Page 1 of 1

Cool little link I found

Posted: Thu Dec 08, 2011 9:10 am
by Mopar Madness
Just surfin google and stumbled upon this server location list, I presume based on the domain that its by Blast, so thanks, Blast :)

http://bzexcess.com/geolist/

It shows the distance of every server from roughly where you live. Distance isn't the only factor in determining which server would be least laggy for you, but it plays a fairly big part, plus I just think its cool to see where in the world all the servers are located and how far they are from my house.

Re: Cool little link I found

Posted: Thu Dec 08, 2011 11:53 am
by blast
Yep, that'd be mine. It's not perfect, of course, but it's about the best it can be. The GeoIP database doesn't have very good info for some countries, so it will just say, for example, "Germany", instead of a more exact location within Germany. And like you said, physical distance isn't the only thing to take into account. Without a way to do ICMP on the browser-side though, it's the best that can be done. :)

Re: Cool little link I found

Posted: Thu Dec 08, 2011 6:22 pm
by Bullet Catcher
Nicely done. A column that shows the round-trip distance in light-milliseconds would give a best-case lag value, and might even be useful as a teaching tool.

Re: Cool little link I found

Posted: Thu Dec 08, 2011 11:51 pm
by blast
Yeah, it's just too bad that when you're in Germany it shows a distance of zero to any of the servers in Germany. ;)

Also, the source is available for anyone to hack away at. The link for that is at the bottom of the page.

Re: Cool little link I found

Posted: Sat Dec 10, 2011 8:07 am
by Yrogirg
Bullet Catcher wrote:Nicely done. A column that shows the round-trip distance in light-milliseconds would give a best-case lag value, and might even be useful as a teaching tool.
Huh, I've been trying to calculate internet channel length in light-milliseconds, but it' not so easy --- I don't know the length. The straight line distance between me and California is around 10 000 km (20 000 km is 67 ms). But this line goes through Greenland! My usual lag is around 220 for GU California.

Re: Cool little link I found

Posted: Sat Dec 10, 2011 9:16 am
by joevano
Yrogirg wrote:Huh, I've been trying to calculate internet channel length in light-milliseconds, but it' not so easy --- I don't know the length. The straight line distance between me and California is around 10 000 km (20 000 km is 67 ms). But this line goes through Greenland! My usual lag is around 220 for GU California.
That makes sense... you are probably never going to get close to the calculated light-second time but you will definitely never ever have a better lag time (because it is physically impossible to go faster than the speed of light [most likely, though some current findings may disprove that soon]) playing on the server. It is a best case scenario in a perfect world where there is no equipment between you and the server to slow the traffic down and it assumes that the wire is perfectly straight between you and the server.

Re: Cool little link I found

Posted: Sat Dec 10, 2011 9:42 am
by Yrogirg
I'm actually quite surprised the equipment is so fast. Assuming the real lag due to wire length is around 100-120, it is responsible to the almost half of my lag. The other half is contributed by various routing hardware between wires. The light is slow indeed, we can feel its speed in game!

Re: Cool little link I found

Posted: Sat Dec 10, 2011 10:45 pm
by dereliction of duty
joevano wrote:you are probably never going to get close to the calculated light-second time but you will definitely never ever have a better lag time (because it is physically impossible to go faster than the speed of light [most likely, though some current findings may disprove that soon]
an out of the box type question would be;

when can we expect bzflag to incorporate "that stuff" which they find can travel faster than the speed of light...

I'd imagine that would take care of the lag problem!

Re: Cool little link I found

Posted: Sun Dec 11, 2011 1:00 am
by quantum dot
ttfkad wrote:
joevano wrote:you are probably never going to get close to the calculated light-second time but you will definitely never ever have a better lag time (because it is physically impossible to go faster than the speed of light [most likely, though some current findings may disprove that soon]
an out of the box type question would be;

when can we expect bzflag to incorporate "that stuff" which they find can travel faster than the speed of light...

I'd imagine that would take care of the lag problem!
There is a tremendous lag in servers when optical signal is treated to be sent to the next point in the line. The reason is that photons (light), which travel at light speed through optical fibers, are transformed into electrons in the servers and electrons do not travel at light speed. This always adds a few milliseconds at each router (not to mention queues, jobs, noise, and software that has be be run for the process to work). Electronics is comparatively slow in front of photonics.

"that stuff" that has been reported to apparently travel faster than light are some type of neutrinos (actually, a minuscule proportion of them), and there are serious doubts in the scientific community that the experiment is correct. And there are good reasons to think so. Besides, we will probably never be able to construct machines based on "neutrinonics" so to speak, so even if some of them can be faster than light, they will be of little use for communication. Neutrinos are among the most undetectable things out there, they hardly interact with matter. You cannot imagine the number of bits that would be lost if we used neutrinos instead of photons or electrons!

So forget it, we will have to live with lag as a part of the game :) ... at least until fully working optical computers can be built massively.

Re: Cool little link I found

Posted: Sun Dec 11, 2011 1:26 am
by JeffM
quantum dot wrote: So forget it, we will have to live with lag as a part of the game :) ... at least until fully working optical computers can be built massively.
Or until the game gets modern networking and lag compensation and grows up to be a real internet game....

Re: Cool little link I found

Posted: Mon Dec 12, 2011 5:51 pm
by FiringSquad
quantum dot wrote:"that stuff" that has been reported to apparently travel faster than light are some type of neutrinos (actually, a minuscule proportion of them), and there are serious doubts in the scientific community that the experiment is correct. And there are good reasons to think so.
We already have faster-than-light technology in BZ. It's called a teleporter. I figure a few of the neutrinos just forgot to pick up their PZ flag and ended up actually using the tele and skewing the averages.

Re: Cool little link I found

Posted: Mon Dec 12, 2011 11:48 pm
by quantum dot
FiringSquad wrote:
quantum dot wrote:"that stuff" that has been reported to apparently travel faster than light are some type of neutrinos (actually, a minuscule proportion of them), and there are serious doubts in the scientific community that the experiment is correct. And there are good reasons to think so.
We already have faster-than-light technology in BZ. It's called a teleporter. I figure a few of the neutrinos just forgot to pick up their PZ flag and ended up actually using the tele and skewing the averages.
Neutrinos are big cheaters indeed!

Re: Cool little link I found

Posted: Tue Dec 13, 2011 1:06 pm
by Cobra_Fast
FiringSquad wrote:
quantum dot wrote:"that stuff" that has been reported to apparently travel faster than light are some type of neutrinos (actually, a minuscule proportion of them), and there are serious doubts in the scientific community that the experiment is correct. And there are good reasons to think so.
We already have faster-than-light technology in BZ. It's called a teleporter. I figure a few of the neutrinos just forgot to pick up their PZ flag and ended up actually using the tele and skewing the averages.
If you take player lag into account, light is still faster ;)

Re: Cool little link I found

Posted: Tue Dec 13, 2011 2:08 pm
by FiringSquad
Cobra_Fast wrote:If you take player lag into account, light is still faster ;)
You're forgetting that the BZWorld is actually viewed in scale and that the HIX map is almost 45 light-seconds across.
Frankly I'm amazed at the acceleration of the tanks. If it wasn't for their minuscule mass, they'd explode just by moving.

Re: Cool little link I found

Posted: Tue Dec 13, 2011 5:18 pm
by Mopar Madness
but it only takes under 30 seconds to cross the hix map... o.O tanks are faster than light? That solves the lag dilemma, just use tanks instead of photons or electrons and lag would drop immensely!

Re: Cool little link I found

Posted: Tue Dec 13, 2011 6:22 pm
by blast
Doesn't seem like 45 light-seconds makes sense, since I don't think the map is 8.3 million miles across. :)
https://en.wikipedia.org/wiki/Light-second

Re: Cool little link I found

Posted: Tue Dec 13, 2011 9:07 pm
by JeffM
where in fact it is actually 2.6685 light microseconds across (800 meters)

Re: Cool little link I found

Posted: Wed Dec 14, 2011 4:25 pm
by yosef
Perhaps on the Discworld, under the heavy influence of magic, light moves that slowly.

Re: Cool little link I found

Posted: Wed Dec 14, 2011 11:11 pm
by FiringSquad
800m? I didn't realise that we dealt with real-world measurements in the map. Thought it was just arbitrary units in 3 dimensions. Oh well. It was fun while it lasted.

Re: Cool little link I found

Posted: Wed Dec 14, 2011 11:59 pm
by JeffM
Technically it is completely arbitrary. But if you assume 1WU == 1 meter then the tank is very close to the size of a small tank, so the approximation works.

Re: Cool little link I found

Posted: Thu Dec 15, 2011 5:14 am
by Yrogirg
FiringSquad wrote:800m? I didn't realise that we dealt with real-world measurements in the map. Thought it was just arbitrary units in 3 dimensions. Oh well. It was fun while it lasted.
Well, the only real unit in bzflag is second (time). The default gravitational acceleration in bzflag is 9.8 (length units) / s^2 that clearly implies length to be measured in meters.