Spooky is correct.
Also sly, I would argue you would need both a networking class and a physics class. The biggest distance possible between two players is a little more than 41000 km here on earth. Practically you will deal with much shorter distances, and lets take the america-aussieland example:http://www03.wolframalpha.com/input/?i=distance+sydney+new+york, 15000km. With C being 299 792 458 m / s your ping would be 50 ms(I think, atleast its super small).
The Speed of light has nothing to do with it, rather it has to do with the number of "hops"it has to make via "relay stations". Now this can be linked to distance, but the speed of light has nothing do do with it.
You have hearby disqualified yourself from this conversation IMO.
1.) By your own link the distance is far closer to 16,000km than 15,000km.
2.) By your own calculations and assuming speed of light (which, as I stated, propagation speed is NOT... it is slower), you get 100ms, NOT 50ms. You clearly do not understand that ping is a ROUND TRIP TIME, meaning TO the destination AND BACK. 100ms, notwithstanding the fact that the distance is actually longer and the speed is actually slower, is already significant delay. Remember: this is the ABSOLUTE MINIMUM time it takes to perform ONE EXCHANGE (send 1 packet and receive its acknowledgement); you are ignoring a multitude of other factors, SOME of which I'll outline as follows. The point? The speed of light simply isn't fast enough for particularly low-latency global networking, no matter what hardware we throw at it.
3.) You assume a straight line from your source to your destination, which routing most certainly is not. The additional propagation delay induced from all the misdirections further increases this base number.
4.) You assume no processing whatsoever. When the packet is delivered to the destination it must be processed. To be processed, the target system must interrupt the OS which must stop what its doing to deliver the packet to the target process which must read the packet, create a reply packet, and send it out the network interface, all of which can take time that may factor in at the millisecond granularity.
I won't get into ignoring the time that the packets spend in the routers themselves as this post dealt mostly with propagation speed, but recall that more distance almost invariably means more routers which logically does bring them into play with this discussion.
I would encourage you not to post if you do not understand what a ping actually is.
Ping does not matter, as long as it is under 350. Even with all pings under 100 the game can lag.
Pings under 100ms can lag is a true statement, as simspeed/packet analyzing programs can come into play regardless of latency.
In general, the higher the ping the more likely the game will lag is also true...
As for the number of hops issues, some "locals" playing in America can have more hops then I do to someone else - especially if they are at a uni or somewhere like that with a complicated network.
Complex networks like universities do increase hop counts, but are you assuming only Americans can reside in such networks? Someone on the other side of the world is just as likely to reside in a complex network, meaning the average number of hops to get to the person will go from high to very high.
It's similar to claiming the people under X (say, 150ms) ping can still lag games due to simspeed/packet analyzing programs. I ignore this because I assume every player is equally likely to run into problems for this reason; higher latency players just have more opportunities to increase lag and therefore a greater chance of doing so.