Print

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

[Weaver]

Forgive me, I forget - is there upstream G.INP on your link now? If so, what does it do for you? You’re getting roughly 4% more downstream speed then.

[forceware]

Has this been spotted by anyone in the wild yet?

[ktz392837]

Far from the wild but 19 Plusnet users have it and it appears to be working for them.  No guarantees though it was working for virtually everyone last time and they still canned it.  Not convinced anywhere near 100000 trial users have it it may only be the 19 on Plusnet at the moment.  More details https://forum.kitz.co.uk/index.php?topic=24403.0

[kitz]

Re-tx low
 
I was getting practically the full 3600 Err Secs per hour for over 3.5hrs.
Aside from a small respite for a couple of minutes at 8:30, I was racking up at the rate of 1 E/Sec per second from ~7:30 until just gone 11am.

INP rein and interleave delay have not been added. 

Code: [Select]

                        Bearer 0
INP:            47.00           0.00
INPRein:        0.00            0.00
delay:          0               0
PER:            0.00            6.15
OR:             0.01            202.87
AgR:            69030.25        20203.27

[kitz]

Forgive me, I forget - is there upstream G.INP on your link now? If so, what does it do for you? You’re getting roughly 4% more downstream speed then.

This is a trial for downstream retransmission on ECI cabs.   ECI's supposedly cant do upstream.
Yes, the potential gains for those on ECI cabs who previously had Interleaving could be even greater.   

[kitz]

As ktz392837 said there are a few of us on here who have it (including myself).   The only cases we have seen so far though are those from Plusnet who Bob submitted and it seems to be working nicely so far for us.   

It was supposed to be ISP opt in which usually means Openreach [randomly] selects how ever many trialists they need from across the ISP's userbase.   I would have thought that at least BTr would have opted in.  That said the average BTr/Sky/TT user is unlikely to notice any difference especially if they are using ISP modems which don't show information about the bearers.

[Chrysalis]

wow trial already suspended thats why not many on it

[burakkucat]

wow trial already suspended

Is it?    Do you have a link to the Openreach announcement, please?

[Chrysalis]

Is it?    Do you have a link to the Openreach announcement, please?

i can pm

[burakkucat]

Thanks. 

[mrk26]

wow trial already suspended

Is ther somewhere officiall info about?

[Alex Atkin UK]

wow trial already suspended thats why not many on it

I kinda expected that, its risky doing anything that might decrease stability during the current situation.

[kitz]

Since that horrendous burst of noise, I thought the DLM may hammer me.    But touch wood it looks like I've gotten away with it 

After I rebooted the other day to clear the noise, when it came back up I noticed my SNRM was at 6.8dB but it immediately dropped down to 6.4 and as my sync was fairly rounded to 68Mbps (67790) and just assumed that it was just the DSM playing with power.   I happened to notice this evening my SNRM had gone back up to 6.8 again so out of curiosity I performed a resync.    Now connecting at 68960 so that's yet another small gain of 1.2Mbps giving a total gain of ~3Mbps since g.inp was enabled on my line.

My daily SHINE is still there as I'm seeing FEC's each morning for 1 min, but for the past few days it hasn't caused any E/Secs

No idea why DLM didn't kick in..  but I'm certainly not complaining. 

I had another stuck SHINE burst this morning.  The email notification took a while to come through as there were that many errors it was difficult to send or receive any data, but I was still able to stop it in about 10 mins. 

Total error seconds for the day was 683...  and the DLM has just taken action.   
Compared to 11,705 a couple of weeks ago when it didn't take any action.

----
Stats

Previous

Code: [Select]

Max:    Upstream rate = 24417 Kbps, Downstream rate = 68365 Kbps
Bearer: 0, Upstream rate = 20000 Kbps, Downstream rate = 68960 Kbps

                        Bearer 0
INP:            47.00           0.00
INPRein:        0.00            0.00
delay:          0               0
PER:            0.00            6.15
OR:             0.01            202.87
AgR:            69030.25        20203.27
Current

Code: [Select]

Max:    Upstream rate = 24776 Kbps, Downstream rate = 67197 Kbps
Bearer: 0, Upstream rate = 20000 Kbps, Downstream rate = 67984 Kbps

                        Bearer 0
INP:            51.00           0.00
INPRein:        1.00            0.00
delay:          0               0
PER:            0.00            6.15
OR:             0.01            202.87
AgR:            68054.16        20203.27


PS. 
I've noticed since I've been on G.INP that my max rate is always below my actual rate despite my SNRM being 6.4 dB. 
1) Is 6.4dB the new Target SNR as every sync I've done since being on re-tx has been at 6.4dB
2) Presumably the modem isn't correctly calculating the overheads for g.inp and making the correct adjustment for coding gain in a similar way to how it doesnt correctly take into account interleave overhead. 




 

[ejs]

I think the target SNRM always was and still is 6.0 dB. If your modems always connect at a slightly higher SNRM, I think that'll just be the modems erring on the side of caution when estimating the rate/SNRM to connect at. Technically speaking, the very definition of SNRM is totally different with G.INP. Without G.INP, SNRM is defined relative to bit error rate (1 error per 10⁷ bits), whereas with G.INP, SNRM is defined relative to MTBE (1 error in 4 hours).

[Alex Atkin UK]

That's an interesting point.  Do the modems actually bother to try to calculate that or just report the SNRm as if G.INP did not exist?