Towers under too much pressure?

[Dovi]

Okay I'm thinking of getting iBurst and live in Emmarentia, so apparently I'd use the Northcliff tower (dunno if that's true ), but it sounds as if it's under too much pressure? If the towers can only handle 24mbps (I read that somewhere on the forum), and there are 2500 customers, which I'm sure many use that tower, and all have 1mb connections, then the tower won't be able to provide all the users with their bandwidth??!

 

[alchamy]

Thats pretty much the problem, the towers are feeling the pressure, been a little better with Randburg & Fourways up but nothing amazing. Northcliff is the worst of the lot, followed closely by Bryanston. Although I believe the problems on Bryanston are more technical then the load related.

The few times that I have been on Northcliff I have gotten speeds similar to ADSL, sometimes a little less. However when connected to the quieter towers things fly, slimothy is a great example of this. He is on Kempton and always gets 1mbit out of his connection.

WBS has tried to convince people that the towers can handle to load, however this is not true, once caps are in place then yes, people will not be sucking the tower dry, but until then its gonna be a fight for bandwidth.

Here is a link showing tower specs, notice the number of channels etc to understand the limitations.

http://global.kyocera.com/prdct/telecom/office/iburst/basestation.html

8 carriers, by 3 channels. Thats 24. Users share channels ard carriers, the system is spatial. However that is per every 5mhz used, unfortunately WBS would not give information on the exact frequency range, So I am not sure of the exact capacity. It could be a lot higher if they use a large spectrum. According to arraycomm its 20mbit per 5mhz, I assume the other 4 is used for signalling etc.

For detailed techinical information on how Iburst functions you can view the following PDF.

http://www.arraycomm.com/docs/iBurstOverview.pdf

I wish I had more information on the limitations, however its clear from the effects on busier towers that it cannot handle the current load. I would hazzard a guess at around 100mbit per tower, but that is being optimistic to say the least.

Ofcourse I am not expert on these things, merely intepreting research. If WBS understood honesty perhaps we would have a better idea.

 

[slimothy]

Is it worth it then for iBurst to put up more basestations for every 100 users on thier system?

 

[zoro]

If I remember right the BS has a bandwidth of 30Mb. They are currently running around 12Mb.

The BS has two MoNerds (Master and slave) that the UT's connect to. Each UT creates a 2-3% load on the MoNerd it is connected to depending on the use.

So you can work it out. if each UT connected creates a max load you will have around 33 UT's on each MoNerd. That makes 66 in total. Now 30Mb / 66 gives you 0.45Mb of bandwidth available. However they around running not at 30Mb but at 12Mb so we have 12Mb / 66 = 0.18Mb.

Dunno if my maths is correct there, haven't had my morning coffee yet

BTW each BS costs around R2m to install and setup so I doubt they will put up one BS for every 100 ppl

 

[guest2013-1]

I sometimes think companies like these (including Sentech who made the same mistake) thinks that because its a broadband 24/7 wireless connection, and that the bandwidth is shared, they can get away with less.

Yes, sharing the bandwidth blablabla... but if you have a 24/7 connection you still need to factor into the whole equation that people will try and download 24/7 if they can. Especially those that DOES go for broadband and need a mobile solution as well. (This is where Sentech f1cked up royally, they wanted to corner the 56k user market, the user that doesn't really need a 24/7 connection)

It's like buying a house to rent it out. You don't go buy the house and not be able to afford the repayments yourself while it's standing empty. You make sure you can afford it, then buy it, then rent it out. That way you don't really lose any money and you have everything to gain.

I doubt the ratio is 100 users per tower, but let's assume you're correct. They place 2 base stations and cover 200 people. Capital layout for 2 base stations is 4,000,000 (I'm not including cost of bandwidth or technical support etc)

200 users paying between R599 and R699. Let's say all of them buy the modems (meaning the modem capital layout from WBS is R0) and each of them pays R599.

(200x599) = R119800

It will take WBS around 33 months to recoup the costs of those 2 towers with a userbase of only 200 users.

So, the math speaks for itself, it's not a viable solution for WBS, so let's rework it a bit (usually a business wants to recoup costs in 12-24 month period, let's take 12 months to see how interesting it gets)

If you want to recoup R4,000,000 in 12 months, you need to generate an income of R333,333.33 per month. To generate that income, you need 557 users paying R599 per month.

Now, I'm unsure exactly how tower capacity works, and I haven't factored in bandwidth/technical support (and normal salaries needing to be paid etc)

But as you can see, it is a very good possibility that WBS will put up more towers to cut on the load. More towers = more users. More users = more money. More money = more towers. Goes on and on.

so in 12 months they'll recoup the cost of 2 towers. Let's say they ONLY have 2 towers and never increase the cost per month for a user (and the user base stays constant forever) They'll make a profit of 333,333.33 per month forever (obviously excluding bandwidth costs.

end long winded... they would probably put up a base station per 275 users. Obviously, it's speculation from my part.

 

[alchamy]

Nice post noone.

I hope WBS can post more detailed information on tower limitations. The vodacom post on tower limitations was very educational.

 

[ic]

Nice post noone.

I hope WBS can post more detailed information on tower limitations. The vodacom post on tower limitations was very educational.

IMO, v3g's post shows Vodacom's maturity in the marketplace - being open & honest with your customers is very important.

WBS haven't got a clue about how important it is to provide Customer Service & Support - sad, but I suppose they will learn not to play with matches sooner or later ; so far WBS is actively trying to burn their own house down before they have even finished building it .