Sean

I noticed recently while running speed tests, the upload test tends to complete right after the initial 96KB test without going on to test larger blocks.  The results below show an example where it did this four times in a row (the '96MB' should be '96KB'), run on Firefox with Windows 10.
In the four highlighted results below, the upload test completed within a fraction of a second and displayed the result based on the 96KB block size:

In the '2.1MB' test above, I did the Express test, so this gives an idea of what the uplink should be at the place I'm visiting in California, whereas the 96KB tests (shown as 96MB) are out due to the tests completing nearly instantly after I start the test.

Based on what I've often heard about satellite Internet, it is considered a "last resort" means of Internet connectivity, i.e. where you cannot get a stable DSL connection, fixed wireless, cellular data or other means of Internet connection.  No matter what the satellite operator claims, it is certainly not a DSL, cable, etc. replacement, with the small exception of very slow (<1Mbps up/down) and unstable connections. 
Your speed tests actually seem pretty good for a satellite connection.  I have visited a few people who had a satellite connection which typically hovered around 300kbps in the evening and incredibly slow to browse the internet, i.e. 30 seconds per page. 
Minimum ping (round trip) time over satellite is generally 600ms to 800ms due to the physical time it takes for the signal to reach the geostationary satellite and back, so satellite is unsuitable for low latency requirements such as online gaming.  I'm fairly sure VoIP works, but with roughly a 1 second delay, similar to a satellite phone.

I think I can do one better, well... on having the slowest tortoise...
 

 
North West Electronics Bluebox, for what's suppose to be up to 10Mbps!

It's still there. I just took the top link down.  Look at the menu under "Extras > Mercury Speed Test"  
 
I'm in development but not working on that.  I just moved it because it isn't a very popular option...
 
I'm juuuust about to hit reply and the site just told me that you replied... I noticed that you found it under the menu.  

I tried a handful of tests between my PC and mobile and this time it seems to be resolved. 
 
Like my home fixed wireless ISP, the Three 3G and LTE networks both need a multi-thread test to check the full capacity available.  I assume they throttle linear transfers to reduce network usage such as force streaming TV to use a lower bitrate as I've never seen a linear test show anything faster than about 10Mb even on 4G.
 
The Meteor 3G network doesn't seem to do that, which I'm currently trying out with a prepay SIM.  I get roughly the same speed whether I use linear or multi-threaded tests and their 3G linear speed is sometimes faster than a multi-threaded test on Three's LTE network. 
 

 
I would be curious to see what Meteor's 4G would achieve, but they charge €20 for 4G compared to €10 for their 7.5GB add-on and their nearest 4G coverage is also a 1.5 hour drive from home.

Want to see something astonishing? 
Go have a peek at how many hidden directories are in windows 10

This morning, I noticed that roughly every second multithreaded test I do ends up running a linear test. 
 
Just in case it was a cache/cookie issue or with Firefox, I tried again in Chrome after clearing my cache & cookies and it did the same thing as shown in the following screenshots:
 
For example, I start by choosing the Multithread option and picking London UK:
 

 
I then click Save and go into start a 25MB download:
 

 
It then shows a screen as if I'm running a linear test and even suggesting I try the multi-threaded test:
 

 
After the test completes, it shows the variation in the throughput throughout the test exactly like if I ran a linear test, yet shows the lightning bolt in the result line showing that I ran a multi-threaded test:
 

 
Finally, when I switch back to Linear test mode, it defaults to the Texas server.  Normally it defaults to the London UK one which is nearest to me:
 

 
Going a little off topic here, but when I went to make this thread, I noticed that the forum will not let me upload images either.  The images in this post are on my own host as I couldn't upload them.
 

 
Sure is a strange morning even back at home.  I noticed that I could not connect to either of my Internet connections (DSL or fixed wireless.)    The router for the fixed wireless crashed (a reboot fixed that) and one of the HomePlug adapters for the DSL has died.

It appears that either your connection to YouTube is being throttled in some way or it's a congested server.
 
As YouTube uses a different codec in Chrome than Firefox, I would suggest testing Firefox (or Internet Explorer) with a few HD videos.  Chrome uses the VP9 codec, where as Internet Explorer and Firefox both use the H264 codec, so YouTube will likely fetch the videos off a different server. 
 
Another possibility is HTTPS based connections being throttled.  One simple test to check is download a large file from a HTTPS connection see what throughput you get.  One example is the standalone Chrome browser package which is served from a HTTPS connection:
 
https://www.google.com/chrome/browser/desktop/index.html?standalone=1
 
Based on your above speed test, you should get a similar speed to what I got:
 

 
If this downloads very slowly at the time YouTube is having trouble playing in HD, then it's very likely HTTPS based traffic is being throttled on your ISP.  Otherwise, if you're just having trouble playing YouTube in HD, it's possible YouTube is trying to play from a congested server or along a congested path to it.

Based on my experience using the Ookla Speedtest mobile app, the Android app can't be trusted!
 
Have a look at the following two tests run on my mobile (OnePlus One):
 

 
The following is the router's DSL status page which I ran both tests with:
 

 
Notice the overinflated uplink result in the Speedtest app result vs. my DSL line speed.
 
On my fixed wireless ISP connection, it underrates the downlink:
 

 
As I don't have access to a status page showing what my Wireless ISP is limited at, I ran an FTP test right after the following tests completed:
 

 
3.4MB = 27.2Mbps, so it's obvious which result is more accurate.
 
These are the results form a YouTube video I created showing these tests in action.  Usually the Speedtest.net app reports about 50% above my actual uplink for any connection, but in this case it was out by over 3 times the link speed!  Plus catching it in a screen recording made was even more exciting as I think I only ever seen it go that high once before.
 

As long as I'm logged in, the tests carried out in the auto test are shown in the "My Results" page, even if the test gets stuck or I inadvertently close the browser ...
 
One way to help make the Auto test results stand out is to choose an identifier, e.g. "Home" if you don't use that identifier with manual tests.  This way it shows the associated icon next to the results on the "My Results" page, so you know which were from the Auto test.
 
If you were logged out while running the Auto test, go back on the connection you ran the auto test and bring up the "My Results" page.  There is good chance it will show the results from the partial run that were carried out, assuming your IP address did not change since running the Auto test. 
 
If the results are not shown, try this with your computer connected to the ISP you ran the Auto test on:
Go into the "Database" menu, then into "My Detected Info" and then into the "ISP: (your ISP)" (4th listing) Click the button "Speed Test Log". Change the "25 results/page" option to something like "250 results/page" Look through the date & times on the left column to see if any match up with the time you ran the Auto test and continue page by page until you do.  This is generally pretty straight forward unless you've using a popular ISP like BT Broadband. A tale-tale sign you've found your results is where the "ConnectID" column has a matching figure at roughly each time interval you chose, e.g. each 30 minute time period has the same ConnectID for a 30 minute interval test. Once you're fairly sure you've spotted it, click that ConnectID figure and you should have the series of test results from that Auto test.

Thanks for reporting this, you've given great information that I can use to make the test better.  I'll try to break it the same way in development and work to make it throw an error instead of calculating a result in that instance.
 
The result has been removed from the database, so your stats look clean again.  
 
Thank yo so much for the support.

Since the day I've run into TestMy, I've been running several download and upload tests pretty much every day.
 
This evening when I ran the Download Test, I picked the 25MB size from the drop-down and the browser stopped responding.  I was just about to close Firefox when a box appeared saying a script stopped responding and whether I should terminate it.  So I did and the following screen appeared:
 

 
Somehow, I don't think my PC's Wi-Fi connection can deliver that let alone my ISP's fixed wireless connection... However, it sure would be nice!!
 
The following is what my connection is right with another test in multithreaded mode:
 

 
I'll leave the spike there in case it is useful for debugging.  Otherwise I think it would be worth deleting that result before anyone starts thinking I'm an employee that accidentally ran a Speedtest within the network.  I consider myself lucky to be in a very rural area with few users on the mast as I've heard of areas where users sometimes get below 1Mbps in the evening.

Some DSL routers I've used in the past did something similar where it seems to be related to how they deal with and recover from noise on the line.
 
For example, on Smart Telecom's (former ISP here) earlier DSL router model, if there was excess noise on the line such as the DSL line passing someone's home who operates welders, power tools, etc., it would resync the connection to achieve a 15dB signal-to-noise ratio.  However, when the interference went away, the DSL router remained synced at that lower speed until it was rebooted.  My brother had this issue and had to keep cycling the power each morning to get the speed back to normal.
 
Most of the newer DSL routers seem to periodically resync to adapt to line conditions.  The main problem I've had with this is the Internet connection being interrupted during each resync with long cable runs, so I'm using one that seems to stick whatever it synced at until the next reboot. One of my past ISPs use to trigger a resync around 12:30am each day. 
 
If this happens again, one idea would be to use a plug-in timer that cycles the power every night at 3am.

Some DSL routers I've used in the past did something similar where it seems to be related to how they deal with and recover from noise on the line.
 
For example, on Smart Telecom's (former ISP here) earlier DSL router model, if there was excess noise on the line such as the DSL line passing someone's home who operates welders, power tools, etc., it would resync the connection to achieve a 15dB signal-to-noise ratio.  However, when the interference went away, the DSL router remained synced at that lower speed until it was rebooted.  My brother had this issue and had to keep cycling the power each morning to get the speed back to normal.
 
Most of the newer DSL routers seem to periodically resync to adapt to line conditions.  The main problem I've had with this is the Internet connection being interrupted during each resync with long cable runs, so I'm using one that seems to stick whatever it synced at until the next reboot. One of my past ISPs use to trigger a resync around 12:30am each day. 
 
If this happens again, one idea would be to use a plug-in timer that cycles the power every night at 3am.

I love the idea, in fact I'm working on it.
 
The reason it wasn't implemented like that in the first place is because the multithread test came many years after the classic test.  It needed it to prove itself and we needed to better understand how it worked before it could be pushed out front.  I wrote every line of it but that doesn't mean I fully understand it.  Obviously I understand what's behind it but it's taken time to understand it's interactions and how it arrives to reflect the users performance.  Very hard to explain.  It's not just about bandwidth.
 
I'm very confident in the multithread test.  I was confident in its ability when it was released too... but now it's proven itself over time.  The data it presents is far more valuable when combined with the linear test (need to put a name up to a vote... linear, classic, single thread... then stick to one name) results.  The information is so valuable, they need to be combined (obviously the user will have a choice).  In my experience a computer and Internet connection combination that can run similar high values in both tests is always going to run better than one that produces a skewed results.  That's valuable information to know and hardly anyone even knows about the multithread test right now (or this whole site for that matter).  And I bet many who do are confused by it.  Sorry to anyone who's confused, kinda making this stuff up over here.    I don't know the right or wrong way to lay it out because I don't have anything to go off of.  I work on functionality first, presentation last.  
 
Thank you so much for the suggestion.  This thread helps a lot.  You're absolutely right, the information must be presented correctly.  There's already too much information for most people... which is why I intend on releasing another version of TMN.  Same core, ultra basic output.  Not for you, you're an obvious 2%'er... it'll be for the other 98% of the population that look at this site and don't get it. 
 
Keep up the great thoughts and suggestions.  Very much appreciated.
 
Very nice...

 
 

While I absolutely agree that more data can make things harder to understand, it's all about how the data is presented. From what I have seen, there already seem to be a lot of people asking why the results are so different from other speed tests that default to multiple connections. And if there are a lot of people asking, there are a lot more who won't bother. I think if both tests are run, and the results are displayed with a simple explanation of what they mean, that will help people understand why their results are so different.

It could be as simple as:

Single connection:
These are the results of doing one thing at a time, such as downloading a file. Ideally they should be fairly close to what you see from multiple connections.
[Graph and results]

Multiple connections:
These are the results of doing many things at the same time, such as loading all of the parts of a web page. If there is a large difference between both results, your connection may not be optimal. Click here for more information.
[Graph and results]

Easy to understand, if slightly simplistic, it gives a much better picture of the health of a connection then just one or the other testing methodology. And since they are run at very close to the same time, they are more accurate then trying to run both tests separately.

Also, Joe Blow that comes here just wanting to see if his ISP is providing what they say they are, isn't going to bother learning the difference between the tests. He is going to go to one of the first links in a search result, and run the first test he finds and if his provider is limiting speeds per connection, he is going to be very confused.

As the multithread test involves transferring lots of small files, I wonder if it is taking an unusual amount of time to establish the individual connections or if something is briefly pausing the start of each individual transfer.  For example, some antivirus products scan the data at the start of each transfer to check for known infections and this sometimes causes a brief pause before allowing further data to download.  I had issues with AVG in the past causing unusually long delays when downloading files.
 
If something is briefly delaying each connection, then this would likely cause the multi-threaded test to take longer as instead of a brief pause in a single transfer, there would be a brief pause in every individual transfer of the multi-threaded test, so with the 50 or so small files downloaded during the test, these pauses probably add-up more than any speed benefit of running batches of transfers simultaneously through the test.
 
If you can easily disable your antivirus product or firewall, I would suggest trying the multi-threaded test again with these disabled.  The issue could also be within the ISP if they are doing something like deep-packet inspection on each connection that is established. 

I get the impression that they apply a tier-type throttling like satellite based providers, where data is initially delivered at full speed, but once a certain threshold is reached in a rolling interval (e.g. 1 hour), the connection is throttled.
 
The next time your connection is at full speed, you could try capturing a graph of the throttling as follows:
Go into the 'Download Test' tab above.  If it says "Testing Global Multithread", click 'Classic Linear'. Ensure the 'Testing' server is nearby, otherwise click 'Servers', pick the nearest and go back into the 'Download Test' tab. For the 'Manual Test' drop-down, pick a fairly large size that you know will lead to throttling, e.g. '50MB'. Don't do anything online during the test and leave it to finish, e.g. start the test just before dinner.  At the end of the test, you should see a graph like the following:
 

 
If the line is fairly flat for a portion of the graph, then plummets to a much lower level for the rest of the test, then the culprit is most likely your ISP throttling after a certain amount of data is transferred over an hour or some other short interval. 
 
On the other hand, if the level remains fairly low or fluctuates all over the place throughout the graph, then it is most likely contention, such as your Wireless ISP being over-subscribed.  In this case, repeat the test early in the morning to see how it compares.  If the throughput is much better in the morning, then it's very likely you are connected to an over-subscribed tower. 

Personally the ads here don't bother me.  The most irritating ads are the door-way type and those with audio.  Generally if I land on a door-way ad, I click back and try the next link, which votes as a "bounce hit" to Google for the site that served the ad.  The audio ones tend to catch me out at work when I forget to mute the speakers.  I don't use an ad-blocker.
 
I try to spread the word when I can.  Pity I didn't run into this site years ago, as I've completely abandoned Ookla's site since.  I also think this site is more closely watched by many ISPs than Ookla's site.  Whenever my main ISP seems to run into bandwidth issues, a 24-hour Auto Test tends to resolve it, probably because they see the low test result scores don't look pretty in the recent test log.  Even after I posted a Thread on the 3 Mobile network here throttling certain traffic, that throttling seems to have disappeared a few days later and I'm fairly sure it was going on for months. Of course this is probably all coincidental...

It's because of the different wireless adapters.
The iPad 4th generation has less bands while the iPhone 6 has more. This allows the iPhone to take advantage of more bands resulting in faster data speeds.
Only a certain amount of the extra speed is because of the processing speed boost between the two. The iPhone 6 and iPad 4th generation are pretty close in speed anyways. It's the wireless card, or adapter, causing the speed difference.
Hope this helped.
I may be a little wrong, but I'm almost positive I'm right.

Yeah, there is a technical roadblock.  It would only work if the browser is left visibly open.  No problem on desktops but no something that could be asked of a mobile user.
 
I'm looking into iOS and Android app development which should open up more possibilities.  
 
I have an array of new features that mobile users will find useful coming out in my next version.  Opening up a whole new world of possibilities in the next release.  I'm already using many of the hidden features myself right now and it's providing me a much deeper insight.  It will take me time to build systems to display all the additional information in useful ways but once it's ready for prime time I think you'll find it useful.

Hi All Souls,
 
Well ladies and gentlemen, it has finally come. Internet speeds faster than your typical cheap 5 port hub, POE devices, and even too fast for some of your laptops with older NIC cards out there. Behold, 300 mbps / 20 mbps - has finally reached into my area. Although, most webservers peak out at 20mb/s, and netflix only allows each device about 5 mbps from what I can see on my network monitoring. 
 
Evidence Below:
 
Download: (on a good day I can get 320mbps)

 
Upload:

 
They complimentary increased our speeds because competitors in the area had 100mbps offers and also, I believe Google Fiber is looking to expand into my area ... I'll keep you posted.
 
As for the obvious questions, torrenting (legal) stuff is infact awesome with the correct settings in your torrent client, I manage about 22-30 mb/s download rate in torrents.
 
The upload speed increase also seemed to have increased my ping times by about 10ms in most games, such as League of Legends, DirtyB, Counter Strike, even agar.io 
 
At any rate, TWC MAXX is awesome! And after 2mo with no outages, I have nothing but great things to say. It is cable, so I know there are speed variances but a variance from 150mbps-320mbps is pretty nice for $69/mo.. 
 
Anyone else out there get there TWC MAXX upgrade? Oh yeah forgot to mention you need the right modem that can handle all the channels....
 

As the multithread test involves transferring lots of small files, I wonder if it is taking an unusual amount of time to establish the individual connections or if something is briefly pausing the start of each individual transfer.  For example, some antivirus products scan the data at the start of each transfer to check for known infections and this sometimes causes a brief pause before allowing further data to download.  I had issues with AVG in the past causing unusually long delays when downloading files.
 
If something is briefly delaying each connection, then this would likely cause the multi-threaded test to take longer as instead of a brief pause in a single transfer, there would be a brief pause in every individual transfer of the multi-threaded test, so with the 50 or so small files downloaded during the test, these pauses probably add-up more than any speed benefit of running batches of transfers simultaneously through the test.
 
If you can easily disable your antivirus product or firewall, I would suggest trying the multi-threaded test again with these disabled.  The issue could also be within the ISP if they are doing something like deep-packet inspection on each connection that is established. 

His signature now shows the time converted to local time:
 

 
For anyone curious, the following is what the combined test ID bug originally looked.  This screenshot is from my blog as WordPress's JetPack Photon feature has the original image stuck in its cache:
 

 
This might explain why I saw confused Chaffinches out my window, one which I snapped for my avatar.

I only just noticed that when I combine speed test results, it shows "GMT-7" in the image, which is incorrect for me.  For summer time (DST), it's GMT+1 here.
 
When I retrieve the speed test result images individually, the time zone appears fine:
 

 
However, when I combine the above two results, it shows "GMT-7":
 

 
The time itself is correct and matches the clock after the test completes.