I was typing a '?' after the + to indicate uncertainty but autocorrect (maisha ya mobile phones)
Interesting results. Reminds me of Telstra in Australia (Telstra, the largest mobile carrier in Australia, is the first company in the world to roll out a commercial Gigabit LTE network) where they achieved near optimal LTE-A speeds at their labs during a demo. They used top of the line gear at the time and were insistent that the demo be so locked down that no devices apart from a handful of test devices were allowed to access the Gigabit LTE cell tower.
With that said, Some questions for you (as a safcom employee with probable advanced knowledge) and the rest of the members:
- How many people have phones capable of 4G+?
- How many of them are in Westlands/Waiyaki Way?
- Of the 4G+ capable, how many have LTE CAT.6 (or above) antennas?
- Will these speeds remain consistent once rollout is complete and people upgrade their phones?
- How will the speeds perform in the CBD through buildings/Urban Canyon?
- Will there be enough signal points within the city to reduce (stop?) 4G capable phones from overheating? (This is a real problem for some if you're 15KM radius away from KENCOM)
I can go on, but the point is, we have a long way before we know exactly how these networks will perform with users and against each other. I look forward to a scenario where we can make such comparisons like 4G UK does HERE
Once people come on board we should have something between 40M-120 Mbps Down, 20-60 Mbps up.
The phones achieving the fastest speeds are the ones whose antennae can pull as many 4G parallel signals as possible (OFDM (orthogonal frequency division multiplexing), which divides a signal into parallel data streams that travel radio sub-channels.)
We had seen the S8+ achieve minimal speeds in town & even been told speeds in westlands before 4G+ were 40Mbps tops (See post HERE) So if @martingicheru still has access to a S8+, kindly test and post your results here
This can be done by using google maps to determine traffic density along waiyaki way then testing during these periods - ensures cell towers are well congested with commuters.
Test while in an uber to determine performance at different speed differentials - faster uber = slower internet speeds
Test at different areas with different population densities - Maybe create a quadrant/cluster, measure and determine average?