dslreports logo
 
    All Forums Hot Topics Gallery
spc
uniqs
12

chris1212
@rr.com

chris1212 to tschmidt

Anon

to tschmidt

Re: Full T1 or Fractional T1 for Gaming

thanks for the response. i pretty much understand what latency and bandwidth is, but for onling gaming, latency is a lot more important than bandwidth (thus why i'm looking for a T1 connection). here's my question, let's say a game uses 50kbps, would me having a 15/15mbps connection over a 256/256kbps connection make a difference? or would that be a waste of bandwidth since the game uses 50kbps?

it just dawned on me...would my connection speed have anything to do with my actual bandwidth usage? i just thought about what you said, my SPEEDS would impact how FAST my information goes out, not how much data i'm using lol is that correct?

tschmidt
MVM
join:2000-11-12
Milford, NH
·Consolidated Com..
·Republic Wireless
·Hollis Hosting

tschmidt

MVM

said by chris1212 :

let's say a game uses 50kbps, would me having a 15/15mbps connection over a 256/256kbps connection make a difference? or would that be a waste of bandwidth since the game uses 50kbps?

All else being equal 15/15 would be faster since it takes less time to send a packet of data. Keep in mind T1 is only 1.5 Mbps not 15.

Lets put some numbers on those speeds. Maximum packet length is 1500 bytes which is 12,000 bits (1-byte = 8-bits)
At 256 kbps it takes 46.8 ms to send packet
At 1.5 Mbps it takes 8 ms
At 15 Mbps it takes .8 ms

More then likely gaming synchronizing packets are somewhat shorter so these numbers overstate the negative effect of low speed.
said by chris1212 :

would my connection speed have anything to do with my actual bandwidth usage? i just thought about what you said, my SPEEDS would impact how FAST my information goes out, not how much data i'm using lol is that correct?

Bandwidth is the amount of data sent over a connection. Speed affects how long it takes not how much you send.

/tom

chris1212
@rr.com

chris1212

Anon

thank you very much for your help. i have one last question..

let's say if i ran a ping test a on a t1 line and averaged a 3-5ms on a nearby server. what' a decent estimate of what i might get on a server located across country ie NY-CA?

thanks in advance

tschmidt
MVM
join:2000-11-12
Milford, NH
·Consolidated Com..
·Republic Wireless
·Hollis Hosting

tschmidt

MVM

Speed of light in a vacuum is 186,000 miles per second. Internet backbone is fiber, light travels slower in fiber about 70% the speed in vacuum or about 130,000 miles per second.

Ping measures round trip time. New York city to LA is about 2500 air line miles so round trip is 5,000 miles. Actual fiber optic cable length is significantly longer so figure cable length is probably closer to 6,000 miles.

You do the math.

That is the best possible time based solely on transmission flight time, in the real word other factors will typically add another 25-50 ms.

Still pretty fast when you think about what is actually happening.

/tom