Turn on thread page Beta

Internet speed goes up if you run a speed test watch

    • Thread Starter
    Offline

    16
    ReputationRep:
    Like does anyone else find that when you download something, it's slow. Then if you also run Speedtest on speedtest.net. Then after the test finishes, your download speed for the thing you were downloading have gone up massively?

    This isn't restricted to one ISP either. The internet connections for 3, EE, Virgin Media have all exhibited that behaviour when I've tested it.

    It seems like the VW cheat device or something, when it knows it's being tested puts on a good show then when it's over dial the speeds down again? I mean, that's just ****ing ridiculous.

    Anyone else notice that?
    Offline

    8
    ReputationRep:
    interesting. never heard of this before and i couldnt find anything about this. have you tried asking the ISP? what test was it?
    Posted on the TSR App. Download from Apple or Google Play
    • Section Leader
    Offline

    20
    ReputationRep:
    Section Leader
    (Original post by nonotyoutoo)
    Like does anyone else find that when you download something, it's slow. Then if you also run Speedtest on speedtest.net. Then after the test finishes, your download speed for the thing you were downloading have gone up massively?

    This isn't restricted to one ISP either. The internet connections for 3, EE, Virgin Media have all exhibited that behaviour when I've tested it.

    It seems like the VW cheat device or something, when it knows it's being tested puts on a good show then when it's over dial the speeds down again? I mean, that's just ****ing ridiculous.

    Anyone else notice that?
    Hi I have moved this to the technology and computer forum where you might get more specific advice. I’ve not personally noticed this
    Offline

    12
    ReputationRep:
    Say you bought a broadband package of 50mb/s down and 10mb/s up. Now you run a speedtest say on speedtest.net and get those exact results. But then you decide to download something online but it's only downloading at 5mb/s. Now this is because of how bandwidth works. You are actually recieving around 50mb/s which translated to about 5mb/s down when downloading. The general rule of thumb is you'll download at 10% of your quoted speed. Now to prove this. On windows, download something online, during the download open task manager and click performance go to your ethernet or wifi adapter and look, It will tell you how many mb/s you're recieving and it should match your quote or come close.

    Now to focus on what you're talking about, downloading and running a speedtest and the same time will split your bandwidth, alongside doing other tasks using internet resourses. Your computer will automatically decide what needs more bandwidth and other things will slow down. After the speedtest finishes, bandwidth is relocated and your download speed will increase.

    Hope this explains whats going on.
    Posted on the TSR App. Download from Apple or Google Play
    • Community Assistant
    Offline

    20
    ReputationRep:
    Community Assistant
    (Original post by Daniel00)
    Now this is because of how bandwidth works. You are actually recieving around 50mb/s which translated to about 5mb/s down when downloading. The general rule of thumb is you'll download at 10% of your quoted speed.
    I'm not really sure where you got this information from but it's not correct. Downloads are not generally capped at 10% of your quoted speed.

    What's actually happening nonotyoutoo is a number of different things. The primary things that will affect your download speed are:
    1. The speed of your connection (i.e. the broadband package you actually have)
    2. The state of the network at that time
    3. The limitations of the server you are downloading from and the network it is connected to

    The first one is obvious how it affects download speed. If you have a 50Mb (that's 50 Megabits, not Megabytes) connection then the basic cap for your download speed is that 50Mb. That translates to 6.25MB and this is where I suspect you might be getting the roughly 10% thing from.

    The second point, state of the network, is also fairly obvious. Companies throttle speeds and there are peak times when people are on the network. During those times, you are more likely to get slower download speeds. If you were to run speedtest at different times in the day you would see different speeds, purely because of this.

    The third point is the more likely answer to this question. When you download something, you don't just automatically start downloading at the maximum speed. Just like how an ISP imposes restrictions on you, the server at the other end will have restrictions placed on it as well. This is overly simplified but if you have a package from your ISP that allows for 50Mb/s but the server is set up to allow 20Mb/s for any one client then no matter how much you try, you won't ever get speeds above 20Mb/s.

    So why does speedtest present different speeds compared to regular downloads? Because speedtest is designed as a service to test connection speed. It'd be a little pointless if speedtest servers were unable to saturate a network connection so speedtest doesn't have that cap in place. Therefore a service like speedtest will get close to your advertised network speed, providing things like ISP throttling allow it.

    Why do download servers not offer the same? Lots of things, primarily linking back to cost. I imagine if we all had Virgins 300Mb/s package and all tried downloading something at once. The company that owns the server would need a lot more infrastructure to deal with that. Thousands of people connecting at 300Mb/s for a download would saturate gigabit and 10 gigabit network connections very quickly. Since there's no real need for that, companies limit the speeds available to people trying to download things.

    A pretty cool example of this was when Google changed their logo. The new logo was a much smaller file. This meant they regained a lot of bandwidth that was previously lost to sending people the more complicated Google logo.

    So the tl:dr of this is that: Speedtest is designed to saturate a connection and test speed. The company you are downloading stuff from does not want you connecting at the maximum speed available to you, they need to control it so as not to overload their server and network. Speedtest tends to grab the nearest server too so it is less affected by ISP throttling and busy networks.
    Offline

    12
    ReputationRep:
    (Original post by Acsel)
    I'm not really sure where you got this information from but it's not correct. Downloads are not generally capped at 10% of your quoted speed.

    What's actually happening nonotyoutoo is a number of different things. The primary things that will affect your download speed are:
    1. The speed of your connection (i.e. the broadband package you actually have)
    2. The state of the network at that time
    3. The limitations of the server you are downloading from and the network it is connected to

    The first one is obvious how it affects download speed. If you have a 50Mb (that's 50 Megabits, not Megabytes) connection then the basic cap for your download speed is that 50Mb. That translates to 6.25MB and this is where I suspect you might be getting the roughly 10% thing from.

    The second point, state of the network, is also fairly obvious. Companies throttle speeds and there are peak times when people are on the network. During those times, you are more likely to get slower download speeds. If you were to run speedtest at different times in the day you would see different speeds, purely because of this.

    The third point is the more likely answer to this question. When you download something, you don't just automatically start downloading at the maximum speed. Just like how an ISP imposes restrictions on you, the server at the other end will have restrictions placed on it as well. This is overly simplified but if you have a package from your ISP that allows for 50Mb/s but the server is set up to allow 20Mb/s for any one client then no matter how much you try, you won't ever get speeds above 20Mb/s.

    So why does speedtest present different speeds compared to regular downloads? Because speedtest is designed as a service to test connection speed. It'd be a little pointless if speedtest servers were unable to saturate a network connection so speedtest doesn't have that cap in place. Therefore a service like speedtest will get close to your advertised network speed, providing things like ISP throttling allow it.

    Why do download servers not offer the same? Lots of things, primarily linking back to cost. I imagine if we all had Virgins 300Mb/s package and all tried downloading something at once. The company that owns the server would need a lot more infrastructure to deal with that. Thousands of people connecting at 300Mb/s for a download would saturate gigabit and 10 gigabit network connections very quickly. Since there's no real need for that, companies limit the speeds available to people trying to download things.

    A pretty cool example of this was when Google changed their logo. The new logo was a much smaller file. This meant they regained a lot of bandwidth that was previously lost to sending people the more complicated Google logo.

    So the Tlr of this is that: Speedtest is designed to saturate a connection and test speed. The company you are downloading stuff from does not want you connecting at the maximum speed available to you, they need to control it so as not to overload their server and network. Speedtest tends to grab the nearest server too so it is less affected by ISP throttling and busy networks.
    Im not saying its capped. But to explain in more depth, I will admit that bandwidth allocation flaws the system and does cause issues, but were confusing ourselfs between the different of Megabytes and Megabits. Our quoted speeds are in Megabits (MB). We download in Megabytes (Mb). So 50MB roughly translates to 6Mb which is just over 10% so thats why I use 10% as a rule of thumb. This is where most of the errors are coming from. Now a days there's very little throttling even at peak times so its almost all just confusion between the two quantities.
    Posted on the TSR App. Download from Apple or Google Play
    • Community Assistant
    Offline

    19
    ReputationRep:
    Community Assistant
    (Original post by nonotyoutoo)
    Like does anyone else find that when you download something, it's slow. Then if you also run Speedtest on speedtest.net. Then after the test finishes, your download speed for the thing you were downloading have gone up massively?

    This isn't restricted to one ISP either. The internet connections for 3, EE, Virgin Media have all exhibited that behaviour when I've tested it.

    It seems like the VW cheat device or something, when it knows it's being tested puts on a good show then when it's over dial the speeds down again? I mean, that's just ****ing ridiculous.

    Anyone else notice that?
    not really, in fact my laptop shows sppedtest as slower than actual speed (task manager shows faster than speedtest.net or fast.com)

    I would say for download speed, fast.com is accuarate for myself but speedtest shows download + upload.

    I should point out that generally VPNs slow down the speed of your browser from my experience but dont show on sppetest or fast.com
    Offline

    12
    ReputationRep:
    (Original post by quasa)
    not really, in fact my laptop shows sppedtest as slower than actual speed (task manager shows faster than speedtest.net or fast.com)

    I would say for download speed, fast.com is accuarate for myself but speedtest shows download + upload.

    I should point out that generally VPNs slow down the speed of your browser from my experience but dont show on sppetest or fast.com
    VPNs will soon be faster than normal speeds if the US decide to slow down connections due to new net neutrality conditions, so much for freedom.
    Posted on the TSR App. Download from Apple or Google Play
    • Community Assistant
    Offline

    20
    ReputationRep:
    Community Assistant
    (Original post by Daniel00)
    Our quoted speeds are in Megabits (MB). We download in Megabytes (Mb).
    Your prefixes are the wrong way round. Megabits is Mb, Megabytes is MB. The easy way to remember it is that bits are smaller than bytes, use a lower case b.


    (Original post by Daniel00)
    So 50MB roughly translates to 6Mb which is just over 10% so thats why I use 10% as a rule of thumb. This is where most of the errors are coming from.
    It's not 10% though, it's literally the same thing. 50Mb is the same as 6.25MB. The value 6.25 is an eighth of 50 but the actual numbers are the same.

    It's also not the cause of the discrepancy. Windows 10 Task Manager (since you suggested it), speedtest.net and ISPs all list the speeds in Mbps. There's no conversion necessary.

    The only time conversion is required is if you are looking at your browsers download box or the file size.

    (Original post by Daniel00)
    Now a days there's very little throttling even at peak times so its almost all just confusion between the two quantities.
    Again, completely incorrect. Throttling is very much a real problem. Virgin Media are notorious for it and at peak times will not offer anywhere close to the advertised speed. During peak times, network operators will throttle speeds if it becomes too much for the network to handle. It's a fairly basic principle of networking really.

    As for whether it is purely down to not knowing the difference between MB and Mb, that depends on factors we don't know. If the OP has a low level package then odds are a download server will be able to support a connection with them at the maximum available speed they can handle. If they have a more premium package then it's far less likely.

    Regardless, that doesn't change the fact that your actual speeds depend on what you are paying for, what you are actually getting and what the server can offer. As I originally said, services like speedtest are designed to saturate the network. Servers are designed to avoid that. Therefore there will always be discrepancies between speedtest services and actual available speeds. I can easily get gigabit speeds showing up on speedtest while at uni. I would never actually get gigabit download speeds for a ton of reasons though (wireless cards don't support it, the speedtest server is local, uni has gigabit backbone but the ISP connecting us to the internet will not offer those same gigabit speeds to individual users and so on).
 
 
 
Reply
Submit reply
Turn on thread page Beta
TSR Support Team

We have a brilliant team of more than 60 Support Team members looking after discussions on The Student Room, helping to make it a fun, safe and useful place to hang out.

This forum is supported by:
Updated: February 18, 2018
Poll
How are you feeling in the run-up to Results Day 2018?

The Student Room, Get Revising and Marked by Teachers are trading names of The Student Room Group Ltd.

Register Number: 04666380 (England and Wales), VAT No. 806 8067 22 Registered Office: International House, Queens Road, Brighton, BN1 3XE

Write a reply...
Reply
Hide
Reputation gems: You get these gems as you gain rep from other members for making good contributions and giving helpful advice.