Internet Speeds

[Stealer]

https://www.speedtest.net/result/11563423863.png
for 24 usd per month and no caps

Nice, your upload speed is much better than mine

[[MiA]Dave]

Y'All can stop ur complaining now.....
USD $129/mnth

https://www.speedtest.net/my-result/d/329797440

My normal ISP got shutdown. New one gives me an 8MB throttled SIM (SMH).
Supposed to get an 'unthrottled' one delivered to me tomorrow.
Paid my deposit for StarLink. Hope they come through in the next few months.

[Piglet]

Why is it so expensive though? Surely at that price there would be more competetion.

Is it the long distances involved?

[[MiA]Dave]

Distance limits my choices. Geo-satellite or Cellular provider is about it. Problem with both is they don't 'really' want us as a customer, they are forced to offer something. They do, by GB. So, somewhere in the hidden abyss of AT&T there's a method to provide unlimited data (up to 1TB) at LTE speeds to a fixed location. I have to pay between $100-250 to leverage that option through a middle-man rather than pay $50 directly.

I figure as soon as Starlink comes out of beta and looks to be a popular option good ol' capitalism will create a wealth of affordable options to me.

[Rage]

https://www.speedtest.net/result/11916053337.png

my new connection =) 35x better than old

[Rage]

[Rage]

wow dave i thought i had a bad connection previously thats half of what i had and i got unplayable pl

[Piglet]

You don't need much bandwidth for UT..but pl is horrible to have.

If much rather have lower bandwidth without pl than higher bandwidth with pl

[Stealer]

I held out for as long as I could, got the upgrade.
Think I'm paying £60 or 70 pcm, but that includes landline and calls.

[Miauz55555]

Code: [Select]

|------------------------------------------------------------------------------------------|
|                                      WinMTR statistics                                   |
|                       Host              -  Loss %  | Sent | Recv | Best | Avrg | Wrst | Last |
|------------------------------------------------|------|------|------|------|------|------|
|                           xxx.xxx.xxx.xxx -    0 |  257 |  257 |    1 |    1 |   15 |    1 |
|                           xxx.xxx.xxx.xxx  -    0 |  257 |  257 |    5 |    8 |   72 |    5 |
|                            xxx.xxx.xxx.xxx -    0 |  257 |  257 |    5 |   10 |   73 |    7 |
|             hbg-b1-link.ip.twelve99.net -    1 |  253 |  252 |    8 |   11 |   45 |   12 |
|            hbg-bb3-link.ip.twelve99.net -    0 |  257 |  257 |   21 |   25 |   85 |   25 |
|            ldn-bb1-link.ip.twelve99.net -    0 |  257 |  257 |   23 |   25 |   87 |   24 |
|             ldn-b3-link.ip.twelve99.net -   30 |  117 |   82 |    0 |   25 |   41 |   25 | ... 62.115.120.75
|upc-ic306717-ldn-b3.ip.twelve99-cust.net -    0 |  257 |  257 |   24 |   28 |   88 |   25 |
|                   No response from host -  100 |   51 |    0 |    0 |    0 |    0 |    0 |
|          uk-lon03a-ri2-ae-2-0.aorta.net -    0 |  257 |  257 |   21 |   25 |   85 |   23 |
|                   No response from host -  100 |   51 |    0 |    0 |    0 |    0 |    0 |
| brnt-ic-1-ae0-0.network.virginmedia.net -   46 |   91 |   50 |    0 |   27 |   66 |   36 | ... 62.254.42.198
|                   No response from host -  100 |   51 |    0 |    0 |    0 |    0 |    0 |
|sotn-core-2b-ae2-0.network.virginmedia.net -    1 |  253 |  252 |   24 |   27 |   71 |   25 |
|sotn-metnet-3a-lag-56.network.virginmedia.net -    0 |  257 |  257 |   25 |   27 |   90 |   26 |
|    drkn-bh1-ia1.network.virginmedia.net -    0 |  257 |  257 |   25 |   28 |   90 |   28 |
|                   No response from host -  100 |   51 |    0 |    0 |    0 |    0 |    0 |
|                   No response from host -  100 |   51 |    0 |    0 |    0 |    0 |    0 |
|                   No response from host -  100 |   51 |    0 |    0 |    0 |    0 |    0 |
|                   No response from host -  100 |   51 |    0 |    0 |    0 |    0 |    0 |
|                   No response from host -  100 |   51 |    0 |    0 |    0 |    0 |    0 |
|                   No response from host -  100 |   51 |    0 |    0 |    0 |    0 |    0 |
|                   No response from host -  100 |   51 |    0 |    0 |    0 |    0 |    0 |
|                   No response from host -  100 |   51 |    0 |    0 |    0 |    0 |    0 |
|                   No response from host -  100 |   51 |    0 |    0 |    0 |    0 |    0 |
|                   No response from host -  100 |   51 |    0 |    0 |    0 |    0 |    0 |
|                   No response from host -  100 |   51 |    0 |    0 |    0 |    0 |    0 |
|                   No response from host -  100 |   51 |    0 |    0 |    0 |    0 |    0 |
|                   No response from host -  100 |   51 |    0 |    0 |    0 |    0 |    0 |
|                   No response from host -  100 |   51 |    0 |    0 |    0 |    0 |    0 |
|________________________________________________|______|______|______|______|______|______|

the connection to the servers is, since ~6 days extreamly bad for me. But I think it's on the routing end. I have midgame for like 10s a total loss .. so not even 1 packe will go in.
---------
As for speedtest, I use http://www.dslreports.com/speedtest it is one of the few which is also meassuring your BufferBloat.
If you want to see an example of a very bad connection, there you go. The ping is 8 to 32 for EU servers btw .. but the BufferBloat and Jitter not. =)
The other speedtest just meassure the idel ping, which is always fine. And even though you will not notice the high upload LAG when streaming and doing other things which aren't realtime.

[holyspam]

Today I would like to introduce you to bufferbloat.

Disclaimer - This is surface-level stuff !!

Bufferbloat simply put, is when you saturate your link speed and everything else that requires some bandwidth becomes unusable.
For example you download a file and your latency ingame jumps to 100-300-1000ms


As Miauz posted, the DSLReports speed-test used to be the go-to tool to test this.
Now all the servers are offline except 1-2 so it's not that useful any more.

But I managed to find another webpage that provides a bufferbloat test. Here
This will measure your latency while the connection is idle and then again while you are downloading and uploading, so you can compare and know if you have an issue.

The solution to this is to reserve some of your bandwidth - around 5-15% of your max measured speed is usually enough.
What reserving bandwidth accomplishes is that when you download...stuff, it will allocate that reserved bandwidth to a service/application that only needs a small portion of it.


So let's say your measured connection speed is 150Mbits Down - 10Mbits Up.
When you download something, it will use all of your download speed - it will really try to use it all.

You setup a simple QoS bandwidth limiter on your router, and set a limit to 90% of your real speed so 135Mbits Down - 9 Mbits Up.
Just doing this will solve 70% of the issues.

The other 30% requires special QoS classification rules for your games and applications which is very time-consuming, although some routers have integrated QoS that will detect games and other low speed - high priority traffic.



Here you can see the difference between on/off for my line:
My router is TP-Link AX1500.

Detailed Results with QoS off - unlimited speed



Detailed Results with QoS on - limited to 135/8.5




Keep reading if you want to get 100% otherwise skip the rest of my post


The most intense UT2004 game requires 40KB/sec maximum download speed and 8KB/sec maximum for upload (for 250fps)
That is < 1% of your Download and Upload capacity.

Similar examples are VoIP traffic, Video Calling(this can easily consume more than 1Mbit/sec) - but these are *standardised* high priority traffic and are handled by default.

Unless you are using some quality router, gaming traffic will be considered "Best Effort", while Video calls, Voice over IP are considered higher priority.

A quick overview:

Purpose		Common Use		DSCP Value
VoIP		VoIP traffic, including signaling and control traffic		46
Interactive video		Two-way video conferencing		34
Mission-critical data		Database queries, LOB communications, video streaming		25
Best effort		All other traffic, including e-mail and Web browsing		0
Bulk data		Backups, nonbusiness applications, file transfers		10

Generally, all traffic has no DSCP value(default 0), the easiest way to get around this, is to assign a DSCP value to your gaming traffic.
This will only apply to your local network so your router knows what traffic is high priority - once you reach the ISP the value will be reset back to 0.

To achieve this in Windows, you need a version that contains the Group Policy Editor(Ultimate, Enterprise) will work. (it does not work in Home edition)

Steps:

• Start menu
• Run
• Type gpedit.msc and press OK/Enter
• Local Computer Policy -> Windows Settings -> Right click on Policy-based QoS -> Create new policy
• Use any name you like, then in "Specify DSCP Value" use 46 or 34
• Click Next, then select "Only applications with this executable name" and enter the name "UT2004.exe"
• Click Next, and again Next on the Source/Destination IP address page.
• In the drop-down, "Select the protocol this QoS policy applies to:" - Pick UDP and click Finish.

Now your UT2004 traffic will be marked as high-priority and hopefully handled as such by your router/network.

If you are on wifi - switch to Ethernet cable, don't game on wifi wtf??
Wifi traffic supports some basic QoS prioritisation by default on 90% of routers/APs which is commonly called WMM you can check your device wireless settings, make sure it's enabled.

[-Adler-]

Working great thank you.

without SQM
Details



In OpenWrt i installed luci-app-sqm 90Mbps Down - 10 Mbps Up
Details



just for info
SQM (Smart Queue Management)
The luci-app-sqm package (or “sqm-scripts” if you don't use LuCI) will mitigate Bufferbloat. After installation and configuration you'll have a much more responsive network connection under load.

[holyspam]

You can use like 30Mbit/s instead of 10

[Stealer]

Ah no way, I bought a TP link router to replace the shitty isp one (which broadcasts a public channel/hotspot which is stupid if you are a gamer!), seemed much smoother - it has more cores and antennae so win win. Think I have a similar one to you holy. Also noted the QOS was referred to as pinholes on the tp link router, confused me for a sec... Lol had another look whilst sober, there is QOS on the AX1500 

[hagis]

find it varies - depending on time of day - sometimes it's quicker but generally around this



vodafone 900/900mbps - got a deal @ £26/month - though they screwed up so much installing it I have 12 months free

edit and really I don't need the speed,. I won't notice a difference if it was 1/10 of that 99.9% of the time