2-Way Affordable Satallite Access this Year.

Dustaz

no, im not beardy, i have a goatee...thats style mate....

jd

Originally posted by chernobyl
longword i found you post so funny as this logic has been the downfall of many many many *infinity* internet via satellite providers.


Actually longword, any bi-directional provider would work on the assumtion that each user would in theory use 100% of their bandwidth whenever they use the service.

No they don't.
And they don't work on ratios either.
What they (should) do is provide the capacity-one it gets to 70-80% get more.
The ratio stuff is marketroid bull****..

chernobyl

*Ratio*
80% users = 20% Bandwidth and vice versa.

It makes much more sense to exclusively sell bi-directional systems to SMEs as you know when bandwidth demands will be highest and you can easily develop user trends from it as "their" use will be consistent but providing a system to a home user without a leash is temtation enough for a warez dood.

longword

I promised myself I wouldn't post on this topic again, but I did a bit of research into the economics behind this and thought I should share with the community. I arrived at a figure of about $2000 per year per customer as an easily attainable price for a satellite service.

Have a browse around http://www.europestar.com/. It's a very poorly designed site - not as bad as the totally broken www.ntl.ie, but worse than the atrocious www.3com.com. Have a look at the Our Satellites->Overview page. They give a price of $450M for two 30-transponder sats with a groundstation. That would put a single transponder around the $7M mark. I figure a minimum 10 year useful lifespan for the satellite. Future technology can make better use of the available RF bandwidth so depreciation over that period shouldn't be much of an issue. Most of the satellite ISPs enforce long term contracts too - three years is common. I'm a bit hazy on satellite technology, so I don't know if a single transponder can cover two polarizations or just one - I've assumed the pessimistic latter.

Start out at about 40Mbit/sec per transponder, with one reserved for upstream, one for down, and offer users a 256k symmetric service with a very generous 10:1 or 20:1 contention ratio. You pay $14M up front for the two transponders (Eurostar and several others are in the business of selling these things as a commodity). With margins for yourself & Eurostar, service provision costs, and some allowance for undersubscription I think it's fair to treble that for a reasonable cost estimate so you'll need to make (14*3)/10 = $4M a year per pair of transponders. On that facility you can reasonably expect to fit (40000*10)/256 = just over 1,500 users - double that if you stick to a DSL/Cable style 20:1 ratio. The cost to each user per year would be $1300 to $2600 - between $110 and $220 per month depending on the contention ratio. As the service acquires more users, you buy or lease more transponders to maintain your low contention ratio.

I know I've made a lot of rough guesstimates, but are there any glaring holes in my numbers? The end result seems to be roughly in line with the satellite internet services offered to date. Looks financially viable to me.

PS. Those satellites were made by the French company Alcatel and launched on the European Ariane 4 from Kourou. My estimate of bandwidth per transponder is also a bit low - it seems QPSK on a 36MHz transponder gets you 60Mbit/sec. Digital TV will tend to use a more robust signal at nearer 40Mbit/sec since even very low rates of corrupt packets can result in completely destroyed pictures on a TV. I think the DVB spec demands a phenomenally high performance, something like 1 damaged bit per hour at most.

chernobyl

Packet Filtering is more a Digibox problem (or lack of it) rather than a DVB problem.

The max of any TP for Digital purposes that we are discusing is just over 38mb/sec.

$2000 per year....hmmm
Now when have Sky ever shown an interest in offering such products directly to business becuase that is not a home user price tag?

My point was that if (1)Sky were to offer this service that it would be (2) to an exisitng customer base and so (3) a single satellite would be required in order to provide this service, BS ratios aside, 28.2 is already feeling the bandwidth squeeze as any Sky customer can testify.

28.2/5 as a luxourious car park is almost full, there is @ most 2 spaces left but in reality you will only see 1 more satellite go up there when it is needed.

Sky are not interested in taking any more massive financial hits, Sky are now making a profit and i am sure they wont risk it any further, infact they are already considering cut backs in their programming offerings.

bottom line, there will not be any 2way IVS service from Sky but i do think that they would start some form of "content provision" with the exisitng platform.