[Snap] 100Mb / 20Mb UFB only connects to the USA via a 10Mb pipe

Forums › 2degrees (including Slingshot, Orcon, Flip, Stuff Fibre, MyRepublic, 2talk and Vocus) › [Snap] 100Mb / 20Mb UFB only connects to the USA via a 10Mb pipe

hellonearthisman

1819 posts

Uber Geek

Trusted

#172105 12-May-2015 01:00

Will Snap be boosting it's international speeds anytime soon?

I though the Southern Cross Cables had a lot of capacity available.

1 | 2 | 3

6018 posts

Uber Geek

Trusted

Geekzone

Lifetime subscriber

#1302387 12-May-2015 01:00

Hello... Our robot found some keywords in your post, so here is an automated reply with some important things to note regarding broadband speeds.

If you are posting regarding DSL speeds please check that

- you have reset your modem and router

- your PC (or other PCs in your LAN) is not downloading large files when you are testing

- you are not being throttled by your ISP due to going over the monthly cap

- your tests are always done on an ethernet connection to the router - do not use wireless for testing

- you read this topic and follow the instructions there.

Make sure you provide information for other users to help you. If you have not already done it, please EDIT your post and add this now:

- Your ISP and plan

- Type of connection (ADSL, ADSL2, VDSL)

- Your modem DSL stats (do not worry about posting Speedtest, we need sync rate, attenuation and noise margin)

- Your general location (or street)

- If you are rural or urban

- If you know your connection is to an exchange, cabinet or conklin

- If your connection is to a ULL or wholesale service

- If you have done an isolation test as per the link above

Most of the problems with speed are likely to be related to internal wiring issues. Read this discussion to find out more about this. Your ISP is not intentionally slowing you down today (unless you are on a managed plan). Also if this is the school holidays it's likely you will notice slower than usual speed due to more users online.

A master splitter is required for VDSL2 and in most cases will improve speeds on DSL connections. Regular disconnections can be a monitored alarm or a set top box trying to connect. If there's an alarm connected to your line even if you don't have an alarm contract it may still try to connect so it's worth checking.

I recommend you read these two blog posts:

- Is your premises phone wiring impacting your broadband performance? (very technical)

- Are you receiving a substandard ULL ADSL2+ connection from your ISP?

I am the Geekzone Robot and I am here to help. I am from the Internet. I do not interact. Do not expect other replies from me.

These links are referral codes: Sharesies | Mighty Ape

sbiddle

30853 posts

Uber Geek

Retired Mod

Trusted

Biddle Corp

Lifetime subscriber

#1302421 12-May-2015 07:37

Are you aware of how TCP works and how latency affects speed? And the massive congestion that exists on many US interconnects (California being somewhere that is really bad) that will give some exceptionally poor speeds to many speedtest servers in California?

Do you actually have some evidence (such as real world downloads) to show that speeds are slow to the US, or just a single speedtest result? It's also worth trying multiple servers on the West Coast, as some are horribly congested.

Behodar

10502 posts

Uber Geek

Trusted

Lifetime subscriber

#1302424 12-May-2015 07:43

While I fully agree that Speedtest isn't too reliable, those numbers do seem very low. I just ran a test myself to the "Unwired" server:

NonprayingMantis

6434 posts

Uber Geek

#1302451 12-May-2015 08:43

Behodar: While I fully agree that Speedtest isn't too reliable, those numbers do seem very low. I just ran a test myself to the "Unwired" server:

you probably should say what plan you are on for that speedtest to have any relevance.

on a 100/20 plan that would be ok. on a 200/200 plan that would be pretty bad.

jjnz1

1358 posts

Uber Geek

Lifetime subscriber

#1302456 12-May-2015 08:52

What about to the UK? I can't get more than 15mbps down on a Snap 200mbps connection!

Edit: using vast array of speed test servers...

Behodar

10502 posts

Uber Geek

Trusted

Lifetime subscriber

#1302471 12-May-2015 09:03

NonprayingMantis: you probably should say what plan you are on for that speedtest to have any relevance.

100/20, as per the thread topic :)

timmmay

20575 posts

Uber Geek

Trusted

Lifetime subscriber

#1302538 12-May-2015 10:01

I can easily get 30Mbps down from a US torrent using Snap UFB. It's possibly just TCP latency limiting speed.

NB: edited as I was quite wrong.

Equinox IT

Cloud spending continues to surge globally, but most organisations haven’t made the changes necessary to maximise the value and cost-efficiency benefits of their cloud investments. Download the whitepaper From Overspend to Advantage now.

dan

1134 posts

Uber Geek

Lifetime subscriber

#1302540 12-May-2015 10:05

timmmay: I can easily get 30Mbps down from a US torrent using Snap UFB. You may have an old switch/router locked on 10Mbps. Alternately it's just TCP latency limiting speed, but limiting very close to 10Mbps is suspicious.

er no. he posted 2 speed tests, one showing 100mbit download, has nothing to do with his switch/router.

timmmay

20575 posts

Uber Geek

Trusted

Lifetime subscriber

#1302557 12-May-2015 10:17

Good point. Then it's probably the TCP thing, though I thought Speedtest.net was multi threaded, which somewhat works around it through parallelisation - kindof like a torrent.

noroad

949 posts

Ultimate Geek

Trusted

#1302582 12-May-2015 10:32

OK, well here is a speedtest on a gigabit connected workstation direct into the Flip core network, so it should give you pretty much the maximum you can achieve from a USA based speedtest server to New Zealand or there abouts -

https://www.speedtest.net/my-result/4353062211

265/138meg

timmmay

20575 posts

Uber Geek

Trusted

Lifetime subscriber

#1302588 12-May-2015 10:36

One speed test isn't evidence of a poor network. I'd suggest doing speed tests to 10 servers in the US at say midday, 7pm, and midnight. If they're all slow then maybe escalate to Snap, otherwise it's just bad luck. Sometimes the internet is not lightening fast. First world problem.

sbiddle

30853 posts

Uber Geek

Retired Mod

Trusted

Biddle Corp

Lifetime subscriber

#1302840 12-May-2015 15:12

timmmay: One speed test isn't evidence of a poor network. I'd suggest doing speed tests to 10 servers in the US at say midday, 7pm, and midnight. If they're all slow then maybe escalate to Snap, otherwise it's just bad luck. Sometimes the internet is not lightening fast. First world problem.

Especially Californian servers. As I explained above if you chat to any network engineers on the West Coast they'll tell you about the poor connectivity between providers. The differences you'll get at different times of the day, and to different servers, are quite amazing.

hellonearthisman

1819 posts

Uber Geek

Trusted

#1302987 12-May-2015 17:07

Thanks for all the feed back, I will collect a number of speed tests from the USA West and East coasts (any recommendations?) as well as other countries connected to the southern cross cable. I have not seen any of my speed tests to the USA top 14Mb btw. I get decent speeds to Australia.

What would you recommend for real world speed monitoring as my browser doesn't give detailed data.

Multi threaded downloads do seem faster but when I have to use a single thread it's like adsl and not a fibre connection.

sbiddle

30853 posts

Uber Geek

Retired Mod

Trusted

Biddle Corp

Lifetime subscriber

#1303037 12-May-2015 18:41

Testing to the US East coast isn't ideal - speeds there can really suck. Once your traffic hits the US there is pretty much nothing your ISP can do about it.

fe31nz

1228 posts

Uber Geek

#1303103 12-May-2015 20:06

Multi threaded downloads do seem faster but when I have to use a single thread it's like adsl and not a fibre connection.

You have to remember how TCP works. There is a window size setting, k, which gives the number of TCP frames that can be sent before a reply packet with an ACK flag must be received. If the ping times are small, the standard k values in Windows work well - by the time k TCP packets have been sent, several reply packets with ACKs will already have been received and packets can keep on being sent continuously. But here in New Zealand, we often have quite long ping times to the rest of the world (except maybe Australia), due in part to the simple speed of light over the distances involved. So if you add all the time taken to send all the bytes in all the k packets that can get sent before an ACK must be received, it can be a shorter time than ping time, so by the time k packets have been sent, the first ACK had not yet arrived and transmission stops until it does. If you are doing a large download, that is why it is useful here in NZ to use a multithreaded download. Once things are going, some threads will be waiting for ACKs, while other threads have received ACKs and can send again. It used to be possible to change the k setting in Windows XP, but in later Windows versions there is no setting that can be adjusted. Instead it is supposed to be "adjusted dynamically", and if you leave a single threaded download running for long enough, you can see it pick up a bit more speed, indicating that maybe the dynamic adjustment is actually working somewhat. But I have never found it to work as well as using a multithreaded download.

Also, when you are downloading something you need to have enough bandwidth available in the upload direction for your ACK packets to get through. A very rough rule of thumb for this is that you need 10% of the download bandwidth in the upload direction. So if you have a 100/10 connection, you should just be able to max out the download bandwidth as long as there is no other traffic using the upload bandwith. If someone ever was stupid enough to sell 200/10 connections, it might well be impossible to ever get the full 200 download speed.

1 | 2 | 3