dslreports logo
site
 
    All Forums Hot Topics Gallery
spc

spacer




how-to block ads


Search Topic:
uniqs
2006
share rss forum feed


bitchtorn

join:2006-10-11
Peterborough, ON

1 edit
reply to Webslingerac

Re: Teksavvy's speedtest server results often aren't reliable

.


Webslingerac

join:2004-05-01
canada

1 edit
reply to Webslingerac

said by Webslingerac:

Someone care to try to explain these results?

I also just started downloading a 7GB game on Steam. I'm getting a 3.3 MB/s download rate, which is decent for 28/1 cable service (the best I should be able to do at a sustained rate is 3.5 MB/s).

The point being that I'm a little skeptical of these Toronto speedtest results I just posted (including the results from Teksavvy's server).


sbrook
Premium,Mod
join:2001-12-14
Ottawa
kudos:13
Reviews:
·WIND Mobile
·TekSavvy Cable
reply to Webslingerac

said by Webslingerac:

said by sbrook :
So the differences between TekSavvy and Nexicom's speedtest results therefore can only be one of a very few things ...
Well, except sometimes I get bad results from Nexicom as well at the same time as I do from Teksavvy.

But we were talking originally how TekSavvy's server was giving a poorer result than Nexicom. You're talking something different and that is probably congestion ANYWHERE enroute between you and 151 Front.
said by Webslingerac:

quote:
It doesn't matter where you are in the province ... that's the route.
Yes, but there are more hops and routers along the path, the further away one is situated.

Irrelevent since anywhere you connect to will be along that same path. Say you live in Ottawa and want to go to www.gc.ca ... your packets will go from Ottawa to Toronto and then back to Ottawa. You want to go to nexicom, you go Ottawa to Toronto TSI to Nexicom. You want to go to TSI Speedtest, you go same route Ottawa to Toronto TSI to TSI speedtest.

The point being made is that the only difference when comparing any speedtest is the difference between Toronto TSI to the speedtest site wherever located. And generally as long as it has adequate capacity and is on a good pipe, the closer to Toronto TSI the better.


sbrook
Premium,Mod
join:2001-12-14
Ottawa
kudos:13
Reviews:
·WIND Mobile
·TekSavvy Cable
reply to Bhruic

[off topic]Bhuric, no, it's not a straight forward rule of apostrophe s if plural. This is a case of "if it makes the result awkard" ... James's (sounds like jameses) is accepted because James's isn't clumsy. Catharines's IS clumsy, so no s after the apostrophe!

It's one of the many rules of the English language where any rule imposed is so bad that nobody obeyed it, so the rule became "what isn't clumsy"

Of course now for recent generations, the apostrophe simply means "look out, here comes an s at the end of the word"!!!
[/off topic]


Webslingerac

join:2004-05-01
canada

4 edits
reply to sbrook

said by sbrook :
But we were talking originally how TekSavvy's server was giving a poorer result than Nexicom.
And the server in St. Catherines

During very specific hours, yes

But yeah, right

quote:
You're talking something different and that is probably congestion ANYWHERE enroute between you and 151 Front.
That's what I thought too, but I'm not positive. I've been pinging different hops, running traceroutes, downloading files from different locations, etc., and I can't pinpoint where this is happening.

Regardless, I don't feel the answer can be purely congestion related given the proximity of Toronto servers and the results from my first post.

quote:
Irrelevent
no--for the reason you stated at the very end (below)

quote:
since anywhere you connect to will be along that same path.
Yes, I know that.

quote:
And generally as long as it has adequate capacity and is on a good pipe, the closer to Toronto TSI the better.

Yes. The point I'm making is that someone testing from Vancouver (to the Toronto Teksavvy server), for example, may not get a result that accurately reflects the service he or she is receiving. Furthermore (and more importantly), along that path, there's far more variables with respect to potential issues (faulty routers, outside of the GTA, dropping packets for a week and suddenly having those routers replaced the next week, being one of several examples). To suggest (I'm not saying you are) that those variables are going to remain constant along a longer path than a more direct route is illogical (I would expect greater inconsistency over time the longer the route to the test server). Consequently, repeatedly testing for issues with the Toronto Teksavvy server from outside the GTA does not seem as useful as running repeated tests from within the GTA. We may have to agree to disagree on this point.

Webslingerac

join:2004-05-01
canada

4 edits

It's after a little after midnight (just to indicate why I feel testing around midnight probably isn't useful to show potential issues with respect to Toronto speedtest servers; that is, try around 9:15 p.m. instead).

Teksavvy Toronto server






Nexicom


Beanfield


Gone
Premium
join:2011-01-24
Fort Erie, ON
kudos:4

Your own test is still showing degradation even at midnight compared to two of the others.

I'm pulling better-than-a-full 31/2 on all four right now, though.


Webslingerac

join:2004-05-01
canada

1 edit

said by Gone :
Your own test is still showing degradation even at midnight compared to two of the others.

The results from all 4 servers are over 36 due to D3 speedboost (and my service is 28/1). Differences over 36 (on 28/1) are fairly negligible, I would think, no?

Anyway, I don't think people will whine and complain about a 36 Mbps result from 28/1 cable service. But they may point to a Teksavvy speedtest server result of 20 or lower and start complaining solely based on that test. And more importantly the first post in this thread, in particular, does point to a potential issue with the Teksavvy speedtest server.

Anyway, I guess people are looking into it.


Gone
Premium
join:2011-01-24
Fort Erie, ON
kudos:4

Ahh, speedboost. Gotcha.



Anaron

join:2005-01-28
North York, ON
reply to Webslingerac

I don't think I've ever had any issues with TekSavvy's speed test server in Toronto.
--
Formerly known as section32


Webslingerac

join:2004-05-01
canada

1 edit

said by Anaron:

I don't think I've ever had any issues with TekSavvy's speed test server in Toronto.

If you're subscribed to a service that's over 16 Mbps, could you do me a favour (I'm just interested) and run a speedtest using those same servers I used around 9:15 p.m. tomorrow and post the result, please?


Anaron

join:2005-01-28
North York, ON

said by Webslingerac:

said by Anaron:

I don't think I've ever had any issues with TekSavvy's speed test server in Toronto.

If you're subscribed to a service that's over 16 Mbps, could you do me a favour (I'm just interested) and post a speedtest result around 9:15 p.m. tomorrow?

I'll try to remember. I'm subscribed to TekSavvy's 18 Mbps and I regularly get over 30 Mbps on speedtest.net.
--
Formerly known as section32

Webslingerac

join:2004-05-01
canada

said by Anaron:

I'll try to remember. I'm subscribed to TekSavvy's 18 Mbps and I regularly get over 30 Mbps on speedtest.net.

Thank you! I'm pretty interested in the results from North York. Do you happen to know what POI you're on?


Gone
Premium
join:2011-01-24
Fort Erie, ON
kudos:4

said by Webslingerac:

Thank you! I'm pretty interested in the results from North York. Do you happen to know what POI you're on?

It's not going to make one lick of difference where he's located if people who aren't using Teksavvy (like me) are getting poor performance at various times, too.

Webslingerac

join:2004-05-01
canada

said by Gone:

It's not going to make one lick of difference where he's located if people who aren't using Teksavvy (like me) are getting poor performance at various times, too.

I'm interested in Anaron's routing path (I'm interested in congestion along the way and not just whether the server is at fault), especially if Anaron isn't having any problems.


Gone
Premium
join:2011-01-24
Fort Erie, ON
kudos:4

said by Webslingerac:

I'm interested in Anaron's routing path (I'm interested in congestion along the way and not just whether the server is at fault), especially if Anaron isn't having any problems.

His path is irrelevant for two reasons. The first is that if there was an issue between his POI and TSI, he would be experiencing issues with more than just the speedtest server. The second is that if people from outside TSI but who otherwise peer at Torix are experiencing issues on the server, it rules out a POI link being the cause, but from the other direction.

It very well could be a congestion issue, but it's most definitely not a POI link. If it was merely a POI link I'd be pulling from it at full speed in the evening when I'm not. The issue is somewhere local, and I trust that the gang at Teksavvy are smart enough to get to the bottom of it.

Webslingerac

join:2004-05-01
canada

1 edit

said by Gone:

His path is irrelevant for two reasons. The first is that if there was an issue between his POI and TSI, he would be experiencing issues with more than just the speedtest server.


Yeah, sorry. You're absolutely right. It's late, and I forgot to mention I also wanted tests to the other Toronto servers and to St. Catharines (if it's not too much trouble).

quote:
The second is that if people from outside TSI but who otherwise peer at Torix are experiencing issues on the server, it rules out a POI link being the cause, but from the other direction.
I'm sure hoping they're (my problems with all 3 Toronto servers) not POI related. But that might not matter in a few months anyway.

quote:
It very well could be a congestion issue, but it's most definitely not a POI link. If it was merely a POI link I'd be pulling from it at full speed in the evening when I'm not. The issue is somewhere local, and I trust that the gang at Teksavvy are smart enough to get to the bottom of it.

I'm sure they are. It's just that I'm not experiencing issues solely with Teksavvy's speedtest server (and I'm able to download fairly quickly from various sites and Steam around the same time I'm getting poor results from the Toronto speedtest servers).


Anaron

join:2005-01-28
North York, ON
reply to Webslingerac

said by Webslingerac:

said by Anaron:

I'll try to remember. I'm subscribed to TekSavvy's 18 Mbps and I regularly get over 30 Mbps on speedtest.net.

Thank you! I'm pretty interested in the results from North York. Do you happen to know what POI you're on?

Yes. Greensboro.
--
Formerly known as section32