fast.com Speedtest results - please (-:

L

lightpixel

Well-Known Member
Joined
Jun 15, 2019
Messages
423
Reaction score
189
Hello fellow interweb folk

I am trying to get a real world comparison on my ISP speeds. Specifically using fast.com.

Not IPERF to Terraco using a special FNO account or anything fancy - Just a simple generic, easy, non technical fast.com test.

If you have a gap please post your result and your ISP line speed.

As sharing is caring, here are my results on Atomic Access and Frogfoot FNO 1 Gbps line.


771112
 
Frogfoot
200/200 CISP
d89fba9d7e9186750d6b5ddba60a7ef0.jpg
 
Fast.com is not the best measure for objective speedtest results. It comes down to Netflix’s cache design..

Basically, if you’re an ISP in South Africa and you peer at NAP JHB, your clients have access to Netflix’s main SA cache - and as such, the JHB fast.com server. There is only one general cache in South Africa, hosted in JHB. All ISPs who don’t have their own in-rack Netflix cache use this. Netflix then selectively provides in-rack caches for ISPs (based either on their load on the JHB cache, or their persistence in requesting one from Netflix). There are a few of these private caches in scattered around SA. This means that fast.com results in a particular region may be influenced by an ISP’s access to either their local cache or the JHB cache.

So sitting in Cape Town. A client on ISP A connects to Netflix JHB and a client on ISP B connects to Netflix in ISP B’s rack in Cape Town. (Basically a speedtest to your ISP) Hardly a balanced test. FAST.com is simply a visual version of the tool that Netflix uses to rightsize your connection. Basically, when you open a Netflix video, it tests your line to choose the optimal resolution it should use to send you.
 
websquadza said:
Fast.com is not the best measure for objective speedtest results. It comes down to Netflix’s cache design..

Basically, if you’re an ISP in South Africa and you peer at NAP JHB, your clients have access to Netflix’s main SA cache - and as such, the JHB fast.com server. There is only one general cache in South Africa, hosted in JHB. All ISPs who don’t have their own in-rack Netflix cache use this. Netflix then selectively provides in-rack caches for ISPs (based either on their load on the JHB cache, or their persistence in requesting one from Netflix). There are a few of these private caches in scattered around SA. This means that fast.com results in a particular region may be influenced by an ISP’s access to either their local cache or the JHB cache.

So sitting in Cape Town. A client on ISP A connects to Netflix JHB and a client on ISP B connects to Netflix in ISP B’s rack in Cape Town. (Basically a speedtest to your ISP) Hardly a balanced test. FAST.com is simply a visual version of the tool that Netflix uses to rightsize your connection. Basically, when you open a Netflix video, it tests your line to choose the optimal resolution it should use to send you.
Click to expand...

Thanks for this and your explanation. :-)
Yeah I know, I've been through the who controls the TCP/IP stack and peering explanations. Check the Frogfoot thread.

The reason I asked for the fast result is that its a simple result of a real world experience when watching Netflix.
;-)
 
DuckKnuckle said:
Vumatel trenched
CISP 1000/100

84e401b55caf039b45439043acb1bc23.jpg


Regarding real world performance, Steam and Origin average 80MBps, Battlenet is all over the show, and the Rockstar and Star Citizen launchers both hit 120Mbps
Click to expand...

Holy crap that's awesome. I rest my case. Vumatel and CISP kicks ass.
 
lightpixel said:
Thanks for this and your explanation.
Yeah I know, I've been through the who controls the TCP/IP stack and peering explanations. Check the Frogfoot thread.

The reason I asked for the fast result is that its a simple result of a real world experience when watching Netflix.
;-)
Click to expand...

And that's what fast.com is great for. Run a fast.com test to see what resolution you can get - but as a measure of linespeed and ISP performance, there are better tools. With free Netflix peering available, every ISP should easily be able to deliver pretty decent Netflix (and Youtube for that matter) performance (even if it means getting it from JHB to CPT).
 
websquadza said:
And that's what fast.com is great for. Run a fast.com test to see what resolution you can get - but as a measure of linespeed and ISP performance, there are better tools. With free Netflix peering available, every ISP should easily be able to deliver pretty decent Netflix (and Youtube for that matter) performance (even if it means getting it from JHB to CPT).
Click to expand...


Did not know they do free peering (Netflix). Thx @websquadza.
 
DuckKnuckle said:
Vumatel trenched
CISP 1000/100

84e401b55caf039b45439043acb1bc23.jpg


Regarding real world performance, Steam and Origin average 80MBps, Battlenet is all over the show, and the Rockstar and Star Citizen launchers both hit 120Mbps
Click to expand...

@DuckKnuckle You mind doing a Speedtest.net to Vox Cape Town please?
 
DuckKnuckle said:
8965368953.png
Click to expand...

You could download the internet before lunchtime at those speeds. :cool:

How much data is on the internet?

www.sciencefocus.com www.sciencefocus.com

"One way to answer this question is to consider the sum total of data held by all the big online storage and service companies like Google, Amazon, Microsoft and Facebook.

Estimates are that the big four store at least 1,200 petabytes between them. That is 1.2 million terabytes (one terabyte is 1,000 gigabytes). And that figure excludes other big providers like Dropbox, Barracuda and SugarSync, to say nothing of massive servers in industry and academia."



Disclaimer: Pls do not try as a test, ever. Besides intl connections from S.A are badly constrained currently.
 
OpenServe 40/20 AH CPT, I'm too cheap to pay for anything faster right now :giggle:


771414
 
You must log in or register to reply here.
Back
Top
Sign up to the MyBroadband newsletter
X