FranzMetcalf
join:2002-09-01
Los Angeles, CA
|  $10 768 Service Speed/Latency
Hi all,
I'm seriously considering coming over from Earthlink 3.0 (roughly 2500kps downstream service) to ATT's court-ordered amazingly cheap $10 768kps service. But I wonder about latency. I really think I can handle the large files/photos/mpgs etc. taking longer to download, but I think I'll go crazy if my periods of waiting to connect to new sites also get longer.
I'm paying $40 a month so saving $30 a month sounds mighty nice, but I can't handle tripling my waiting time to navigate from site to site and page to page. Is that going to happen with the 768kps service?
What do you all think?
Franz
--
Franz Metcalf, Ph.D.»mind2mind.net |
|
d_l
Barsoom
Premium,MVM
join:2002-12-08
Reno, NV
1 edit | I think the FCC isn't a court, but a regulatory agency.
That is a plot of pings to DSLR over a regularly priced, Basic line and an Elite line. They alternate frequently over this route.
Note the ping response time for the first hop. It is shown for the Basic line and lies between 7 and 9 ms. Note also that DSLR's router isn't responding well, but that the latency to it really doesn't change when the traffic is arriving to it on the Basic line or on the Elite line.
How does this compare to your Earthlink first hop latencies. |
|
Brian Grif_n
Wireless Guru
join:2001-01-17
Reno, NV
| reply to FranzMetcalf
In my area Earthlink just resells AT&T service using AT&T's copper. The difference between 2500 vs. 768 download might be a few milliseconds - nothing you should notice.
--
Check out my website: www.renowirelessinfo.com |
|
FranzMetcalf
join:2002-09-01
Los Angeles, CA
| Gang,
Thanks for the info so far. And for the correction: the FCC is of course a Federal Agency, not a court. And through its regulatory powers it forced ATT to offer $10 DSL service in 22 states until 2009. Lucky for us.
Now, for the latency issue. I did the DSLR line quality test and got these results:
»/linequality/nil/2325174
Being fairly ignorant, I don't know what much of it means, but the test did clearly say my first hop was 9ms, so that's comparable to what d_l found for ATT. Does this suggest I wouldn't be too hard hit on the latency front by switching?
Oh, and I think that Brian is right about Earthlink using AT&T's lines in my neighborhood, too. I recall that from when EL was provisioning my line a few years ago. So things should be fairly comparable.
Cheers, Franz
--
Franz Metcalf, Ph.D.»mind2mind.net |
|
d_l
Barsoom
Premium,MVM
join:2002-12-08
Reno, NV
| reply to FranzMetcalf
Do a tracert to DSLReports and compare the latency your first hop to my PingPlot for the most accurate results. Reverse tracerts such as the line quality test here can sometimes be inaccurate (usually they are correct though) on their estimates of first hop latencies.
I don't think you will have any increase in first hop latencies by switching to AT&T. Normally the first hop delay is a function of the upload speed and the Basic plan has 384 upload speeds. Faster uploads than that don't seem to reliably lower the first hop delay.
Then you have to consider the latency of the routing your ISP will use. AT&T normally has pretty good routing. I don't know what has happened to my routing to DSLReports. It used to be better than it is now.  |
|
