Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Edinburgh_lad]

My line within 'acceptable parameters' (3 days but all these errors from 11am this morning), whatever is adding to the noise eventually made the line drop.

[Edinburgh_lad]

(Indirectly) encouraged by this community, I connected my Zyxel once again, to see how it fares against the Fritzbox.

I'm connected to a Huawei cabinet, but my line has an 'interference' issue (SHINE type, rather than REIN type). Stats included below. Not much of a difference in terms of errors etc., apart from that Zyxel can so far keep up the connection. But the real test will be this week, when my regular Teams calls resume.

My conclusion so far is that if there's an issue on the line, no matter what technology you try, the technology (PhyR) won't make the issue go away.

[tubaman]

Interesting to see that you have G.INP running on the upstream side as that is quite unusual. I've had it once or twice on the upstream side of my line following high noise events, but it never lasts more than a day or so before being removed again.

[Edinburgh_lad]

Interesting.

Is G.INP then automatically applied by the DSLAM equipment on the basis of the performance on the line over a period, or by ISP or by Openreach (engineer)?

[Weaver]

With enough SNRM and enough interleave, you should be able to withstand pretty much anything. It may be that the situation you have doesn’t produce a large enough error count to persuade BT DLM to take strong enough action - it’s not taking the SHINE seriously enough. Someone with FTTC has little control over these things though, so I believe- I may be wrong.

Note: you cannot have PhyR; you will have G.INP instead if you have VDSL2 / FTTC. PhyR is confined to ADSL users, not VDSL2 / FTTC. The two do the same thing though.

[tubaman]

Interesting.

Is G.INP then automatically applied by the DSLAM equipment on the basis of the performance on the line over a period, or by ISP or by Openreach (engineer)?

It's automatic and is controlled by DLM.
If the SHINE type noise is reasonably regular you may find it'll leave it on the upstream side. It'll certainly be interesting to see.

[Edinburgh_lad]

Interesting. It's been like that since openreach number 17 swapped pairs for the nth time last week. Senior openreach visit 2 (openreach visit 18) scheduled for this Thursday, so let's see, though I'm not very hopeful.

[j0hn]

My conclusion so far is that if there's an issue on the line, no matter what technology you try, the technology (PhyR) won't make the issue go away.

PhyR is only used on some BT Wholesale ADSL DSLAM's in the UK.
It isn't used on FTTC/VDSL2. That uses G.INP instead.

G.INP isn't proprietary like PhyR so it is used by both the Fritzbox and the Zyxel with your Broadcom DSLAM.

Looks like your downstream is capped/banded at 55Mb.

[Edinburgh_lad]

So, would this suggest that it doesn't matter if you're using Zyxel or Fritzbox? Also, I note that the new Fritzos to be released contains an improvement to a feature, which I think is GINP then:

DSL: Improved Better interoperability with ADSL2+ and VDSL remote sites

Is the Fritzbox going to be on par with Zexel then, except getting better?

[Weaver]

Regarding your question about "is the Fritzbox going to be on par with "ZyXEL" - I can answer this as I have no experience with Fritzboxen. Some ZyXEL modem models are substantially better than others on certain lines due to the presence of a high quality filter in the analog front end. However, I assume this doesn’t apply in the case of FTTC, as lines are all so incredibly short. Can anyone else comment on that?

Anyway, all ZyXELs are not created equal at least in the case of G.992.1/G.992.3/G.992.5. I get a feeling that the front-end filter component increases the build cost for ZyXEL significantly.

[tubaman]

Regarding your question about "is the Fritzbox going to be on par with "ZyXEL" - I can answer this as I have no experience with Fritzboxen. Some ZyXEL modem models are substantially better than others on certain lines due to the presence of a high quality filter in the analog front end. However, I assume this doesn’t apply in the case of FTTC, as lines are all so incredibly short. Can anyone else comment on that?

Anyway, all ZyXELs are not created equal at least in the case of G.992.1/G.992.3/G.992.5. I get a feeling that the front-end filter component increases the build cost for ZyXEL significantly.

The versions with the front end filter certainly give an improvement on my FTTC line (~800m long), which is why I stick with my VMG8924-B10A despite it no longer getting any firmware updates.
Whether it is the filter or other hardware or firmware differences that do this is hard to tell of course.

[Weaver]

Wow, thanks Tubaman, I didn’t know that. So the ZyXEL good AFE is not solely relevant to long lines. That's good to know.

Now I come to think of it, Kitz herself is a relevant case too, no? FTTC and a ZyXEL with the good AFE filter? And so is Burakkucat now.

[dee.jay]

The versions with the front end filter certainly give an improvement on my FTTC line (~800m long), which is why I stick with my VMG8924-B10A despite it no longer getting any firmware updates.
Whether it is the filter or other hardware or firmware differences that do this is hard to tell of course.

Depending on the use case, if one is using that simply as a modem then it wouldn't be the end of the world if there were no firmware updates...

[burakkucat]

Now I come to think of it, Kitz herself is a relevant case too, no?

Yes. Kitz has a relatively short line, to a cabinet with ECI electronics, and a regular daily burst of SHINE.

[Edinburgh_lad]

It's automatic and is controlled by DLM.
If the SHINE type noise is reasonably regular you may find it'll leave it on the upstream side. It'll certainly be interesting to see.

G.INP value indeed disappeared (= become 0) on the upstream, but the number of ES has increased as a result since then.