TEST SERVER Central US — Dallas, TX, USA

[bigmakow]

Im from wisconsin and automatically connected to Central US — Dallas, TX, USA test server.  With that server i was getting any from 20% up to maybe 60 % percent of my isp's rated download.   I can connect to any other server and get my rated dowload speed of 200.  At first i thought it was my modem or router or ethernet cables.  Even if i test to server in UK, ASIA i get my rated down plus sum.  is there sumthing wrong with the texas servr???     https://testmy.net/quickstats/bigmakow

[CA3LE]

It may be routing, I'll test the Dallas servers out from a bunch of different locations and get back to you.

[bigmakow]

thank you

[CA3LE]

Heads up: I'll get all of the Dallas servers switched out within the next two weeks.

 

I have servers much further performing better for me personally.  Testing between servers I see good speed but on my client-side they aren't routing as well as they used to.  I feel it's on their end and it isn't getting better, so... the host and servers will be replaced.

 

I'm always looking for better performing hosts anyways and it literally takes 4 minutes to replace a mirror (once I pick one) so no big deal.  Give me a little bit to test out some fresh Dallas hosts.  I have to switch out various hosts sometimes, nothing unusual.  Your input helps. 

 

Please use the mirror option to select a different hosts if you feel like Dallas is under performing.  But please always inform me like you did if you feel something is persistent and affecting more than just one connection.  I've had multiple people now inform me of lower than normal Dallas results.  IMO -- Pulling off 4 different servers in the same datacenter... it has to be related to the peering in that datacenter.  Goodbye!

 

By the way, the new version of TestMy.net will allow you to select not only the location but the specific server for the location.  I plan to diversify my bandwidth over more providers more and give you the option to select.  The difference between what I'm offering and my competition is that I have to pay for all of my bandwidth out of pocket.  My website and all of its resources are hosted where normal websites host.  I don't have donated bandwidth or servers (nor do I accept them... at all).  In my opinion this is only truly unbiased if I host like everyone else.  --- another thing that different is that I don't think that you should test TOO close by.  I'd say put at least 1000 miles between yourself and the server for a real result.  Anything less is too easy.   If you can perform at a distance, that's the real test.  Everyone wants to get the big score while they ignore the true result, at a distance.  I used to get 30 Mbps to the UK... now I get 125 Mbps.

 

TestMy.net Test ID : f0NuzHlmR

Is that what Comcast sold me?  HAHAHAHA, no.  This is "1000 Mbps"!   "GIGABIT BLAST!!"  (whatever they call it this week.  Not gigabit service.)

 

Los Angeles gives me my best results right now.

 

TestMy.net Test ID : l19Qz1yBv

 

Is that Gigabit... Still, no.

 

And then I multithread and get...

 

 

Still not gigabit.

 

So I try speedtest.net (hey, maybe I'm the dummy, even though my test has done the same thing since the dawn of high speed internet... I could still be wrong) --- and get the expected best case scenario, 12 ms ping, multithread result. 

 

 

 

Even best case, that's not 1000 Mbps.

 

Let's check Comcast (Xfinity's) test, real best case, right?

 

 

wrong.  ...the same test as speedtest.net/ookla -- it's made by ookla so it's not surprising.  But even their own test has me at <50%. 

 

Actually,                   ---I'm going to request a credit next month.  50% is unacceptable.

 

Obviously this next part isn't meant for a TMN vet like you @bigmakow (nearly 8 years as a member) -- you already get the idea by now but for anyone who's new to testmy.net... keep in mind, I'm not trying to be your friend with my results but if you use them properly you WILL see an improvement.  I always make friends in the end with the truth.

 

Hold them accountable.  Ever realize that when you order service they PURPOSELY make you not want to call them back.  It's part of the 'plan'.  So make sure to mess up their plan and call them back when they don't serve you properly!  Make them pay for not delivering on their promises!  They effed up your internet, so eff up their little no call back plan!  I've made my providers pay MANY times over the years.  THIS... is by far the most extreme level of NOT delivering on package speed.  >50%!!!!!  Even admitted by their own account.

 

Multithread to UK, pretty much across the world.  4500 miles direct, much longer by wire.

 

 

Don't get me wrong, it's a good connection.  But you can't sell this to people and call it GIGABIT INTERNET SERVICE!       --- BS.  --- that's multithreaded!!!  Liars!

 

Truth is, they set up service in an area and then pray that people don't complain before they get the REAL speed available in the area. ("let 1 out of 10 complain while we roll around on our MONEY! MUH HA HA HA!)  I originally made this service so that we can BOTH have proof why our bills should be credited when we aren't getting what we pay for, when they aren't providing what they quoted us.  --- (actually, I originally made this for myself and it was an accident that anyone found it.  True story.)

 

"...sell it before you can deliver it.  -- two years later nobody will notice, then do it all over again!  WOOT-WOOT!  We're in the monnnneeeeyyy."

 

 

I'm only 37 but I love me a 1930's Gal!

 

When you see your landlord and you can look that guy right in the eye... THAT'S when you know,   ----you're Comcast.

 

... and then once they have a grip on everyone (everyone is in debt to them), the ISPs parade out and start speaking tongues...

 

To get it, keep watching till the end...

 

 

I hope at least one other person think this is funny.   

 

[CA3LE]

Dallas servers are all switched out to a different host.

 

I put 10 new servers online in Dallas tonight.  This will better parallelize tests.  Basically when you test... there's an even better chance that you're testing on a server all by yourself.

 

Also relocated DE and added a 2nd DE server, again to help parallelize.

 

I see better results on both.

 

Dallas for instance...

TestMy.net Test ID : a_jeL_PBv

I was consistently below 200 Mbps before on the same test.  I see improvement, I hope you do too.

 

?  ?

[mudmanc4]

I've not been able to max out my connection for years on the Dallas server, Bravo!

 

 

[CA3LE]

1 hour ago, mudmanc4 said:

I've not been able to max out my connection for years on the Dallas server, Bravo!

 

 

 

Bravo VULTR VPS (you get $50 free on their platform with that referral link).  

 

Edit: found a better link (above) -- I like this better because it gives you $50 to play with.

 

All I did was move the same thing to a better host.  I tested them in Dallas before... been a while, they've obviously stepped it up since then.  They're kickin' ass wherever I test them actually.   -- exact same configuration as before, better host.

 

I'm building a better self auditing system that will be doing benchmarks between the servers more in the background automatically.  Hopefully that will alert me in the future if a server or cluster of servers starts to under perform... hosts that continue to under perform will be marked for replacement.  My users should never notice this in the future.  But I may provide ways to show you that it's being done in the background. 

Edited February 7, 2019 by CA3LE
found a better referral link

[mudmanc4]

Heck I should have tossed my vultr aff ID your way

 

Either way, nice!

Was that 'old' host the same as I remember from back in the day?

 

If so, this all started about the same time they were bought out.

[CA3LE]

39 minutes ago, mudmanc4 said:

Heck I should have tossed my vultr aff ID your way

 

Either way, nice!

Was that 'old' host the same as I remember from back in the day?

 

If so, this all started about the same time they were bought out.

 

I've been using Vultr since 2015... 

 

yeah, IBM bought softlayer and it all went to shit.  Oh well, time to give them back the LAST grandfathered server... they killed the bandwidth on purpose.  Deals like mine, had to die.  They want all the old customers DEAD -- gone.  No joke.  Well, you win IBM.  BYE!  I never wanted to be on your platform in the first place... WAY to rich for my blood.

 

In hindsight, I should have picked up your linode link -- you turned me on to them long before I used them.

 

Edited February 7, 2019 by CA3LE
linode comment

[OCR]

Well, since I joined this site on...

 

 

...which was about ten years ago, it would probably make sense that I use the tests provided here, also... .

 

 

I just ran one, and show these results...

 

 

The results are real close to our "plan", and what we pay for... . ?

 

 

I also had my Task Manager running at the same time, and show these results...

 

 

Couldn't the Ethernet (Receive) R:  30.5 Mbps,  be considered a reliable "check" test... if a person is concerned about the test

 

accuracy  here??

 

.

 

 

 

 

 

[Sean]

The Ethernet throughput read-out can be pretty close to the actual user data throughput.  Its read-out is always a little higher as it includes overhead data, such as for routing, handshakes, error checking/correction and so on.  You can see this for the speed it shows in the other direction, i.e. the 0.3Mbps upload shown above is mainly overhead data as this test is taking place.  So around 0.3Mbps of that 30.5Mbps is likely overhead data also.

 

If let's say you run this over Wi-Fi or there was a poor Ethernet connection, that figure could be a lot higher than the actual throughput, such as if much of the data over the Ethernet connection to the Router consists of retransmissions.  It was one thing that used to bug me with the Ookla test as it would filter out dips that accounted for less than 30% of the test duration, giving inflated test results over wireless connections. 

 

If there's a server on the network (e.g. workplace), the Ethernet throughput will include data that occurs over the local network.  For example, Windows will periodically index network shares, Exchange databases and so on.  Windows 10 also distributes Windows updates between PCs on the same network.  Network devices that broadcast data (e.g. printers) will also add to the Ethernet throughput figure.  Then again, running an Internet speed test at workplace can also be challenging as it's difficult to make sure the Internet connection itself is idle at the time of testing.

[OCR]

Thank you for responding, Sean... .

 

.