Increase speed with Internet Booster

IB2 with a copper connection and a throughput of approx. 40 - 50 MBit/s

The Internet Booster is connected to the IB2 via WLAN.

My expectation would be that I can increase my throughput with the Internet Booster.

However, the reality is different: Every device for which the booster is selected in the device list only achieves the throughput of the booster in the speed test, which is well below the Internet Anschluss.

As soon as the booster is deactivated, the throughput increases to the usual value, which is actually what you want to increase.

Are my expectations of “increased throughput” wrong or is there something wrong here?

Show original language (German)

@mf37 wrote:

….

Are my expectations of “increased throughput” wrong or is there something wrong here?

Hi,

in my opinion “no” and “yes”. What about the speed if you connect the kit directly to the IB2 with a LAN cable? And what happens if you allow bonding by default on all devices? And have you already rebooted? I remember that I had this issue once too. But that was with a very early software version. Today I no longer use a bonding kit.

Thomas

Show original language (German)

Hello Thomas

Thank you very much for your answer. I have implemented your suggestion. After I connected the booster to the IB2 via cable, rebooted the IB2 and activated the booster in the settings on all devices, new measurements showed that the booster now provides a good 5 Mbit more throughput. However, the average is still significantly below that for copper. If I remove the activation of the booster, I have the usual throughput again.

It really seems to be the case that the booster does not increase the maximum throughput, but only enables more throughput overall when using several devices in parallel if you assign the devices individually to the booster or the IB2. If you want to increase the throughput of a device, the booster won’t help.

Unfortunately, this is not clear from the description of the Swisscom booster.

Martin

Show original language (German)

@mf37

It can’t be in the description because it wasn’t designed that way.

The bonding proxy on the IB2 bundles two Internet channels using Distributed TCP, i.e. the IB2 distributes the TCP packets from a single client to both channels and incoming packets for a client can also arrive at the IB2 in parallel over both lines and are then reassembled by the proxy for the individual client.

The only known limitation for bonding is (at least it was when I had a booster myself) that it really only works with TCP traffic and not with UDP traffic (e.g. access from outside to a UDP-configured Open VPN server).

Had 50 MBit/sec copper + 50 MBit/sec LTE download for two years and never excluded any device from bonding (except the TV boxes, which are blocked by Swisscom by definition) and never had a problem with 100 MBit/sec download on a client as a direct individual download.

If this doesn’t work for you (e.g. for a normal speed test like CNLab), it’s not due to the bonding in general, but must be due to a local limitation.

By the way, you can easily see how many MBit/sec the booster actually delivers in the LAN diagnosis of the IB2’s expert menu during a long-running speed test, provided you have connected the booster to the IB2 via a LAN-Anschluss.

If you are really interested in reliable speed tests:

- Don’t use a browser but rather the CNLab test program installed locally under Windows (free download from the Internet)

- Always test with the PC connected directly to the IB2 via cable

Show original language (German)

Hobby-Nerd ohne wirtschaftliche Abhängigkeiten zur Swisscom

Thank you very much for the detailed information. Then it should work at some point.

But I have no idea what internal limitation might exist. I only have the IB2 and a switch behind it to get more connections. In this respect, the limitation could only lie with the IB2. I received this new with the booster and did not make any adjustments to the configuration. So I’m at a loss as to what the internal problem could be here.

Show original language (German)

@mf37

If your booster is connected to the switch, or the WLAN is connected to the switch, this would lead to special requirements for the switch in connection with the booster.

If the booster is connected directly to the IB2 with a LAN cable, or directly to the IB2’s WLAN, there should actually be no fundamental configuration problems with the booster…

Show original language (German)

Hobby-Nerd ohne wirtschaftliche Abhängigkeiten zur Swisscom

The booster is connected directly to the IB2 via cable, so I don’t see any reason for problems.

I have now reinstalled the firmware on the IB2, the results are slightly better, but there is still no noticeable combination of copper and mobile. In addition, the throughput in the cnlab test fluctuates greatly.

Show original language (German)

l@“x”#p567162

It is normal for the throughput on a copper line or LTE transmission to fluctuate greatly over the course of a single second of measurement; what is important is the average over a certain period of time.

With the Cnlab test program, it is best to set the measurement duration in the settings to 30 to 60 seconds as standard, then you will receive relevant numbers.

Since you have the Internet booster on the LAN cable, you can see its current performance during operation directly on the LAN port used for it:

To do this, go to the LAN analysis on the IB2.

It looks something like this:46BF3065-D86F-41CC-BE5A-E2E25AD40295.jpeg

By the way, the picture is an old Netflix streaming example of mine to show that, unlike Swisscom TV, the Internet booster contributes generous LTE performance for external video streaming (and constantly, even when the copper line is not yet at capacity is, because conceptually speaking, the Internet booster is actually not a “booster” at all, which, as one might assume, is “switched on when you step on the gas”, but rather a constantly running fix coupled “second motor”)

Shown in the screenshot are:

- the Internet-Booster on LAN port 1 as an upload from the perspective of IB2

- The home network with 2 Smart TVs and two running Netflix apps with UHD streams on LAN port 2 as a download

- the UHD-TV-Box, which was switched off at this time, on LAN port 3

(the total LAN traffic of 51.8 is irrelevant)

Show original language (German)

Hobby-Nerd ohne wirtschaftliche Abhängigkeiten zur Swisscom

Thank you very much for the additional tips. I have now taken measurements with the LAN analysis. I noticed that when measuring the Ethernet Anschluss, over which the booster - IB2 connection runs, it does not show any throughput, even though the booster is running and activated.

Is this correct since the connection is not considered traffic, or is this an indication that there is a problem? (If I swap the cables, the same picture occurs).

Martin

Show original language (German)

You should see it in the menu exactly as @Werner has already shown. The traffic that the booster brings should also be displayed. If nothing is displayed, there is no data traffic going through the port.

\=> So yes, seems to be an indication of a problem.

Thomas

Show original language (German)

@mf37

If the booster never draws data at all, it may be that when you activated your Internet booster, your quota on the LTE mast assigned to you was “lost”, which happened again and again in the pioneering days of bonding.

If this were actually the case, you would only be able to get to the second/third level via support to have the misconfiguration corrected in the background of the LTE network.

Show original language (German)

Hobby-Nerd ohne wirtschaftliche Abhängigkeiten zur Swisscom

14 days later

Update
After support couldn’t find any further information on the phone, a technician came to my house. He also couldn’t solve the problem and didn’t understand why the booster didn’t provide additional throughput. I was then informed by Swisscom that I had to install the booster somewhere else. The connection is too bad and leads to constant interruptions. Although this was not understandable to me (connection display on the booster is solid white, the time is displayed in the IB2; status: connected, network: LTE, signal strength good with 2 green lines), I have now tested the Internet booster in two other locations:
1. Location: same status with 2 green lines and no effect
2. Location: in the garden connected via WiFi, LTE with 3 lines and minimal contribution of 7 - 11 MBit/s

Conclusion: I can’t use the booster because I can’t get a signal strength of 3 lines in my apartment.
But I find it very troublesome that the booster lights up in a stable white color when the connection is still not good enough. This should be adjusted, then I could have saved myself many hours of testing.

It’s amazing to me that 2 strokes aren’t enough and 3 strokes only do so little. Apparently I live in the wrong place for Swisscom: no fiber optics and poor mobile reception.
Thanks again to everyone who gave me tips.

Show original language (German)

Hello @mf37

The choice of location where you place the booster makes a lot of difference.

If I have to install or relocate this at the customer’s house, I first use my Android cell phone with the following app “NetworkCell Info” and walk around the house for miles. If possible, I always make sure to connect it via cable

There you have a level indicator of how strong your cell phone signal is. The booster is not always on the nearest antenna that is directly in front of your house, it can also be the neighboring antenna that is on the other side of your house.

You may find out which antenna you are connected to, and not every antenna has the same power output.

In which area does the pointer move?

Greetings Lorenz

Show original language (German)

Hello Lorenz

Thank you very much for your information. It is absolutely clear to me that reception strength plays an important role. What surprises me is that the booster does not correctly indicate whether the connection is good enough based on the measured reception strength (booster via white/red display and IP2 via signal strength). Apparently 2 lines of LTE are simply not enough, it needs at least 3 lines. Today, however, two lines indicate that everything is good. The color scale of the tool is correct: it only turns green from -98.

I measured inside and outside the apartment and got the following results:

Apartment: LTE -108 to -128

Garden: LTE -96 to - 119

outside the window: LTE -106 to -114

(where possible)

So it’s clear to me that this will never work. I have signals from 3 antennas, but none of them provide enough power to my apartment. Therefore the reception in the apartment is insufficient or just barely sufficient in places. I was hoping that this was due to the thick walls and good windows and that I would have better reception outside.

Thank you again

Martin

Show original language (German)