upload/ download speeds.

can anyone explain this?

i have a 10 megabit connection.
when i test this connection at 2 different sites i usually get widely differing results.
the first site, testmy, gives a current reading of 5.6.
the other site, speakeasy, has a number of different servers to choose from.
testing at the speakeasy site using all the servers gives anywhere between 16 and 27 megabits.
widely fluctuating yes?
but, that isn't really what bugs me.
it's my upload speed.
i believe my upload is supposed to be 1.5 megabits.
after performing the download tests at the above sites and getting widely fluctuating results i made the upload tests.
now, get this:
i get a CONSISTENT 0.93 to 0.95 upload speed.
what's the story here?

 

Log in or Sign up to hide all adverts.

It is probably dropping off due to the sustained data rate of the test. Could be on your end though. bottelneck at you puter:

http://www.makeuseof.com/tag/technology-explained-understanding-speed-the-internet/

 

Log in or Sign up to hide all adverts.

That's normal for a home connect...even fiber to the house is not given it's own QoS I don't think. Basically they are always talking max throughput in ideal conditions for your connection. (no one else on your main switch/router is doing anything). They can divided it many ways.

To get even 10meg dedicated QoS is like hundreds of dollars a month.

 

Log in or Sign up to hide all adverts.

I've read that transfer protocols on ADSL and DOCSIS transfer uploads and downloads at different frequencies 6-48MhZ (UP) vs 54-1000MhZ (DOWN) this is done to (I believe) prevent crossing signals but has the obvious down-side of bandwidth reduction.

Judging by the fluctuation I am going to guess you have a DOCSIS connection (Cable). Your modem is designed to throttle you based on a specific flag sent by your ISP. If your modem rates by frequency without accounting for latency...and you have high latency due to your locale (maybe on the periphery of a hub?) you should see reduced real bandwidth.

Throttling incoming data is more complicated as it cannot be throttled at the source and variations need to be made to the operating frequency to compensate, which is why you'll likely see higher variation. I will say... this seems to be less true on ADSL lines which I cannot explain.

 

Connection analysis websites can be misleading in the sense that they aren't just measuring the quality of your network connection to your ISP, they are also measuring the quality of the connection between your ISP and whatever server is performing the analysis. It's kind of like trying to measure the top speed of your vehicle in conditions where there's no guarantee that the roads you're driving on will be free enough of traffic for you to be able to reach it, or whether there will be imposed speed limits/roadworks/detours and such.

When I test my own connection, I usually do both of the following: 1) download a large file directly from my ISP's local mirror and 2) initiate a multi-threaded download of a large file (see download acceleration) from a couple of different sites. I don't even bother with internet speedtest sites. They're close to useless if you ask me (and cause all sorts of confusion among people who don't understand exactly what the internet is and how it works).

 

I find that depending on the link I'm downloading from the speed will fluctuate greatly due to the amount of people that are downloading from that site even though I am supposed to have a 3 Meg download time I usually get less than a meg.

 

Leopold you're so lucky to have asked you question here, practically everyone is an export.

 

speedtest.net
pingtest.net

Giver.

You will never get 10meg download or 1.5 upload...probably not even to your own ISP FTP servers. Not without a QoS Router right in your house.

 

you are the only one that gave a plausible answer to the question.
yes, i have cable from the wall outlet but my ISP says their network is fiber.

i've scored well over the advertised download speed on numerous occasions.
i got a 51megabit download last night from the speakeasy site which was off the scale.
the upload for this score?
0.94

 

well, i check at the above site.
results:
download 41.22 megabits.
upload 0.95 megabits.

 

lol you don't have a 10 meg download throughput limit then do you? You probably shouldn't complain to your ISP if you are paying for 10meg lol

 

they were going around offering incentives to join up.
stuff like 3 months of reduced prices, a month free service, free this and that.
they knocked on my door and went into their pitch.
i wanted my net back so i told them forget all the free stuff just jack me up and tell me what i gotta pay each month.
boom, done deal.

also, i am not running the usual windows configuration.
oh yeah, i HAVE windows, but i've been toying around with a configuration utility called autoruns.
this in itself could have a lot to do with this.

edit:
use caution with the autoruns utility.
i've crashed my OS to the point i had to reinstall.
i can usually get into safe mode to undo the changes but a couple of times i couldn't.
so beware.

 

Leopold, download speed is NEVER constant. It depends on so many things:

-time of day
-other users (load) of the website
-physical distance of the website
-your browser
-your PC running other programs
-your IPS and their load
-etc.

So you will get different results just by checking the speed in every few minutes. Your ISP advertises the highest speed achievable under the best circumstances. In real life you should be happy if you get up to 80% of that...

 

I got an XP install down to .99gb. Full boat with installded proggies like photshop etc.

Funny thing is my media player is missing the images for the interface so it shows broken image icons. Sitll works fine though.

I did cheat and use XPLite and a reg cleaner mdae by a chap in NZ that opens with a page peel but for the life of me i cannot remember the name of that program. Best reg cleaner ever made though. Some reg cleaners actually put a few entries in there upon install to show them as deletable when you run the scanner. That is my own observation, i have no solid evidence though. Other than that i trimmed it down manually.

Please Register or Log in to view the hidden image!

 

john99
if you realized how retarded "puter" and "proggies" sound you would NEVER use those words again.

 

i know.

Please Register or Log in to view the hidden image!