Kitz Forum

Broadband Related => Broadband Hardware => Topic started by: ktz392837 on January 19, 2017, 03:52:52 PM

Title: Comparison Of Three Modems
Post by: ktz392837 on January 19, 2017, 03:52:52 PM
Finally got around to doing some soldering and applying the Openwrt firmware to my ECI modem and thought a quick comparison between three modems may be a useful post. 

Each modem was connected for 30 minutes then a copy of the statistics were taken.  No surprise but as this wasn't a very long period of time errorred seconds or corrections were none/few so see little benefit of posting these statistics.  I was "bad" and only left 5 minutes or so between swapping modems as I wanted the time of day to not impact too much on the connection statistics.

The results for my connection was the ECI modem was by far the best (no surprise probably with being connected to ECI cabinet at approximately 450m). 

It is a pity dslstats/MDWS does't support just a basic subset of the ECI data as I would use the ECI without hesitation if that was the case but as it doesn't I am currently using the ZyXEL as there is not a great difference between that and the Huawei and the ZyXEL is a combined unit hopefully using less electricity. 

ECI B-FOCuS V-2FUb/r Rev B (OpenWrt Chaos Calmer)
Max          : 65.520 Mb/s / 19.237 Mb/s
Actual       : 65.157 Mb/s / 19.141 Mb/s
Noise Margin :      6.1 dB / 6.1 dB
Conclusion   :         1st / 2nd

ZyXEL VMG8324-B10A (v15)
Max          : 61.455 Mbps / 18.758 Mbps       
Actual       : 61.358 Mbps / 18.734 Mbps       
SNR Margin   :      6.3 dB / 6.0 dB
Conclusion   :         2nd / 3rd

Huawei EchoLife HG612 3B (Unlocked GUI / No BT Agent)
Max          : 60644 Kbps / 20364 Kbps
Actual       : 60882 Kbps / 20000 Kbps
SNR (dB)     :        6.3 / 6.1
Conclusion   :        3rd / 1st

Title: Re: Comparison Of Three Modems
Post by: burakkucat on January 19, 2017, 05:17:51 PM
Thank you for sharing those results.

I would not have been able to predict outcome of your experiment.
Title: Re: Comparison Of Three Modems
Post by: ktz392837 on January 19, 2017, 06:02:23 PM
Yes it surprised me -  it will be interesting if we do ever get 3db/Ginp on ECI cabs whether the ECI modem will further increase its lead.  I can remember people reporting thinking the ECI modem was much better the last time around before it was removed.

If anyone has their own comparison figures it would be interesting to see how cabinet type, distance from cab and modem type makes a difference to their connection.
Title: Re: Comparison Of Three Modems
Post by: burakkucat on January 19, 2017, 07:04:59 PM
But (and there's always a "but" . . . ) you are not testing the ECI hardware with its default, supplied software. So how much must be down to the OpenWrt code with which your CPE is operating?  :-\

Is it fair to expect the average user to learn how to open up an ECI /r device, solder appropriate header pins to the PCB and then install a new software package?  :no:  Perhaps British Telecommunications plc and ECI Telecom (a private company since 2007) should reconsider of what that hardware is capable.
Title: Re: Comparison Of Three Modems
Post by: ktz392837 on January 19, 2017, 09:06:57 PM
Totally agree - I have not tried to hide the ECI is not standard. 

What is also interesting the modem can quite possibly make a surprising difference to how well your connection performs.  If you have the ability researching different modems may give a non-insignificant performance boost.

Let's hope I will be performing the same tests again in a few months time with 3db/Ginp active ;)
Title: Re: Comparison Of Three Modems
Post by: underzone on January 19, 2017, 09:27:06 PM
Well done ktz392837  ;D
I look forward to seeing the 3dB results at some point. It will be very interesting!
Title: Re: Comparison Of Three Modems
Post by: Weaver on January 19, 2017, 10:45:02 PM
My experience with ADSL1 modems on an ultra-long line many years back showed just how much difference modems can make, and also the same is true for some DSL cables and even micro-filters in certain conditions but not others.

I would expect the results to vary slightly  - although not necessarily enough to affect ranking order - according to distance for VDSL2, because presumably the frequency spectrum will differ as well as the absolute voltages of both signal and noise. It would be interesting to results compared for long vs short VDSL2 lines.
Title: Re: Comparison Of Three Modems
Post by: psychopomp1 on January 20, 2017, 07:50:30 AM
You should also have tried the Draytek Vigor 130 which is arguably the best vdsl2 modem for those on ECI cabinets. It uses a Lantiq chipset, supports G.INP and gives full line stats.
Title: Re: Comparison Of Three Modems
Post by: Chrysalis on January 20, 2017, 08:55:31 AM
does the vigor work with any of the MDWS tools?

On my line both the stock ECI modem and friztbox (which has a lantiq chipset) sync lower than all my broadcom devices.  I am on a ECI cabinet, on the flipside tho they also have lower error rates.

Also of interest is that when interleaved the fritzbox syncs super low, it seems to apply the interleaving sync penalty twice.
Title: Re: Comparison Of Three Modems
Post by: roseway on January 20, 2017, 10:44:27 AM
does the vigor work with any of the MDWS tools?

No, both programs only work with Broadcom-based modems (Vigor 130 is Lantiq).
Title: Re: Comparison Of Three Modems
Post by: skyeci on January 20, 2017, 11:41:25 AM
Broadcom always been best on my  eci line when compared to open reach eci modem that I unlocked.