Satellite Speed checks

Ok Newbie here so don't be harsh We recently installed an Avanti broadband system with headline speeds of 15/2 and it doesn't feel anything like. Here is my issue, when we run speed tests on the Avanti portal the speed looks quite good, in the areas of 10/1 but when we run the speed test anywhere else on generic speed tests it reads more like it feels . sub 2Mbps.
When I have asked why this is the supplier says its to do with the "turbo" and latency not being recognised on the generic speed checker. OK, I understand latency and I understand the accelerators the platform uses in the hope to make the internet over satellite feel usable. BUT IT ISN'T IT IS A BAD USER EXPERIENCE. What I now don't understand is I have a friend on the tooway platform and when ever we run speed tests on this platform on thier speed test or the genric they read identical??? Doesn't tooway on the KA band employ the same accelerator systems.. I believe they do. Any Advice? Is Avanti doing a VW??Not providing real speed test results. I am later today going to trial some real FTP and time it ! Let me know your thoughts and Thanks

Without having their speed test to dissect hard to say, can you post a link to a test done via http://www.thinkbroadband.com/speedtest and also https://labs.thinkbroadband.com/speedtest

The second forces things over SSL so should be a better end to end test, since it should be encrypted between browser and our server.

Caching/acceleration techniques can improve the web browsing experience on satellite, but with the growth of in page refresh technology this is negating some of those tricks. Our test certainly identifies latency on satellite connections very well and download tests are pretty good, upload seems to suffer but then suspect some actual real world upload techniques would suffer similarly.

The world of satellite reselling is complex and there can be issues even if using the same satellite as another provider with regards to capacity the package buys and slower packages are often lower priority ones.

I have Tooway, and my experience is similar to yours. It is supposed to be 20Mb down and 6Mb up. They have a test which you can run that shows that this is what it delivers, and it is true, it can deliver that, technically.

But the reality of using the service is quite different. The available bandwidth has been oversold, so the contention rates are incredibly high. You are competing with too many other users for the limited bandwidth.

Here is how you can prove that this is the issue, and it has nothing to do with their ridiculous excuses about the problem being in other people's speed checkers. Use the thinkbroadband speed checker (and your other favourite checker) at about the following times:
1. When you get up in the morning.
2. Late afternoon or as soon as you get home.
3. 10pm (peak time).
4. This is the painful one, but the most important. Set your alarm clock. 3am.
I bet you will find that the service at 3am is pretty good.
Please keep a log and tell us what you find.

Satellite is the solution of last resort for the consumer. I expect if one pays for a business service one can get it less contended, but this will be far far more expensive. I have found that if one can get a landline service giving just 1Mb it is preferable over Tooway, except during the middle of the night.

I am looking forward to getting rid of my satellite service.

It's the same story with the SES service too. I agree with growltiger to try the service at 3am because that will help determine if your daytime problem is congestion (most likely) or if its just something more mundane like a bad cable connector.

As regards the 'sub 2Mbps' feel, that is because of the high latency of satellite signals. Satellite is at its best when downloading large files and at its worst when downloading lots of small files. Unfortunately most web pages are built up as a series of elements which have to be separately downloaded and then put together to make the page work, and because each element is effectively a small file then you can easily end up with eight or ten small downloads, each one with 750ms or so of latency, and so pages seem to take forever to load.

One way of reducing page load times is use browser add-ons like Adblock because ads are usually downloaded as separate elements, and the more elements you can eliminate the faster the page load will be. It also reduces the awful visual clutter that ads cause. Increases security too now that some ads are being used to deliver malware.

The speed test at http://labs.thinkbroadband.com/speedtest is deliberately ad free to make testing on weird and wonderful devices easier.

From looking at tests people with satellite carry out, the difference for many between 11pm and 2am is so big that only massive congestion or deliberate traffic management can be the cause.

In reply to a post by MrSaffron:

From looking at tests people with satellite carry out, the difference for many between 11pm and 2am is so big that only massive congestion or deliberate traffic management can be the cause.

Yes, I agree. And proof that satellites don't really have the capacity to support unlimited download packages, as Tooway found to their cost a couple of years ago.

Pity really because when used out of peak hours then you see what satellite is really capable of.