Fastest Speedtest.net speed tests in SA

yep my friend got 300mps speedtest before with vodacom, he work at the main building in jhb :P

its more like 80mps to cpt etc, but most of time he gets 200mbs+
 
Vodacom? Really? :rolleyes:
Again, useless rubbish as it does not translate into real world speeds

It's Red and Blue now baby :D

Fasted speedtest I got was via vodacom on 5/9/11 8:45am in hatfield being 6.19Mbps. However 1pm same spot it was crawling there...

And I am sure NASA also get great speeds on their core network :rolleyes:
1Gbps here on my core ;)
 
Is notable that no one reaches the 1Gbit mark! WTF Only one manages but half that speed. Some countries offer this speeds to home users! South Africa F.u.k. and wake up!
 
Is notable that no one reaches the 1Gbit mark! WTF Only one manages but half that speed. Some countries offer this speeds to home users!
That doesnt mean they'll necessarily fare any better on speedtest.net. At these speeds roundtrip latency (ultimately distance) starts becoming the limiting factor. Speedtest.net uses TCP so packet streams need to be continually acknowledged, and the longer the distance/latency the slower this happens.
 
Last edited:
Apparently http://testmy.net/ is better ;)

Don't be fooled about server location
TestMy.net will provide you with real-world broadband speed test results in real-world conditions. TestMy.net does not inflate scores to boost our users egos or host our test servers on the edge of ISP networks.

Some speed test sites lead users to believe that testing off a server close to them is necessary. This isn't always true. Our speed test servers are configured to maintain quality of service for thousands of miles. Most internet consumers don't think about the internet service providers responsibility in peering beyond their own network. Many internet providers send users to their own internally hosted connection test. When you surf the net, how often are you visiting a server down the street... the logic doesn't even make sense.
 
You missed 80% of the tcp spec, and the core of what makes tcp great.
That doesnt mean they'll necessarily fare any better on speedtest.net. At these speeds roundtrip latency (ultimately distance) starts becoming the limiting factor. Speedtest.net uses TCP so packet streams need to be continually acknowledged, and the longer the distance/latency the slower this happens.
 
I can beat that, all I need to do is set up a SpeedTest.net server on another PC, and test it via LAN and I can beat Vodacom and the other one, which is close to what they doing.
 
I can beat that, all I need to do is set up a SpeedTest.net server on another PC, and test it via LAN and I can beat Vodacom and the other one, which is close to what they doing.
Testing between two hosts a GbE switch should provide roundtrip latency measured in the micro seconds (< 1ms) so the network wont be limiter. The inability to saturate a GbE link has more to do with other limitations/inefficiencies on the two host PCs e.g. performance of NIC, NIC bus type, quality of NIC driver, OS IP stack performance & general performance of PCs.
 
This demonstrates a lack of security policies and allowing the posting of such results to public servers is poor judgement. An engineer needs no more than 64kb/s to do his job, any more and they are impacting the bandwidth of a paying customer.

An engineer with unresicted access on a core network can easiliy congest customers especially when he has left his torrent and news clients running in the background. Networks consist of domains and within these domains all resources should be constrained to provide fair use to all services. As an example, no constraints means a misguided download call kill the authentications server by starving the server's bandwidth or alternatively a fault will result in cascading problems as it saturates the network.
 
Am I missing something? Coz this is the speed I'm getting:

1425233964.png
 
And how relevant is this article to the general populus? zzzz

Well, some of it clearly shows your tax money at work for a change. Results posted by the universities, CSIR (including SAC) and CHCP were all obtained on SANREN, which is a 1Gbps+ (10Gbps on certain links in future) backbone connecting these research centres. Although the original article is quick to dismiss these as "core network tests", this in fact represents the speeds that these centres see in actual use. Note that this is local bandwidth, but international bandwidth via SANREN is not too bad either ... :)
 
Don't know why people are getting their knickers in a knot about these articles. They're lighthearted reading reporting on something purely of academic interest. Especially the 'core' tests. Also, myBB needs something to fill up the weekend slots. :)

Reporting speeds of many hundreds of megabits per second merely indicate there is some bottleneck somewhere in the system, be it the client PC, the network or the actual test server itself. Or any of their sub-components. We've got no way of knowing which, so these could easily not even be 'network' speeds but something else, maybe the NIC in the desktop that ran the test.

Take some time and understand the test and you'll understand what the results tell (and don't tell).
 
People reading these tests that do not understand what they really mean think that the "winners". Read the headline "Fastest-Speedtest.net-speed-tests-in-SA". It is rubbish and misleading to the public.
 
People reading these tests that do not understand what they really mean think that the "winners". Read the headline "Fastest-Speedtest.net-speed-tests-in-SA". It is rubbish and misleading to the public.

Since when does "speedtest.net" imply it must be a result anyone, and specifically the public, can achieve? Why are you making that connection? Next time myBB reports on the fastest computers, will you say it's misleading because your 386 can't achieve it?

There's nothing wrong with that headline. IF you understand the context. Which might require *shudder* reading up a bit. ;)
 
Since when does "speedtest.net" imply it must be a result anyone, and specifically the public, can achieve? Why are you making that connection? Next time myBB reports on the fastest computers, will you say it's misleading because your 386 can't achieve it?

There's nothing wrong with that headline. IF you understand the context. Which might require *shudder* reading up a bit. ;)

Riddle me this then what's the point of making it public? Seems to me that it's nothing more than an epeen achievement/competition...
 
Top
Sign up to the MyBroadband newsletter
X