiweb have already publically said Metro will have a TV service, maybe HDTV.

IMO there can be a licence issue too.

I thought that, am I correct in thinking the 3.5GHz licences do not permit TV? How will Digiweb get around that?

Can you write some more about how you think Digiweb will do TV over Metro (is that an Airspan product?) I understand it's a DOCSIS system but where will they get the bandwidth from? 2x14MHz just isn't very much spectrum - what am I missing?