General Discussion
  >> Fibre Broadband


Register (or login) on our website and you will not see this ad.


Pages in this thread: << 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | [19] | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | (show all)   Print Thread
Standard User j0hn83
(member) Sat 04-Feb-17 08:44:48
Print Post

Re: Openreach to rollout 3dB target SNRm for FTTC next year


[re: MrSaffron] [link to this post]
 
are you talking about vdsl2?
if 6dB target gives attainable just above sync
and 9dB target gives attainable way above sync
why does 3dB target give attainable just above sync.

A 9dB target SNRM should give an attainable just above sync (on fastpath). I've never seen a 9dB target though.

It's simple! The attainable changes in a similar pattern to sync when the target snrm is lowered, remaining just above sync. This is the case on 6dB, 5dB, 4dB and 3dB. The same would happen in the other direction, only lowering together rather than increasing. If the attainable ignored the actual target SNRM the above wouldn't happen.

Sometimes, just sometimes, what you assumed to be true is wrong, and someone comes along and points it out. Today that someone is me. I've made around 20 posts on here in the past couple months. 5 or 6 of those posts someone has wrongly tried to correct me. I've been right every time smile

I have no more to say on this matter. I'm peeing into the wind.
Administrator MrSaffron
(staff) Sat 04-Feb-17 10:24:04
Print Post

Re: Openreach to rollout 3dB target SNRm for FTTC next year


[re: j0hn83] [link to this post]
 
Can you point us all at any documentation for this then that details the figure being used for attainable/max rate calculations is told to the modem by the DSLAM, so will vary, and thus reflect the target margin set.

Certainly on ADSL2+ this was NOT the case, and does not seem to be the case on VDSL2.

Edited by MrSaffron (Sat 04-Feb-17 10:36:37)

The author of the above post is a thinkbroadband staff member. It may not constitute an official statement on behalf of thinkbroadband.
Standard User Zarjaz
(eat-sleep-adslguide) Sat 04-Feb-17 10:34:16
Print Post

Re: Openreach to rollout 3dB target SNRm for FTTC next year


[re: MrSaffron] [link to this post]
 
You don't want to see that Mr.S, sounds like there may be wee on it grin


Register (or login) on our website and you will not see this ad.

Standard User R0NSKI
(fountain of knowledge) Sat 04-Feb-17 22:11:31
Print Post

Re: Openreach to rollout 3dB target SNRm for FTTC next year


[re: j0hn83] [link to this post]
 
I happen to think your correct j0hn83 smile I don't think the target is hard coded for the attainable calculation, at least not in most modems, and certainly not in Williams.

PS Anybody using HG612 stats and my GUI should have current stats recorded for each resync (dependant on settings) in the current stats folder - they have RESYNC appended and include a plink log in text format.

Standard User R0NSKI
(fountain of knowledge) Sun 05-Feb-17 09:43:57
Print Post

Re: Openreach to rollout 3dB target SNRm for FTTC next year


[re: R0NSKI] [link to this post]
 
Bear in mind that the stats recorded by monitoring software will not be exactly what they were at time of resync.

For instance, say there is a power cut, Modem A resync's first @ 10:05 and 6 seconds (just after HG612 logs the last stats), this will see minimum interference from disturbers who's modem haven't yet resynced. Modem A will sync much higher speed than usual at 6db, but as other modems come on line the SNR will drop. At 10:06 HG612 stats will run, and see the new resync and generate graphs for it, but the SNR will be much lower due to all the other modems that have connected over the last 54 seconds (so this is not the target SNR). In this situation the attainable will always be lower than the sync, some modems will hang fine for many weeks like this, others will create many errors and the DLM will take action. It depends on a lot of factors. If you look in the Plink log there is a value label AS, this is the uptime in seconds, the closer to 0 the more accurate the reported SNR will be to the taget.

DLStats collects data at 30 second interval's by default, so will likely be more accurate but there still a lot of room for the SNR to change prior to it being logged.

Administrator MrSaffron
(staff) Sun 05-Feb-17 10:14:54
Print Post

Re: Openreach to rollout 3dB target SNRm for FTTC next year


[re: R0NSKI] [link to this post]
 
So if the modem was to resync at 3,6,9,12,15 dB target, we would see an attainable very close or the same as the sync at the moment of sync? Then depending on how the real world SNR vary we would see the attainable just wander around that figure.

The author of the above post is a thinkbroadband staff member. It may not constitute an official statement on behalf of thinkbroadband.
Standard User R0NSKI
(fountain of knowledge) Sun 05-Feb-17 12:52:54
Print Post

Re: Openreach to rollout 3dB target SNRm for FTTC next year


[re: MrSaffron] [link to this post]
 
Yes, as long as the line is on G.Inp or fast path. As far as I'm aware BT FTTC doesn't have a target higher than 6, but if it did it would react the same.

If the line is at maximum sync or banded then the SNR will be higher, and the attainable will be higher than the sync.

If the attainable is lower than sync then then SNR has drifted downwards below the target SNR, that is my understanding.

Administrator MrSaffron
(staff) Sun 05-Feb-17 13:52:20
Print Post

Re: Openreach to rollout 3dB target SNRm for FTTC next year


[re: R0NSKI] [link to this post]
 
So how attainable works in modems is different for ADSL2+ and VDSL2?

The author of the above post is a thinkbroadband staff member. It may not constitute an official statement on behalf of thinkbroadband.
Standard User j0hn83
(member) Sun 05-Feb-17 16:13:33
Print Post

Re: Openreach to rollout 3dB target SNRm for FTTC next year


[re: MrSaffron] [link to this post]
 
I know very little about the attainable with regards to ADSL2+. Every observation I've made with VDSL2 circuits is the attainable moves in conjunction with the target SNRM.

If the target is 6dB and the DSLAM has a reboot and the EU syncs say 10mb higher, the attainable remains what it usually is, considerably below the higher sync.

But as we're seeing this trial progress, the attainable is rising with the sync, suggesting it's using the actual target SNRM. If it was fixed at 6dB then as DLM lowered the target SNRM we would see the sync rise, but the attainable would stay the same.

This is definitely the case with modems we can record stats properly with, Broadcom & Lantiq chipsets are a couple that come to mind.

User A usually syncs at 49.9mb with attainable of 50mb
User B has identical stats.
User A's DSLAM reboots, they sync at a higher rate, still with a 6dB target.
User B resyncs as part of the SNRM trial on say the 3dB profile.
User A's new sync is 59.9mb, the attainable remains 50mb
User B's new sync is 59.9mb, the attainable rises to 60mb.

This is what is being seen in this trial. Usually the attainable wouldn't rise, as we see User A's scenario all the time.

Other chipset modems may not follow this, and may well use a fixed 6dB to work out the attainable. I really don't know as it's much more difficult to monitor locked down modems. Many ISP modems provide very little in the way of stats via GUI's that often display back to front figures, or just plain incorrect figures.
Standard User R0NSKI
(fountain of knowledge) Sun 05-Feb-17 16:26:21
Print Post

Re: Openreach to rollout 3dB target SNRm for FTTC next year


[re: MrSaffron] [link to this post]
 
I've no idea with ADSL, but that is my observations and understanding with VDSL.

I would think that wwwombat would be able to give a much more technical answer.

Pages in this thread: << 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | [19] | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | (show all)   Print Thread

Jump to