Kitz Forum
Broadband Related => Broadband Hardware => Topic started by: roseway on April 10, 2017, 11:34:44 AM
-
I've had a strange report from a DSLstats user with this model: he finds that the SNR per tone data is offset by 32 tones. In other words, the data reported for tone 0 is actually the data which applies to tone 32, and so on up the scale. You can see this in DSLstats if the SNR per tone graph is combined with the bitloading graph, where the offset is quite obvious. See here:
https://www.dropbox.com/sh/926q88am0vlu31p/AABbvSS3bCYr_TXZ8eg6hckfa?dl=0
Do other users of this model see the same anomaly?
-
Mine is fine on a VMG8924-B10A.
A different model, but I believe the firmware is pretty much identical between the VMG 1312, 8324 and 8924 range.
-
It's a different Broadcom chip in the VMG1312-B10D to the other models though, which uses different Broadcom DSL PHY firmware.
-
Mine is fine on a VMG8924-B10A.
A different model, but I believe the firmware is pretty much identical between the VMG 1312, 8324 and 8924 range.
The VMG8924-B10A and VMG8324-B10A use identical firmware. You would likely brick your VMG1312-B10(A/D) trying to use the 8x24 firmware, it's a completely different device.
-
The VMG8924-B10A and VMG8324-B10A use identical firmware. You would likely brick your VMG1312-B10(A/D) trying to use the 8x24 firmware, it's a completely different device.
Whoops. Got that wrong. I thought the B10A used the same chip as the 8x24 and it was the only B10D that was different. However having checked on the ZyXel site I can see the B10A firmware although with a similar nomenclature to the 8x24 series is different. Apologies for any confusion caused.
-
Do other users of this model see the same anomaly?
No, not I. :no:
-
Thanks. :)
-
I dont't have a Zyxel running but some time ago my bitloading graph looked similar (bcm6368, PHY: A2pv6C030h):
(https://abload.de/thumb/bitloading-2014-06-14k1un1.png) (https://abload.de/img/bitloading-2014-06-14k1un1.png)
Now with version A2pv6C038q it looks fine.
-
Thank you. So it's possibly related to the firmware version.
-
My VMG1312-B10D is fully up-to-date.
The current firmware version is: V5.11(AAXA.4)CO
From the BusyBox shell --
$ xdslctl --version
xdslctl version 1.2
ADSL PHY: AnnexA version - A2pvI042c.d26d
$
-
Thanks for that. I believe that the model which showed the issue was supplied by Zen, but I could be wrong. I'll check back with him.
-
I believe that the model which showed the issue was supplied by Zen, but I could be wrong.
Ah, now that is interesting for the one I own was a Zen supplied device. The original owner never actually used it and sold it as essentially new.
One of the first things I did was to update the firmware with the latest available from the ZyXEL Download Library (http://www.zyxel.com/uk/en/support/download_library.shtml) and then to clear the ROM-D, which is where the ISP/CP customisations are held.
It is only recently that I have been able to use your DSLstats utility (the -5.9-experimental version) and so all previous plots, the Bitloading, the SNR, the QLN and the Hlog per sub-carrier, were created with my standalone graphing utility. I have never been aware of any misalignment.
I attach both the Bitloading per sub-carrier and the SNR per sub-carrier plots, which were created earlier today.
-
The plot thickens. Of course, you're using ADSL2, whereas the original reporter and morphium are using VDSL2, so that's another variable which might be relevant.