Bandwidth speed better = not better?

After using various tweaking programs I had tested at 5152kbps download speed and 652 upload but was not happy with my actual download speed of around 420KB. So I ran TCP optimizer and set my ISP max speed and then ran EvID4226Patch.exe.
After doing so I retested my badwidth speed and tested at 5111kbps download and 611 upload. Even though this was a bit slower on the 2 different bandwidth tests my web pages (though fast before) are now even faster and my actual download speed is now 620KB. I don't get it. I'm testing a bit slower but working faster. Can someone enlighten me please as to why this is or may be.
The one major change was my half open ports was set to some obscene number like 16667 and then changed it to 50 with the EvID4226 Patch.

 

5152 kbits/sec allows for a max download speed of 644 KB/sec (simply divide it by 8, but account for TCP overhead as well, you'll never reach 644), so this begs the question, were you testing your 'actual download speed' from the same site both times? was it during peak hours? i suggest runnign speed tests from dslreports.com , just click on tools and go to speed tests. find a location near you and see what it comes up with, chances are this will be your most accurate indicator.

 

Bandwidth has less to do with how fast you browse, instead your latency to a website does.

Bandwidth is how much packets you can get at a time, latency is how fast you get em.

Using the same provider, for example, I will not browse the internet any faster at 8Mbps, than I would at 4.

Speedtests test your ability to download, not your ability to browse. This will change as the server load does, as well as network traffic changes.

I've downloaded from a site at 1.5 megabytes per second at one point, and downloaded from it at 200 kilobytes a second, another.

I've found that any speed test thats on my same ISP, I get full bandwidth. While it makes for a pretty number, its not representative of real world performance. Beware of judging performance on bandwidth tests alone.

 

I'm using ATT dsl and did my bandwidth tests at speakeasy speed test and speed test.net (same thing different link) 1 test each both times. My actual download speed was from a site where I can see how many people were on it and I know what I get there. I was in the low 400 KB for weeks and now am consistenly 617-622 KB and it started immediately after my change which took less than 6 minutes. My latency initially was .37-.58 and is now down as low as .31 and stays in the .3's now.

Needless to say I'm happy,,,for now but just didn't get it. Thnx for the replies but what I still don't get is if I could handle more of a download with higher bandwith then why would I download slower with it. I understand my latency is better so the web page would load faster now. thnx

 

Download slower in regards to, what?

The speedtest vs. real world speed?

There are so many factors...network traffic, routing, file type.

For example, a text file will download a whole lot faster than say, an mp3.

Observe:
http://www.speedtest.net/result/93429657.png

I cant hope to download that fast most of the time. It takes a download accelerator to do it. So, as I mentioned prior, bandwidth tests are rarely a measure of true performance. I remember when Speakeasy messed up recently with one of the files used to test bandwidth, and I was getting almost 20Mbps down.

I even found a bandwidth tester that used a .gif file to test....it scored me 1Mbps faster than the above test. I've even had web browsers affect bandwidth tests!

 

Slower in regards to the amount of bandwidth your getting....In my example I was getting a higher bandwidth but a slower actual download speed with everthing being equal. The bandwidth test was the same before and after what I did. And the program I downloaded was the same and done for weeks at low 400's KB even though the bandwidth was testing higher consistenly for weeks also.
Then I do the change I mentioned and decrease my bandwidth on the same bandwidth tester and yet increase my download speed on the same program from the same site and it's been dozen's of times for days.
Anyway, not a big deal. Performance is better and I'm happier.