|
|
|
michaelmurfy:
3) I am not going to disable 5GHz WiFi off my Wireless LAN at the moment however if you're using this your speeds will be slower again (I get around 20-40mbit down, 20mbit up).
Jase2985: okay if you say so, you know better than the spark rep
sbiddle:jaysin:
The traffic shaping done by Spark was only on streaming and P2P ports also.
Browsing was fine, but streaming you tube and P2P / torrent connections were shaped down to miserable speeds.
Changed back to the capped 150GB plan, and this no longer occurred
Spark do not and have not shaped traffic. End of story. They don't even have hardware in their network right now capable of doing this.
But, surely traffic shaping is done via port or traffic tag types (software based) rather than hardware?
Talkiet:jaysin: [snip]
The traffic shaping done by Spark was only on streaming and P2P ports also.
Browsing was fine, but streaming you tube and P2P / torrent connections were shaped down to miserable speeds.
Changed back to the capped 150GB plan, and this no longer occurred
You. Are. Wrong.
I was involved in the design of reporting system which we used to look at the amount of P2P and streaming traffic etc in the network. I was also involved in the decision not to implement the traffic management side of that solution.
It didn't happen. It didn't go into the network.
I'm 100% right and telling the truth on this from a position of intimate knowledge about the situation.
Something else was impacting you. (Did you have the same IP address on both plans? There were other changes a few weeks ago which improved performance across the board.
Cheers - N
My views are my own, and may not necessarily represent those of my employer.
jaysin:sbiddle:jaysin:
The traffic shaping done by Spark was only on streaming and P2P ports also.
Browsing was fine, but streaming you tube and P2P / torrent connections were shaped down to miserable speeds.
Changed back to the capped 150GB plan, and this no longer occurred
Spark do not and have not shaped traffic. End of story. They don't even have hardware in their network right now capable of doing this.
They did when I signed up with them last August (2014).
It was specifically stated on the plans terms and conditions, and was confirmed by the rep when I requested to go back to the 150GB data plan.
The plan has now changed obviously, as the disclaimer isn't noted on the plan on the website.
I should have mentioned this was nearly a year ago to clarify things. But, surely traffic shaping is done via port or traffic tag types (software based) rather than hardware?
My views are my own, and may not necessarily represent those of my employer.
cbrpilot: August last year, Spark did not shape traffic on Unlimited.
We have never shaped traffic on our current unlimited plan. The only plans that had traffic shaping were some time ago (Go Large and Big Time). They were from well well before August last year.
At the time Spark launched its Unlimited plan (not Go Large or Big Time) a clause was present in the terms and conditions so that Spark would have the right to shape if it viewed that this was necessary in the future. It was never done. The clause was subsequently removed from the Terms and Conditions.
If you were experiencing degraded throughput while on the unlimited plan, there must have been other factors at play. The unlimited plan had the same IP ranges as other plans. I.e. it was (and continues to be) completely identical in the network to all other plans (capped and overage). I.e. the network does not know which plan you are on.
There is probably no point in theorising as to what the cause of your poor speeds were during that period of time seeing as it no longer effects you and there is no way to prove or test any conjecture.
And yes, Taiket and I can speak very authoratively on this topic.
cbrpilot: There is no congestion of the Hamilton UFB (UFF) handover connection. The OP needs to test with a wired connection.
jaysin:cbrpilot: There is no congestion of the Hamilton UFB (UFF) handover connection. The OP needs to test with a wired connection.
There may be no congestion on the handover, as long as the customer base is not passing traffic higher than the E-NNI throughput of the handover the ISP has ordered.
But the end customer may have congestion due to the OLT ports throughput restriction, which is 2.5Gbps per second. This one OLT port will be feeding all customers on the cabinet's splitter. The OLT is completely seperate to the capability of the EAS, which is where the E-NNI is programmed for the ISP's handover connection.
This is why businesses generally pay for high CIR's. A business paying for a 10Mbps CIR will get more guaranteed throughput than a residential folks on a plan with a 2.5Mbps CIR.
Please note all comments are from my own brain and don't necessarily represent the position or opinions of my employer, previous employers, colleagues, friends or pets.
Talkiet:jaysin:cbrpilot: There is no congestion of the Hamilton UFB (UFF) handover connection. The OP needs to test with a wired connection.
There may be no congestion on the handover, as long as the customer base is not passing traffic higher than the E-NNI throughput of the handover the ISP has ordered.
But the end customer may have congestion due to the OLT ports throughput restriction, which is 2.5Gbps per second. This one OLT port will be feeding all customers on the cabinet's splitter. The OLT is completely seperate to the capability of the EAS, which is where the E-NNI is programmed for the ISP's handover connection.
This is why businesses generally pay for high CIR's. A business paying for a 10Mbps CIR will get more guaranteed throughput than a residential folks on a plan with a 2.5Mbps CIR.
You really shouldn't be trying to tell us how UFB works. We know in excruciating detail (more than you will ever ever ever want to know) how these things should work, do work currently, and might work in the future.
If it was likely that there was congestion on the OLT port we'd have thought of it by now. It's not likely. And your earlier split ratio isn't correct for the whole country either.
Cheers - N
jaysin: I was not trying to tell YOU how UFB works. I was trying to explain to the original poster what might be causing the issue. I did not address anyone else to begin with.
I thought the idea of such forums was for people to try to explain / help with other peoples issues. Not to interrupt such conversation to flex the testosterone of your I.T. knowledge.
The fact that I have worked extensively in the LFC space does give me some sort of knowledge about how UFB works.
However, I don't see the need to tell other people about my vast knowledge of UFB, and the details I know, as it won't make me sleep better at night, and I don't see the need in portraying a large ego on a geek forum.
So, if your aim here is to try to put other people, myself in this case, in their place with your protruding extensive phallic like knowledge of I.T. you have failed, sorry.
Please note all comments are from my own brain and don't necessarily represent the position or opinions of my employer, previous employers, colleagues, friends or pets.
Referral links: Quic Broadband (free setup code: R587125ERQ6VE) | Samsung | AliExpress | Wise | Sharesies
Support Geekzone by subscribing (browse ads-free), or making a one-off or recurring donation through PressPatron.
Please note all comments are from my own brain and don't necessarily represent the position or opinions of my employer, previous employers, colleagues, friends or pets.
|
|
|