Speedtest limited to about 500 Mbit download?

thomasmerz

Why can't my Fingbox measure any speedtest much more than about 500 Mbit on my Gigabit (1000 MBit) connection? Upload is always correctly measured with around 50 Mbit.

If I use any other speedtest on my Macbook Pro (for example Ookla), I'm measuring up to 700 Mbit via WiFi. Why can't my Fingbox speed-up to the 1 Gigabit via LAN that my ISP provides me? Fingbox says it's connected to 1 Gigabit so it should be able to speed-up to 1 Gigabit if also my Devices can this via Wifi.

Chris_MLAB

M-Lab is a very different Internet measurement platform than Speedtest.net/Ookla, so it is not unexpected that our measurements are different and sometimes lower than Ookla. M-Lab values and encourages people to run both our test and Ookla's regularly, as they measure different aspects of your connection. Regarding the suggestion that M-Lab's test caps at ~500Mbps, this is demonstrably false-- we regularly see very fast measurements above 1Gbps, which anyone can query for in our public dataset. For details on how to gain free query access to M-Lab data, please see: https://www.measurementlab.net/data/ .

Though the metrics (download/upload speed for example) are the same as other tests, the architecture of M-Lab's platform and our test methodology is not. Further, what M-Lab's test, NDT, measures is also different than what Ookla's. You can read more about why this test measures differently than other platforms in this FAQ and this blog post. More details follow.

How M-Lab measures the Internet

Running an exclusively off-net platform, M-Lab currently hosts three measurement services, of which the most widely known is the Network Diagnostic Tool (NDT). NDT measures the single-stream performance of a connection’s capacity for bulk transport, as defined by IETF RFC 3148. NDT reports the upload and download speeds and latency metrics. When using M-Lab’s NDT data, people should keep in mind these points of clarification:

1. Off-net vs. on-net measurement: All of M-Lab’s measurement services, including NDT, are hosted on our off-net platform. “On-net” refers to measurements performed on the same network as the network it is measuring, such as an Internet Service Provider (ISP) measuring itself. It only captures one segment of any path that data is likely to be traversing. In contrast, “off-net” measurements extend beyond a user’s access provider’s network to measure the complete path across the Internet from user to content including interconnections. By definition, on-net measurement can not even detect the effects of any performance limitations at interconnects between ISPs. All of the measurement services hosted by M-Lab inherit the off-net platform methodology for nearly all users.
2. Link capacity vs. bulk transport: When using NDT tests specifically, Internet users are sometimes confused when their individual results don’t confirm the speeds promised by their Internet service provider. “Speed” is often associated with "link capacity," which is the maximum bitrate of a link; in other words, the best performance possible. However, NDT measures “bulk transport capacity” -- the rate that TCP can deliver data across the end-to-end path; in other words, the reliability of that connection. It is important to note that many link problems (such as low level packet loss and reordering) typically adversely impact both MLab measurements and real application performance.  These two ways of measuring performance, link capacity and bulk transport capacity, are different and are often conflated when both concepts are referred to as “Internet speed.” When using NDT data to discuss speed, it is important to clarify these terms to have more effective conversations about Internet speed.

3. Single-stream vs. multi-stream tests: NDT measures the single-stream performance of bulk transport capacity. While modern web browsers will use multiple streams of data, testing for multiple streams can compensate for data delivery problems that are exposed by a single stream. A multi-stream test can return measurements closer to link capacity but it would not represent the adverse performance impact of low-level packet loss. By testing for single-stream performance, NDT is an effective baseline for measuring a user’s Internet performance.

I hope this clarifies the differences between our test results and Ookla's. Both are valid measurements, they're just measuring different things. If you have additional questions please let us know.

 

Best, Chris

M-Lab Support - [email protected]

Robin_Ex_Fing

Hi @thomasmerz

We have been in contact with our Speedtest partners, M-Labs. They confirmed that they have recently rolled out an upgraded of their infrastructure & migration of machines to a new platform. This is being done to improve the overall Speedtest performance going forward. By doing this, initially they have cut of the multi-stream download Speedtest that we rely on for our Speedtests. However, they have greatly improved the single stream download Speedtest (previously the single stream download could measure up to 250 Mbps, where now it can measure up to 500+ Mbps). With speeds above this, the measured speed reported will be lower than before, due to the single stream limits.

To note for Fingbox users: this will also affect Fingbox Speedtests. We have released a new firmware version for the Fingbox which will allow single stream download to work again, firmware version 1.14.13 and onwards. If you have not yet received the new release, please power-cycle the Fingbox, this will automatically update the firmware on the Fingbox.

M-Labs are working on re-introducing the mutli speed download speedtest, to include it again in the future. Currently, we have no time-frame on when this will be re-introduced. Unfortunately at this stage, there is no further steps we can take to expedite this process.

thomasmerz

Here's Ookla:

https://www.speedtest.net/result/10183353236.png

And here's MLAB:

For real Fingbox-Support I could sent them my IP-Adress so you can check with MLAB more speedtest from IP. I won't post it here for data privacy.

thomasmerz

I already have 1.14.19 running on my Fingbox :-)

PeterP

I also have the latest firmware on my Fingbox, but continue to get very poor speed test results. M-Labs single stream test is garbage. Those of us with home networks are interested in keeping our ISPs honest and we typically have multiples users. A single stream test provides no value.  I have 1GB service from Comcast and see over 800mbps speed tests from other sources, but the Fingbox test never shows more than 160mbps, totally useless.

TheDawgFather_48

I am all about knowing/reviewing/investigating different speed results from different servers.

A very easy solution which I would be ecstatic to see deployed is a variety of set speed tests that are completed one after the other and each result is displayed to show the different speeds results.

If there is a specific difference with each test site as your article describes, a info icon could be beside each provider to provide data context.

I would love this feature. Show me the highs from my ISP but also the lows from a single stream source.

thomasmerz

@Chris_MLAB , do you have an news regarding speedtesting? It's till stuck at about 500 Mbit which is only half of my 1000 Mbit that I could speedtest (and which my computers connected via LAN are testing at Ookla for example)…

PeterP

Mlab is a joke, single stream speed tests are totally useless. It’s been like this for over two years and nothing has been done to resolve the situation. It’s obvious that Fing doesn’t care either. If they did, they would dump Mlab and use a different service.

thomasmerz

Single Stream Speed Tests are "what users usually get" while surfing/downloading etc.

But such a Speed Test should measure the maximal throughput! I totally agree that this SSTS doesn't make sense at all

I better switch to a Raspberry Pi 4 with Gigabit-Ethernet and speedtest with Ookla…

RHH

Do we need to pay a monthly fee to unlock the real speed test ? 😜 or will the hardware we have purchased actually do what it’s advertised to?

Must be the actual FingBox as if I use FingApp speed test on phone or even the MLab speed test I get very accurate readings, FingBox with direct cable to router is about half the gig bandwidth i know the router is receiving, as can confirm that in the modem and router, just the FingBox can’t handle it.

thomasmerz

@Chris_MLAB , do you have an update for the community here after more than half a year regarding doing/providing better speedtests with Multiple Stream Speed Tests (MSST) instead of Single Stream Speed Tests (SSST)?

This would really be a pleasure to check the real speed provided by ISP with wired Fingboxes directly to Modem/Router/ISP. Running manual speedtest on speedtest.net at my Gigabit-Connection measures up to nearly 1 GBit instead of max. 500 Mbit via Fingbox Speedtest.

PeterP

MLAB speed test sucks, plain and simple. Prior to their methodology change, my Fingbox would report over 800 mbps speed tests. After MLAB changed to single stream, my Fingbox, Fing desktop and MLAB direct though a browser all report no more than 155 mbps downstream speed. On speedtest.net, I get around 940 mbps (I have 1GB service from Comcast) which is close to the best you can get.  Many, including myself, have complained to the Fing folks about this, but they clearly have no interest in changing speed test providers.

MIBSWE

Based on my test, MLAB isn't the ISSUE. Using their Chrome plugin, it identifies the correct speed. With Desktop Fing, it's way below what is expected. I find it concerning that Fing aren't at least updating what they're doing to solve this. Is the company still an ongoing concern? Is anyone from Fing still on here? 

Karl_From_Fing

Hey @MIBSWE
We are still monitoring the community to the best of our abilities but for any and all concerns that need to be troubleshooted it is a requirement for the correct support channel to be reached. The community is intended to be used for users to come together and talk technical, and of course it is a great opportunity to discuss some issues you may be facing as others may have faced and rectified similar. I would be happy to open a support ticket for you for your concerns if you'd like?

MIBSWE

Hi @Karl_From_Fing!

Thanks for your reply. Yes please, can you create a support case? The reason I didn't use that route at the first instance is that the issue seems to have been ongoing for a very long time without any resolution (others have reported the issue based on comments in the community) or even an update from the Fing team. 

I use Speed tests to ensure that my ISP is delivering what's promised. It worked well on my original Fing box when I was on 500Mbits/s where I could identify a consistent failure in delivering the contracted speed in the evenings. 

An example is I've used Fing Desktop right now and it recorded a download speed of 501 MBits/s and when I use MLabs Chrome extension, it records 781 Mbits/s. That's a big difference.

Mlabs - https://share.getcloudapp.com/5zu2jY4k (Via 2019 Macbook Pro connected via RJ45 to router)
Mlabs via Fing Desktop - https://share.getcloudapp.com/8LugqoRj (Running on 2019 MacBook Pro connected via RJ45 to router)

Average performance over 7 days - https://share.getcloudapp.com/RBuBrL6k. Ironically, it did seem to be correct for a short period of time yesterday and on the 11th, but on one day it was 23MBits/s. Considering I have children watching netflix, YouTube and we're streaming as well, that would be clerly shown but all services have run very well.

Regards,
Mike