General Discussion
  >> Fibre Broadband


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


Pages in this thread: 1 | 2 | 3 | [4] | 5 | (show all)   Print Thread
Standard User NormanB
(experienced) Sat 08-Jun-13 12:10:16
Print Post

Re: HG612 Firmware


[re: BatBoy] [link to this post]
 
Loss of Sync not Signal?

plusnet -> entanet -> Andrews & Arnold -> BE -> plusnet Fibre
Standard User BatBoy
(legend) Sat 08-Jun-13 12:13:33
Print Post

Re: HG612 Firmware


[re: NormanB] [link to this post]
 
No I think it's loss of signal, obviously you will lose sync though so it's the same thing wink


__________________________________________________________________________The back pedalling starts here__________________

Edited by BatBoy (Sat 08-Jun-13 12:16:04)

Standard User ryant704
(member) Sat 08-Jun-13 12:16:09
Print Post

Re: HG612 Firmware


[re: NormanB] [link to this post]
 
It is Loss of Signal and not Loss of Sync. smile


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

Standard User WWWombat
(fountain of knowledge) Sat 08-Jun-13 12:17:14
Print Post

Re: HG612 Firmware


[re: StephenTodd] [link to this post]
 
LOS - loss of signal
LOF - loss of framing

These occur when the modem doesn't detect the appropriate bits in the inbound data stream, though I'm not sure exactly which bits these are right now. Certainly it is more to do with the control data that surrounds the blocks of user data, rather than problems in the user data itself.

RSUnCorr is the count of the RS blocks that have uncorrectable errors, which is subtly different from the number of errors themselves. An RS block is the smallest unit that the FEC process works on, and has size "N" (in the modem stats).

RSCorr is therefore the number of RS blocks that had correctable errors.

RS is just the count of RS blocks.

OHFErr is a count of the number of X blocks or frames (can't remember the name of X) where CRC checks fail, but these are considerably larger than the RS blocks (variable, but perhaps up to 100x). A single RS block that is Uncorr is always enough to cause an OHF frame to count an error, but there can be multiple faulty RS blocks within the same OHF frame. Therefore OHFErr is always <= RSUncorr.

PFE = Path Far End (CRC is calculated per path)

FECS - correct, but I've never seen it either.

Your ES, SES, and UAS description are correct (strictly it is a CRC error that triggers these, but an uncorrected RS error always leads to a CRC error), but not quite complete. The counts also increment if LOS and LOF happen, but they're not so important in the grand scheme of things.
Standard User WWWombat
(fountain of knowledge) Sat 08-Jun-13 12:23:37
Print Post

Re: HG612 Firmware


[re: BatBoy] [link to this post]
 
In reply to a post by BatBoy:
"The SIN includes FECS-LFE" implies FEC/RSCorr rate is used.

Indeed - but perhaps it is used more in the decision whether to reduce intervention, rather than when to add to it.
Standard User StephenTodd
(experienced) Sat 08-Jun-13 13:16:29
Print Post

Re: HG612 Firmware


[re: WWWombat] [link to this post]
 
Thanks everyone for the very helpful replies.

FECS is clearly closely related to FEC, but not quite the same. Does FECS appear in the modem stats?

Is it effectively the same as ES? Whoops, I can see it isn't as FECS relates to corrected errors and ES uncorrected errors.

--
Moved (with trepidation turned relief) to BT Infinity 2 for upload speed. Happy BE user for several years.

Edited by StephenTodd (Sat 08-Jun-13 13:20:14)

Standard User BatBoy
(legend) Sat 08-Jun-13 13:19:19
Print Post

Re: HG612 Firmware


[re: StephenTodd] [link to this post]
 
FECS are for corrected errors.
ES are for uncorrected errors.


__________________________________________________________________________The back pedalling starts here__________________
Standard User Chrysalis
(legend) Sat 08-Jun-13 15:58:16
Print Post

Re: HG612 Firmware


[re: StephenTodd] [link to this post]
 
is it FEC second count?

Forward Error Correction Seconds.

BT Infinity 2 Since Dec 2012
Standard User StephenTodd
(experienced) Sat 08-Jun-13 16:00:48
Print Post

Re: HG612 Firmware


[re: Chrysalis] [link to this post]
 
Yes, I we think that FECS is FEC second count as you say.
But the modem doesn't seem to report that as part of its statistics (eg from xdslcmd info --stats).

--
Moved (with trepidation turned relief) to BT Infinity 2 for upload speed. Happy BE user for several years.
Standard User Alexir
(regular) Thu 13-Jun-13 10:12:49
Print Post

Re: HG612 Firmware


[re: StephenTodd] [link to this post]
 
Finally getting round to unlocking the HG612 and and wondering if there is a preferred firmware to upload? This is my shipped version http://goo.gl/xAPJf If there are different unlocked versions I'd rather go for the most stable one, I'm not looking for bells and whistles, just an unlocked modem with basic stats. I am slightly nervous about bricking it, hopefully Asbokid's directions are foolproof but if anyone wants to offer any further unlocking tips feel free to throw them at me - if you think you're advice will be too obvious, don't worry it won't be!
Pages in this thread: 1 | 2 | 3 | [4] | 5 | (show all)   Print Thread

Jump to