Forum Discussion
Best way to test latency?
Yes, 600-700 is normal for satellite internet.
Keep in mind that websites add to the delay because they have so much stuff on them that needs to load, such as images or videos. So on top of the normal server to server delay, you will have added delay from website content.
How to read:
The number at the left is the hop number.
Hop 1 is 192.168.42.1, your modem.
Hops 2 and 3 are the HughesNet path to the gateway (thought there should be another one somewhere).
Hop 4 is Qwest (now owned by CenturyLink), your gateway provider.
Use 525mS as an easy to remember guide, because the radio path takes that much time to go to from you to the satellite, down to the groundstation, and then the return trip back to you.
These are all showing about 1-1.5 seconds more than that, which sounds to me like either a network compatibility problem or simple network congestion (congestion being too many users, causing a processing/routing backlog). CenturyLink is well-known for sporadic 1-minute gaps in service, not 1 second, so it's not the provider.
It could feasibly be peak time congestion, but it'd have to be pretty congested and I'd have to see what one of these looks like during a non-congested period. If it's that high (readings of 1500-2000mS) during non-peak time, then it's not congestion, it's a network compatibility issue.
But again, I don't know what tracepath is actually doing. I'm working from the assumption that it's something similar to traceroute and I might be wrong.
MarkjFine, So one doesn't add up the ms values on each hop to determine one's total round-trip latency? But rather, the ms value at far right on each of those lines (i.e. 1471.185ms on the third hop 3 line) reperesents a total round trip latency?
When I do a ping command, I get ms numbers on the right of each line in the 600-900 range, which seems more expected. Does that indicate that those 1-1.5 sec. values on each line after a tracepath command are probably not accurate? Or is the ping command not accurate?
When you talk of the possibility of a network compatibility problem, what networks are you talking about? I'm a guy in central florida using my Hughes service to ping amazon.com, why would any of the networks used to execute that very common route for me have incompatibility issues?
