Originally Posted by Justin M. Giza
Well, I started off in WoW at a healthy 52ms. This jumped to 354ms. Then 1700ms. Then 4200ms. Then 6363ms.
But I didn't get booted from the server! Hoo-ray!
In addition, PingPlotter read much less packet loss. If you'd like me to post the new charts, I will.
I'll give it another shot late-night, when there's ideally less usage.
After I made that initial post, we noticed that the changes were a bit less than ideal and were in fact making things a lot worse after a short while. I ended up having to go back in to the office around 10 - 11 pm and work on the packet shaper.
In the end I put it back into a similar configuration to how we had it earlier, but with some additional modifcations and adjustments which should hopefully make it work a little better all around. That is of course the hope anytime we make an adjustment to this device, it's just sometimes a bit unpredictable how it will turn out what with the chaotic nature of the internet and such. There are a great many ideas that seem wonderful until they are implemented.
So what I'm saying is, I know it must have been bumpy last night, with only http really working well. We didn't do that on purpose. We are actually trying to make things better, believe it or not, and spend a great deal of time on trying to get the most of the internet connection. Beyond that we are of course attempting to see about increasing our bandwidth in a couple of ways, but until that day arrives, we've got to work with what we have and we are not going to give up.
Justin, still drop by today and we'll test out whatever we can.
Even if not as many students were on campus last night and were potentially not as many users on the internet, we were mostly maxing out our connection the entire time I was at Drew- 90-95% in both directions. This is not actually typical believe it or not, so it would have been a good time to test latency in applications.