I had no idea some internet connections were so fast! Can you guys get, like the WHOLE internet downloaded in an hour?
I think there is such a thing as TOO fast!
My theory only said that OSX screws up the test, making the speed appear 50 times faster than it really is, not that there were any actual differences.
Evidence so far:
rowrrbazzle, are you using OSX?
drm, are you using the standard browser?
The Manxnet test gave me:
I thought that looked slow and it also didn’t mesh with my bandwidth meter I have running on my desktop (which shows current speeds at an average of 1401 KB)
I headed over to www.dslreports.com and they gave me a reading which matches what I’ve been seeing.
I don’t think manxnet is correct.
Your line speed is:
953.3 Kbps
116.8 K bytes/sec
Comcast Broadband.
**709.6 Kbps
87 K bytes/sec
Verizon DSL**
Seems a bit slow but I used to have an AT&T Cable modem and I’d say it is about the same.
Prolog Cable, NE PA, Mandrake Linux 9.2, Mozilla 1.4 (home)
2566 Kbps
314.5 K bytes/sec
Now that I’m at home: (WinXP, Earthlink Cable, Mozilla FireFox)
However, using the speed test available at www.midsouth.rr.com (which is a local server to me) gets me:
What a difference a choice in servers make!
Safari does appear to mess up the test. Looks what happens when I test using KDE’s Konqueror (which Safari was based upon, then had some of Apple’s modifications built into it.)
Your line speed is:
151111.1 Kbps
18518.5 K bytes/sec
I always thought Kbps WAS kilobytes per second.
dutchboy, nope, kiloBITs per second, one eighth as much. The correct way is to capitalize Bytes, and use lowercase for bits. In practice even the most prominent websites fuck it up.
Thank you, Chief Crunch. So then it’s Linux’s fault, not Apples! 
Not that I really give a rat’s ass, it’s just fun for me to get to the bottom of it.
Your line speed is:
1642.5 Kbps
201.3 K bytes/sec
Insight broadband (cable).
Little ‘b’ is bits, big ‘B’ bytes, and there are 8bits per byte of course.
Still waiting to have this one answered 
I know pages will load quicker in certain browsers but I always thought that was because they were streamlined: ie. had minimal plugins and used the original formatting as much as possible and all that.
I’m sorry, I could have answered that earlier, but I stupidly hypothesized that it was the operating system, and not the browser, which was causing the inconsistencies within this thread.
Yes, the browser can make a difference. In general, however, the differences would be limited to overhead loading the browser into memory in the first place, ie. 5 seconds vs. 45; or how quickly the browser can download many small files at once, ie webpages.
There are really two different criteria to define your functional internet speed: latency and bandwidth.
If you open a command prompt, and type “ping yahoo.com” and “tracert yahoo.com” this will give you an idea of your latencies. My pings to Yahoo usually hover between 75 and 250 ms. Hopefully you have less than 10 or 15 hops in your trace route.
For a pure speed test of your bandwidth, however, you should really send a much larger file or send several large files and average the speeds together. The proprietary test in the OP is not the best IMHO. As others have mentioned, www.dslreports.com is a much better gauge of speed.
For now, all I can conclude is that the default browser, Safari, which is shipping with Macintosh OS X, which is based on an offshoot of the open source linux KDE Konqueror; and the Manxnet test linked in the OP, cannot communicate properly.
The inconsistencies reported could be due to programming errors (read: not up to spec) either with the website or the browser.
I did another test (speakeasy @ dslreports.com) and got slightly different results:
2004-02-14 07:49:05 EST: 2612 / 321
Your download speed : 2612876 bps, or 2612 kbps.
A 318.9 KB/sec transfer rate.
Your upload speed : 321207 bps, or 321 kbps.
Seems like broadband … above the 1mbit barrier!
Again, using RoadRunner broadband via Brighthouse cable/ Win XP and IE 6.0.2
Roadrunner is our cable provider, bnth computers are attached to the cable modem via a Netgear Linksys router.
My PC (Dell latitude) with Win XP using internal centrino for wireless connnection to router:
With Mozilla:
Test 1: Line Speed is 302.9 kbps
Test 2: 1725.9 kbps
Test 3: 1787.1 kbps
I think the first test may have been affected by the initial page load, which seems much slower on Mozilla.
With IE:
Test 1: Line Speed is 1010.9 kbps
Test 2: 977 kbps
Test 3: 1669 kbps
Her iMac with OSX with hardline to router:
1656.4 kpbs (this was more or less the mean of three different results, ranging from over 2000 down to about 800).
Do you think the Safaria browser skews the test, or is just a better browser (as Apple claims)? Per their website:
I’ll have to set up the Safari browser on our Mac and see if it affects test speed or apparent speed at all.
So I used the Safari browser on the Missus’ iMac, and the result (everything else the same as the post above) was:
92,727 kbps
It is noticeably much faster, although I think that both are probably overstated (results for IE and Safari). But we’ll just use the Safari browser, which came with our 10.3 upgrade, over the old IE browser. Plus there’s the added benefit of one less Microsoft product in use.
Your line speed is:
502.5 Kbps
61.6 K bytes/sec
Verizon DSL, IE6, Win98.
541.8 Kbps
66.4 K bytes/sec on adsl
Damn I want cable.
2559.6 Kbps
313.7 K bytes/sec
Adelphia Cable, eastern suburbs of Cleveland, Ohio.
Win98
NYC
Roadrunner (Time Warner)
Manx sez:
Can’t do dsl reports because my Java is broken. Has been broken for a while. I tried everything including re-installing.