WiMAX could be a non-starter

always a catch if isn't telkom it's something just like telkom
 
The way I see it is wireless technologies seem to be popping up in all different shapes and form, theres something new almost every 3 months, there was 3G, then iBurst, then WiMax, then CDMA. Each with their own set of pros and cons. I rate we leave the economics up to the big people. At the rate things are going, if a certain technology costs too much for the end user, one would simply move to the next best thing. At the moment its just a matter of time until all major role players have achieved decent coverage.
 
R 75 mil for 50 towers (R 1.5 mil per tower). I think that those figures are right for a 'large faceless corporation', but that they can be brought down by a small niche company with dedicated, skilled employees. Rather have your own people doing the work, than subcontract the whole infrastructure project to someone else, who will further subcontract it down the line.

What exactly do you get for R 1.5 per pole. A big steel tower, a few sectoral antennas, a router and a battery backup (and 15 sub-sub-contractors get the other R 1 mil per pole).

50:1 contention. At R 1 000 /Mbps per month wholesale, that's R 20 per user per month. Not quite R 999. Where does that R 35 mil / year come from? Two okes in a bakkie can get the same job done for a fraction of that price.

Profit is of course hidden away in those figures, probably because its 50% of the price.

V3G, if you're bored, maybe give us your cost estimate of the components in these towers. I'm interested to know, and you might have figures available.

These articles about 'why prices must stay high' are old news. The various WUGs around the country have proven that wireless can work, and cheaply. Stop trying to hoodwink the uninformed masses with this bs.
 
Drunkard #1 said:
R 75 mil for 50 towers (R 1.5 mil per tower). I think that those figures are right for a 'large faceless corporation', but that they can be brought down by a small niche company with dedicated, skilled employees. Rather have your own people doing the work, than subcontract the whole infrastructure project to someone else, who will further subcontract it down the line.

What exactly do you get for R 1.5 per pole. A big steel tower, a few sectoral antennas, a router and a battery backup (and 15 sub-sub-contractors get the other R 1 mil per pole).

50:1 contention. At R 1 000 /Mbps per month wholesale, that's R 20 per user per month. Not quite R 999. Where does that R 35 mil / year come from? Two okes in a bakkie can get the same job done for a fraction of that price.

Profit is of course hidden away in those figures, probably because its 50% of the price.

V3G, if you're bored, maybe give us your cost estimate of the components in these towers. I'm interested to know, and you might have figures available.

These articles about 'why prices must stay high' are old news. The various WUGs around the country have proven that wireless can work, and cheaply. Stop trying to hoodwink the uninformed masses with this bs.
Click to expand...


+1
you've hit the nail on the head.
 
I don't think it's as simple as Drunkard might suggest, but R1.5 million per tower seems excessive.

Manufacturers of Wimax technology need to get smart fast. They need to drop prices. Network Providers will say why implement wimax now, let's wait until everyone else does and the price (of equipment) goes down. If they adopt that attitude, no one actually adopts wimax and eventually it dies out.
 
Drunkard #1 said:
R 75 mil for 50 towers (R 1.5 mil per tower). I think that those figures are right for a 'large faceless corporation', but that they can be brought down by a small niche company with dedicated, skilled employees. Rather have your own people doing the work, than subcontract the whole infrastructure project to someone else, who will further subcontract it down the line.

What exactly do you get for R 1.5 per pole. A big steel tower, a few sectoral antennas, a router and a battery backup (and 15 sub-sub-contractors get the other R 1 mil per pole).

50:1 contention. At R 1 000 /Mbps per month wholesale, that's R 20 per user per month. Not quite R 999. Where does that R 35 mil / year come from? Two okes in a bakkie can get the same job done for a fraction of that price.

Profit is of course hidden away in those figures, probably because its 50% of the price.

V3G, if you're bored, maybe give us your cost estimate of the components in these towers. I'm interested to know, and you might have figures available.

These articles about 'why prices must stay high' are old news. The various WUGs around the country have proven that wireless can work, and cheaply. Stop trying to hoodwink the uninformed masses with this bs.
Click to expand...

50 towers maketh not a network.

Subscribers don't use towers, they use services, like accessing the internet, doing e-mail, playing games, etc.

There's a crapload more stuff in a complete network than just the towers, all the bits, technology, transmission lines, buildings, services that interconnect the towers. The centralised services such as international internet access, etc. The data centers with lots of systems in them. Etc, etc.

So saying the network is R75M / 50 towers = 1.5M per tower is a bit simplistic.
 
vodacom3g said:
50 towers maketh not a network.
Click to expand...
Base stations are also pretty dumb devices: Turn on... pump out RF. Turn off... no RF. Fried electronics... put in new component.

Its all the stuff behind the towers that will cost R35 million per year to maintain. Most of that will end up in paying for skilled labour. Sure you can skimp on the labour costs but then you won't have a network with any kind of decent uptime :D
 
I know it's not that simple, but I'd guess that R 500 k per tower is reasonable, considering that, if the company made the effort, a lot of those 'towers' could be attached to existing buildings. The idea of a 'Pretoria and Jo'burg only' isp seems to be a good starting point, so lets assume that 50 base stations can support the network. That's R 25 mil.

What does a small isp need to provide. A radius server, an email server, and a bit of hosting space. That can all fit into 1 rack. Shouldn't cost over R 500 k, if you negotiate hardware prices, and setup the software in house.

Local and international connectivity is a big issue, but shouldn't be impossible to set up, plenty of ISPs have done it already, poach their skilled people. If JINX is too costly, make deals with key ISPs individually, it benefits both parties. Tower interconnects (and direct ISP links) could be wireless, using 802.11, and a few directional antennas included in the R 500 k price of those towers.

I still cant see what would cost R 75 mil to set up, R 35 mil / year to maintain, and R 999 per month per user to provide.

Like a previous poster said: 'Maybe VANS will have better success with this'. An ISP should raise the capital and give it a go. Of course, this all hinges on getting an I-ECNS licence, but maybe the consumer will win for once. I hope someone does build a cheap and efficient WiMax network, and makes a profit. WiMax could still provide the jolt to kick start competition in this sector, and even increase local hosting. Good luck to anyone who tries.
 
I agree with the drunkard here. We tend to over complicate things. Build it well and build it cheaply like the WUG's and your costs will be nowhere near the figures being quoted. Also, you do not have to offer services such as e-mail to a user as most people already have these services available to them FREE. Most people just want the connectivity. Ok so for those who need services... Google will now give you in excess of 7GB free e-mail, full online office, cloud computing and more, Free and there are already many companies offering the same. OpenDNS is a free great dns service for getting your name resolutions done with security, antiphising features, etc in place. There is already a very comprehensive radius/auth infrastructure in place that can authenticate these connections, whether you use IS, SAIX or any of the others...

So yes, towers maketh not a network, but each time you create a network you also do not need to re-invent the wheel. Imagine if we had tried to build the internet that way... we'd get nowhere fast.

My point: Leverage and build on top of what is already there.
 
So does that mean that they will need 35 000 users on their network just to break even each year? I doubt iBurst has that many users on their network and they still seem to be going strong. Am I missing something here?
 
Last edited:
Drunkard #1 said:
Profit is of course hidden away in those figures, probably because its 50% of the price.

These articles about 'why prices must stay high' are old news. The various WUGs around the country have proven that wireless can work, and cheaply. Stop trying to hoodwink the uninformed masses with this bs.
Click to expand...

QFT
 
Hmm as much as i'd love to call Dr Arnab an LTE loving nazi [T-Systems = T-Mobile] he has a valid point ! Another issue with wimax is spectrum the 3.5Ghz is all used up, 2.5Ghz is still being allocated but won't provide adequate coverage & wall penetration , 2.3Ghz is for military use & 700mhz won't be available till 2011. Companies like altech & is should partner with sentech & start a new company akin to clearwire in the u.s.
 
Similar to what InfraCo and DFA are supposed to be doing for fibre/wholesale broadband supplyling, I think some company should build a WiMAX infrastructure and lease off portions to the other ISPs at reasonable costs etc to cover their maintenance/initial capital outlay.

Then, instead of having 3 or more companies trying to build WiMAX networks.

Just my R0.02
 
Oh well, some more years before I see internet out where I stay, 24 hour internet that is
 
dgcarter said:
The way I see it is wireless technologies seem to be popping up in all different shapes and form, theres something new almost every 3 months, there was 3G, then iBurst, then WiMax, then CDMA. Each with their own set of pros and cons. I rate we leave the economics up to the big people. At the rate things are going, if a certain technology costs too much for the end user, one would simply move to the next best thing. At the moment its just a matter of time until all major role players have achieved decent coverage.
Click to expand...

Every 3 months? More like one a year at most.
 
You must log in or register to reply here.
Back
Top
Sign up to the MyBroadband newsletter
X