PDA

View Full Version : Broadband Speed?

Pages : [1] 2

abh!shek
04-08-2008, 09:40 AM

Also tell me the difference between kbps and KB/s. My ISP tells me the speed in kbps but my download manager displays in KB/s so I'm confused with it and I think I'm not getting the speed I should get.

Thanks for the help.

SlyMaelstrom
04-08-2008, 09:47 AM
b = bit
B = byte

There are 8 bits to one byte. There for 800kbps (or 800kb/s) is the same as 100KB/s. So, since your time measurment is a factor of 8... if it take 8 hours to download N kilobytes, then it takes 1 hour to download N kilobits. The rest of the math you should be able to handle yourself.

...or you can just take the easy route and http://www.speedtest.net to get a significantly more accurate measurement.

indigo0086
04-08-2008, 09:49 AM
speedtest.net

abh!shek
04-08-2008, 09:52 AM
So do I calculate speed like this 700*100*8/(8*3600) ? But 19 kbps is very less....ISP says 256 kbps....are they fooling me?:mad:

abh!shek
04-08-2008, 09:54 AM
speedtest.net

I know that and I have no problem with the speed of my browser, I can surf at good speed, but my downloads are f slow. I think they are throttling my downloads.

SlyMaelstrom
04-08-2008, 09:59 AM

...anyway, a MB is not 100 times a KB, it's 1024 times... so the proper formula would be:

800*1024*8/8*3600 = 227.5kbps

Now... considering that I have a 15Mbps (with a 30Mbps burst) connection... I'm sure I wouldn't get much better speed than that downloading off of the same server at the same time.

abh!shek
04-08-2008, 12:33 PM
Now... considering that I have a 15Mbps (with a 30Mbps burst) connection... I'm sure I wouldn't get much better speed than that downloading off of the same server at the same time.

What is a burst..? Can it make my connection faster? And is there anyway to bypass my ISP's throttle-filter to get more speed without letting him know?

SlyMaelstrom
04-08-2008, 12:39 PM

abh!shek
04-08-2008, 12:50 PM
aaaaaw.....your speed is like 100 times more than mine (http://www.speedtest.net/result/256858861.png) :(

Mario F.
04-08-2008, 12:54 PM
Since we are on the subject, I'm often asked what's my connection speed when configuring software or using some web services. Usually I'm asking to pick from a list looking something like the following (concerning ADSL only): DSL 256Kb, DSL 512Kb, DSL 1.2Mb, DSL 2Mb

And... another thing that confuses me with ADSL is that since this type of connection doesn't suffer from the same problem as cable modems in which their speed gets slowed as new modems are added to a node, why is that my ISP announces my connection is 12Mb when all my speed tests never go higher than 500Kb download and 200Kb upload?

brewbuck
04-08-2008, 01:00 PM
Lots of broadband providers are starting to throttle P2P data rates. Some of them are even admitting that they do it.

abh!shek
04-08-2008, 01:05 PM
Lots of broadband providers are starting to throttle P2P data rates. Some of them are even admitting that they do it.
But I think you can get around it easily by encryption. Many torrent clients support encryption.

abachler
04-08-2008, 01:29 PM
They dont actually look at the data you are passing, just the hostname of the other end of the connection, if its another dynamic IP then they throttle. Encryption wont change the hostname.

SlyMaelstrom
04-08-2008, 01:33 PM
aaaaaw.....your speed is like 100 times more than mine (http://www.speedtest.net/result/256858861.png) :(I'm not embarrassed to say that I spend a healthy sum of money (and time) on my internet connection. However, the US is not too expensive compared to some other countries. Right now, I believe my household tosses about \$58/month for our connection.

Maybe if I get a big raise at work I can afford the \$150 50Mbps connection that Comcast will be offering soon. :devil:

abh!shek
04-08-2008, 01:44 PM
Maybe if I get a big raise at work I can afford the \$150 50Mbps connection that Comcast will be offering soon. :devil:

.....and I will sell my arms and legs:rolleyes:

BobMcGee123
04-08-2008, 01:46 PM
I've found that the only way to effectively deal with my internet problems is to kill myself.

:(

Mario F.
04-08-2008, 01:54 PM
hmm... maybe my question was too difficult.

whiteflags
04-08-2008, 02:36 PM
hmm... maybe my question was too difficult.

Pick the one that matches your upspeed and downspeed the closest. Web services and stuff like proxies may also have to configure itself to upload your items to the internet in the most economical way.

And modem potential is better advertising, but if nobody else was sharing your bandwidth, you would be going that fast.

Mario F.
04-08-2008, 02:52 PM
Hard to do since these ADSL connections are asymmetric and my upload speeds are less than half the donwnload speed. I'm just not used to them since I've always used symmetric cable modem connections and only recently switched to ADSL.

But I guess it's most of the time asking about download speeds, unless i'm configuring something that will act as a server... thanks.

What about this thing with ISP announcing I have a 12Mb ADSL when my speed is in fact 500Kb/200Kb? I see this everywhere. Not just with my ISP.

SlyMaelstrom
04-08-2008, 03:52 PM
Usually Mario, when they ask you for that speed, they're looking for your downspeed. By the way, most Cable internet (if not all) is asymmetric as far as I know... just look at my down speeds and up speeds.

Mario F.
04-08-2008, 04:19 PM
By the way, most Cable internet (if not all) is asymmetric as far as I know... just look at my down speeds and up speeds.

They are indeed. But only marginally when compared to ADSL which has an usual ratio of 2:1 in favor of downstream. Cable modem have rather more close ratios of 1.3:1 and even 1.2:1

matsp
04-08-2008, 04:39 PM
Or even 6:1 (1779:311 measured, I think the official number is 2MB/512KB which is 4:1)

--
Mats

SlyMaelstrom
04-08-2008, 06:31 PM
Mine is nearly 23:2.

Mario F.
04-08-2008, 06:42 PM
Different experiences then. I've ran through two version of DOCSIS; 2.0 and 3.0, during about 7 years of cable access. Both offering very close ratios.

Maybe an european thing... or specific to this country service. Even though the same was true of my Australian experience - I believe DOCSIS 2.0.

SlyMaelstrom
04-08-2008, 06:47 PM
Different experiences then. I've ran through two version of DOCSIS; 2.0 and 3.0, during about 7 years of cable access. Both offering very close ratios.

Maybe an european thing... or specific to this country service. Even though the same was true of my Australian experience - I believe DOCSIS 2.0.It's hard to say with cable, but it's quite possible that they are throttling it the same way the throttle the up-speed on FIOS in the US. I know for a fact that with the same line, Verizon can offer either 15d/2u or 15d/15u for different prices.

Mario F.
04-08-2008, 07:18 PM
It's hard to say with cable, but it's quite possible that they are throttling it the same way the throttle the up-speed on FIOS in the US. I know for a fact that with the same line, Verizon can offer either 15d/2u or 15d/15u for different prices.

Actually you just reminded me of an important thing... I've always requested for static ip access. That almost surely explains it. Hmm...

DavidP
04-08-2008, 10:33 PM
Weeeeee....I scored quite well. :)

http://www.speedtest.net/result/257046236.png (http://www.speedtest.net)

brewbuck
04-08-2008, 11:49 PM
Weeeeee....I scored quite well. :)

http://www.speedtest.net/result/257046236.png (http://www.speedtest.net)

Oh yeah?!?!

http://www.speedtest.net/result/257060781.png (http://www.speedtest.net)

17 Mbps is... somewhat more than I used to get. When I started using computer networks, I used a 2400 bps modem.

It's actually funny how I don't really notice a difference in speed of almost 7400 times. Gives you an idea of how much "bloat" is inherent in the Internet. 2400 bps used to be juuust fine. Then again, I wasn't downloading videos or music like everybody does today.

(I'm not trying to sound old -- just got started early :) )

abachler
04-09-2008, 12:18 PM
When I started using computer networks, I used a 2400 bps modem.

Had to get the Ma Bell's permission to hook up my 300 :D

dwks
04-09-2008, 02:35 PM
Okay, this is just weird. My upload speed (6798 kb/s) is over four times faster than my download speed (1684 kb/s)!

(In case anyone's wondering how a dial-up user got such high values: this isn't my connection. It's a public wireless one.)

mike_g
04-09-2008, 02:47 PM

dwks
04-09-2008, 02:53 PM
Erm yes, you're right. I got it backwards. Oops. :)

Mario F.
04-09-2008, 03:42 PM
Hmm... that's odd indeed.
MTU Settings? Try to check/fix those and probably also reset your modem to factory default settings. That download speed is already very good though. And with that upload speed, I would probably just leave it as is. That's server world for you :D

As for my Internet history... I was kind of dragged into it kicking and screaming. I was completely into BBS for many years before that, having ran two as sysop. The internet was invading my so far beautiful world and I really hated it :)

It was in 1995 that I first connected to the internet from my home. It was in 1992 I think that I first got contact with it on the university. During these 2 or 3 years I was forced to work on it (there was high demand for HTML coders and, despite all, I was curious about this markup language), but when coming home I would simply get back into my BBS world only to see my subscribers slowly dwindling away and many other BBSes losing their sysops.

Elysia
04-09-2008, 04:10 PM
Oh my, you've just broken a traditional rule.
Congratulations! :D

My connection sucks :p

TEST_DATE: 09/04/2008 10:11 PM GMT
LATENCY: 25 ms
DISTANCE: ~ 600 km

zacs7
04-09-2008, 04:12 PM
Just TELUS dwks!

Pun intended ;)

dwks
04-09-2008, 05:35 PM
Hmm... that's odd indeed.
MTU Settings? Try to check/fix those and probably also reset your modem to factory default settings. That download speed is already very good though. And with that upload speed, I would probably just leave it as is. That's server world for you :D
Well, like I said, it isn't my connection. I seriously doubt they would listen to anything I said, especially if it had phrases like "MTU settings" in it. ;)

DISTANCE: ~ 600 km
Pick a closer server, if there are any. I tried a server that was really far away and my speed went way down.

Just TELUS dwks!
It would be funny if it was actually serviced by Telus. I don't know if it is or not.

Anyway, here I am on a different internet connection. This one's better. ;) See attached.

Mario F.
04-09-2008, 06:31 PM
North Americans showing off. bah!

whiteflags
04-09-2008, 09:39 PM
Don't feel too bad Mario. My connection is pretty slow even by my own country's standards. I even used dialup up to two years ago.

http://www.speedtest.net/result/257478686.png (http://www.speedtest.net)

And other people in the world beat the U.S. (which I am aware is only part of North America, thanks) at all sorts of things, including broadband internet penetration, by the way. So many Americans still use over-the-air analog televisions that our government has to launch a significant readiness program (http://www.hdtv.gov)to prepare people for the time when domestic analog broadcasting ends. The world did this decades ago. :D

Elysia
04-10-2008, 12:41 AM
Pick a closer server, if there are any. I tried a server that was really far away and my speed went way down.

I would if I could. No closer server, or if there was, not much closer.
And testing servers inside your own country boosts speed more than trying one outside.

heras
04-10-2008, 01:24 AM
but if nobody else was sharing your bandwidth, you would be going that fast.
I doubt that that is the case. In stead I think it has to do with the fact that DSL speeds degrade sharply (http://my.opera.com/boxopen/blog/show.dml/57260) with increasing distance to the phone exchange.

whiteflags
04-10-2008, 01:44 AM
And testing servers inside your own country boosts speed more than trying one outside.

Why would that be an issue? Servers are pretty dense around the world and most requests pass through domestic servers anyway. Do you know where the servers for the sites you visit most often are located?

I doubt that that is the case. In stead I think it has to do with the fact that DSL speeds degrade sharply with increasing distance to the phone exchange.

That was not the point.

Elysia
04-10-2008, 01:46 AM
All I know is I tried 3 different servers and the one in my own country yielded better results than those off-country, despite being farer off into the distance.

heras
04-10-2008, 02:01 AM
That was not the point.
Ok. I thought you were suggesting that not reaching advertised speeds was due to other users on the networks as an only factor. My bad, probably.

Elysia
04-10-2008, 02:09 AM
Yep, speed can degrade on several factors, including how many are using a single line (typically fiber or cable), how far it is to your station (DSL), slow servers, speed throttling, etc.
DSL is especially horrible since the speed degrades sharply with distance. ADSL2+ is supposed to max out at 24 mbit / 1 mbit, but can go way lower if the distance to the station is a far way off. So they may state correctly, or advertise correctly, but due to how it works, you might not get everything, but it isn't the ISP's fault.
It's a terrible world...

abachler
04-10-2008, 08:13 AM
Its false advertising IMO, same as HDD's that advertise 500GB when in fact they only have 465, but they choose to use fuzzy math and say they have 100 billion bytes, thats a 100 gigabyte, which it isnt. Same with ISP's, they advertise the maximum theoretical speed, even though 99&#37; of their lines wont sustain that speed.

abachler
04-10-2008, 08:17 AM
All I know is I tried 3 different servers and the one in my own country yielded better results than those off-country, despite being farer off into the distance.

Map distance is irrelevant, its line distance that determiens it.

laserlight
04-10-2008, 08:21 AM
Its false advertising IMO, same as HDD's that advertise 500GB when in fact they only have 465, but they choose to use fuzzy math and say they have 100 billion bytes, thats a 100 gigabyte, which it isnt.
The problem is, they are correct. 100 billion bytes is 100 gigabytes (GB), by conventional SI unit reckoning. Now, if they claimed a capacity of 100 GiB for an actual capacity of 100 billion bytes, that would be a different matter altogether.

abachler
04-10-2008, 08:41 AM
The term GiB was added after the fact to 'resolve' this issue. GB was always 1048576 KB (KiB by the new 'standard') when dealing with computers, just as the term calorie in nutrition is known to mean kilocalorie. The adertising retards were just takign advantage of the jargon used in 2 disciplines having a different meanign in laymans terms. Naturalyl they chose the one that made their product look better. This was not always teh case, fro many years they correctly used the proper values and only switched when the encountered a client base that was not saavy enough to know the difference.

If you say kilobyte, anyone who is a programmer automatically thinks 1024 bytes, not 1000 bytes. The term kilo when applied to computer math was borrowed from the metric system, but it is not in fact the same. Or at least ti wasn't until the retards at SI cow towed to industry and made the distinction. IEEE never made such a distinction afaik, wqhich is why industry prefers SI, and engineers prefer IEEE.

laserlight
04-10-2008, 09:29 AM
If you say kilobyte, anyone who is a programmer automatically thinks 1024 bytes, not 1000 bytes.
That's true, but the market consists of more than just programmers and other engineers ("a client base that was not saavy enough to know the difference"). Consequently, they can defend their choice of units, even though it is intentionally misleading.

abachler
04-10-2008, 09:38 AM
Consequently, they can defend their choice of units, even though it is intentionally misleading.

To the end user yes, I don't think we as programmers and engineers should encourage the use amongst ourselves. I don't use the term GiB, and I generally look with disdain upon those that do, unless it is specifically refering to the difference betwene the two. In my life at least, the people I socialize with are mostly either engineers themselves, or educated enough to know the difference. Granted the drooling masses neither know nor care as long as they can surf the web for 'movies', so then why even bother having seperate terms?