ADSL performance indicators released (Q1 2012)

[jes]

ADSL performance indicators released (Q1 2012)

Telkom and MWEB have released their latest ADSL key performance indicators. Here is how they compare

 

[snail]

Dam telkom's kPI nationally suck compared to mweb....not even 100% worse.......over 1000% worse

 

[somewherenotthere]

Mweb are cheating the system. Prioritising speednet servers...

 

[Anthro]

ADSL performance indicators released (Q1 2012)

Telkom and MWEB have released their latest ADSL key performance indicators. Here is how they compare

Related Articles "ADSL KPIs: a waste of time?"

Do these REALLY matter all that much ?

 

[Nicko]

Related Articles "ADSL KPIs: a waste of time?"

Do these REALLY matter all that much ?

Well I don't know if it has anything to do with this, but I found that my ping with mweb is miles lower than anyone else's ping. 20-23 range.

 

[drukkie]

Mweb are cheating the system. Prioritising speednet servers...

Then maybe telkom should do the same LOl...will give us an accurate reading!

 

[Tharaxis]

It really boils down to this:

Without knowing the methodologies MWeb and Telkom employ to come up with these figures, you can't actually compare them. If M-Web is testing their network from within their network vs. Telkom testing it from outside their network, then obviously these numbers will display a vast difference (since M-web for example would not be taking into account IPC or SAIX -> MWeb routing).

For these numbers to be comparable, the tests need to be identical. ICASA should have indicated a standardized testing suite that all ISPs that publish KPIs have to conform to (and these tests may be annually revised) to allow their numbers to be considered "officially sanctioned". I don't understand why this wasn't done. Well actually I do, gross incompetence, but still.

 

[MickeyD]

It really boils down to this:

Without knowing the methodologies MWeb and Telkom employ to come up with these figures, you can't actually compare them. If M-Web is testing their network from within their network vs. Telkom testing it from outside their network, then obviously these numbers will display a vast difference (since M-web for example would not be taking into account IPC or SAIX -> MWeb routing).

For these numbers to be comparable, the tests need to be identical. ICASA should have indicated a standardized testing suite that all ISPs that publish KPIs have to conform to (and these tests may be annually revised) to allow their numbers to be considered "officially sanctioned". I don't understand why this wasn't done. Well actually I do, gross incompetence, but still.

Spot on.

And we say this every time these indicators are published!

 

[gregmcc]

These figures are pretty meaningless imo.

 

[saliehendricks]

Add this KPI: Average local loop distance

wait, then how does mweb compare. I dont see the value in comparing Mweb with Telkom?

 

[saliehendricks]

It really boils down to this:
...

For these numbers to be comparable, the tests need to be identical. ICASA should have indicated a standardized testing suite that all ISPs that publish KPIs have to conform to (and these tests may be annually revised) to allow their numbers to be considered "officially sanctioned". I don't understand why this wasn't done. Well actually I do, gross incompetence, but still.

Because nobody at ICASA knows how to come up with a standardized test? Perhaps they will ask Telkom to give them one

 

[somewherenotthere]

Speedtest.net is the worst way to test your speed

 

[somewherenotthere]

Then maybe telkom should do the same LOl...will give us an accurate reading!

Then we get silly articles about broadband speeds in SA increasing... MWEB don't allow users to get their linespeed yet they allow a bit extra for speedtest? not cool.