View Single Post
Old 07-09-2012, 15:39   #94
Chrysalis
cf.mega poster
 
Join Date: Sep 2003
Posts: 12,047
Chrysalis is cast in bronzeChrysalis is cast in bronzeChrysalis is cast in bronzeChrysalis is cast in bronze
Chrysalis is cast in bronze
Re: Finally had enough of VM BS...........

Quote:
Originally Posted by qasdfdsaq View Post
I've modified the test slightly to simulate a typical high-quality VOIP connection - 100pps, 80B payload size. On your connection:

Packets: sent=1000, rcvd=1000, error=0, lost=0 (0.0% loss) in 10.029377 sec
RTTs of replies in ms: min/avg/max/dev: 20.327 / 32.502 / 53.147 / 5.985
Bandwidth in kb/sec: sent=10.768, rcvd=10.768
Out of sequence packets: 21

VM in my area:
Packets: sent=1000, rcvd=1000, error=0, lost=0 (0.0% loss) in 10.020790 sec
RTTs of replies in ms: min/avg/max/dev: 25.218 / 30.334 / 54.009 / 3.774
Bandwidth in kb/sec: sent=10.777, rcvd=10.777
Out of sequence packets: 2

BT Infinity:
Packets: sent=1000, rcvd=1000, error=0, lost=0 (0.0% loss) in 10.021413 sec
RTTs of replies in ms: min/avg/max/dev: 31.378 / 31.789 / 32.796 / 0.170
Bandwidth in kb/sec: sent=10.776, rcvd=10.776
Out of sequence packets: 0

Statistics for pingbox1.thinkbroadband.com:
Packets: sent=1000, rcvd=1000, error=0, lost=0 (0.0% loss) in 10.001254 sec
RTTs of replies in ms: min/avg/max/dev: 11.144 / 11.257 / 12.889 / 0.085
Bandwidth in kb/sec: sent=10.798, rcvd=10.798
Out of sequence packets: 0

Note the Std. Dev figures don't tell the whole story, as packet to packet jitter on VM in my area is quite low, with occasional "gaps" of 20ms+ e.g.:
Spoiler: 
Reply from 92.235.190.31: seq=0386 time=29.341ms TTL=50 ID=3500
Reply from 92.235.190.31: seq=0387 time=27.262ms TTL=50 ID=3501
Reply from 92.235.190.31: seq=0388 time=29.315ms TTL=50 ID=3502
Reply from 92.235.190.31: seq=0389 time=27.361ms TTL=50 ID=3503
Reply from 92.235.190.31: seq=038a time=45.572ms TTL=50 ID=3504
Reply from 92.235.190.31: SEQ=038c time=29.594ms TTL=50 ID=3506
Reply from 92.235.190.31: SEQ=038b time=39.729ms TTL=50 ID=3505
Reply from 92.235.190.31: SEQ=038d time=27.296ms TTL=50 ID=3509
Reply from 92.235.190.31: seq=038e time=29.279ms TTL=50 ID=350a
Reply from 92.235.190.31: seq=038f time=27.285ms TTL=50 ID=350b
Reply from 92.235.190.31: seq=0390 time=29.370ms TTL=50 ID=350c

Yet on yours it's constantly variable but most jumps are <10ms:
Spoiler: 
Reply from [your ip]: seq=03dd time=36.349ms TTL=51 ID=dc08
Reply from [your ip]: seq=03de time=34.260ms TTL=51 ID=dc09
Reply from [your ip]: seq=03df time=37.076ms TTL=51 ID=dc0a
Reply from [your ip]: seq=03e0 time=40.866ms TTL=51 ID=dc0b
Reply from [your ip]: seq=03e1 time=35.427ms TTL=51 ID=dc0c
Reply from [your ip]: seq=03e2 time=25.542ms TTL=51 ID=dc0d
Reply from [your ip]: seq=03e3 time=26.181ms TTL=51 ID=dc0e
Reply from [your ip]: seq=03e4 time=31.852ms TTL=51 ID=dc0f
Reply from [your ip]: seq=03e5 time=43.401ms TTL=51 ID=dc10
Reply from [your ip]: SEQ=03e7 time=37.374ms TTL=51 ID=dc12
Reply from [your ip]: SEQ=03e6 time=47.570ms TTL=51 ID=dc11
Reply from [your ip]: SEQ=03e8 time=38.876ms TTL=51 ID=dc13

And despite st.dev. being less than 2x as high in your area, out of sequence packets are over 10x higher, the latter being, as mentioned, quite disruptive to VOIP communications
you think thats ok constantly variable during the morning?
Chrysalis is offline