Kitz ADSL Broadband Information
adsl spacer  
Support this site
Home Broadband ISPs Tech Routers Wiki Forum
 
     
   Compare ISP   Rate your ISP
   Glossary   Glossary
 
Please login or register.

Login with username, password and session length
Advanced search  

News:

Pages: 1 ... 13 14 [15] 16

Author Topic: ZyXel VMG8324-B10A Preliminary findings.  (Read 116468 times)

roseway

  • Administrator
  • Senior Kitizen
  • *
  • Posts: 40933
  • Penguins CAN fly
    • DSLstats
Re: ZyXel VMG8324-B10A Preliminary findings.
« Reply #210 on: February 11, 2015, 08:35:56 AM »

For the moment I've taken down the VMG8324-B10A and put back the 8800NL. The reason is that the Zyxel loses its network connectivity after a few days. In my case I could no longer connect my Sky box (Catch-up TV), an internet radio, a laptop or a Raspberry Pi. All of these use DHCP, whereas the two PCs which continued working have fixed IP addresses. So I believe that it's a DHCP problem. This is using the 6b1 firmware.

I'll come back to this when I have more time, but for the moment it's the 8800NL for me.
Logged
  Eric

les-70

  • Kitizen
  • ****
  • Posts: 1255
Re: ZyXel VMG8324-B10A Preliminary findings.
« Reply #211 on: February 11, 2015, 01:03:18 PM »

It is most probable that you will get back fastpath in a few days time ( ~10days) as your ES error rate with interleaving has been close to zero.  Hopefully the Billion will then hang on fast path.
Logged

roseway

  • Administrator
  • Senior Kitizen
  • *
  • Posts: 40933
  • Penguins CAN fly
    • DSLstats
Re: ZyXel VMG8324-B10A Preliminary findings.
« Reply #212 on: February 11, 2015, 01:19:07 PM »

I wonder whether DLM takes FECs into account in deciding whether to restore fastpath after it's applied interleaving? That would have some logic, because FECs are an indication of how much error correction is needed to maintain the low ES level.
Logged
  Eric

Ixel

  • Kitizen
  • ****
  • Posts: 1208
Re: ZyXel VMG8324-B10A Preliminary findings.
« Reply #213 on: February 11, 2015, 02:56:39 PM »

I wonder whether DLM takes FECs into account in deciding whether to restore fastpath after it's applied interleaving? That would have some logic, because FECs are an indication of how much error correction is needed to maintain the low ES level.

It would be a bit stupid if DLM did, as FEC errors are also an unreliable statistic unfortunately. ASUS for example produce a stupidly high number. Different modems produce different amounts FEC errors (some significantly higher or lower than others). A high number FEC errors doesn't always mean that there's going to be a high number of CRC errors if the connection goes back to fastpath. Anyway, I thought I recall kitz saying that DLM doesn't consider FEC errors at some point?
Logged

les-70

  • Kitizen
  • ****
  • Posts: 1255
Re: ZyXel VMG8324-B10A Preliminary findings.
« Reply #214 on: February 11, 2015, 02:57:27 PM »

 While nothing seems to be 100% certain re the DLM, the information available suggests that it does not consider FEC.  That said a number of people with low ES rates remain interleaved.  That may just be the DLM having a long memory of the line history?
Logged

Ronski

  • Helpful
  • Kitizen
  • *
  • Posts: 4170
Re: ZyXel VMG8324-B10A Preliminary findings.
« Reply #215 on: February 11, 2015, 03:58:17 PM »

For the moment I've taken down the VMG8324-B10A and put back the 8800NL. The reason is that the Zyxel loses its network connectivity after a few days. In my case I could no longer connect my Sky box (Catch-up TV), an internet radio, a laptop or a Raspberry Pi. All of these use DHCP, whereas the two PCs which continued working have fixed IP addresses. So I believe that it's a DHCP problem. This is using the 6b1 firmware.

I'll come back to this when I have more time, but for the moment it's the 8800NL for me.

Funny you should mention this, went home the other day and my phone wasn't connecting to the wifi, but my PC was fine. Then my PC stopped connecting to the internet. Couldn't access the router so had to reboot it, then all was well again. I'm sure I've had this one before.
Logged
Formerly restrained by ECI and ali,  now surfing along at 390/36  ;D

Chrysalis

  • Content Team
  • Addicted Kitizen
  • *
  • Posts: 6684
Re: ZyXel VMG8324-B10A Preliminary findings.
« Reply #216 on: February 11, 2015, 04:20:47 PM »

ronski you not considered the 8800nl?
Logged
AAISP - Billion 8800NL bridge & PFSense BOX running PFSense 2.4 - ECI Cab - LINE STATISTICS CLICK HERE

Ronski

  • Helpful
  • Kitizen
  • *
  • Posts: 4170
Re: ZyXel VMG8324-B10A Preliminary findings.
« Reply #217 on: February 11, 2015, 05:59:52 PM »

No I haven't, I'll carry on with the ZyXel for the time being.  I do need to update the firmware to the latest version on the ZyXel. That's something I need to do when I have a bit more time as everything needs setting up again, it's not simply a case of restoring a backup of the settings.
Logged
Formerly restrained by ECI and ali,  now surfing along at 390/36  ;D

jid

  • Content Team
  • Kitizen
  • *
  • Posts: 1945
Re: ZyXel VMG8324-B10A Preliminary findings.
« Reply #218 on: February 11, 2015, 09:10:26 PM »

Likewise I had issues with the stability of the Zyxel so will be selling mine.
Logged
Kind Regards
Jamie

BT FTTP - 75meg | Sky Q |  Bridgend Weather

NewtronStar

  • Kitizen
  • ****
  • Posts: 4854
Re: ZyXel VMG8324-B10A Preliminary findings.
« Reply #219 on: February 11, 2015, 10:18:18 PM »

A high number FEC errors doesn't always mean that there's going to be a high number of CRC errors

The way I look at FEC's counts is that they are more sensative to noise on a line than CRC's as a forward error correction is the first line of defence against corrupted data and then it's over to CRC cyclic redundancy check to fix the corrupted data so if your getting 10,000 to over 90.000 FECs per minute this says your line has a medium amount of noise and i very much doubt the DLM would be able to use non-interleaved in this case
« Last Edit: February 11, 2015, 10:42:21 PM by NewtronStar »
Logged

WWWombat

  • Kitizen
  • ****
  • Posts: 1674
Re: ZyXel VMG8324-B10A Preliminary findings.
« Reply #220 on: February 12, 2015, 01:35:21 PM »

I wonder whether DLM takes FECs into account in deciding whether to restore fastpath after it's applied interleaving? That would have some logic, because FECs are an indication of how much error correction is needed to maintain the low ES level.

Contrary to other people's opinions, I *do* think DLM pays attention to FEC counts in some way. I recently commented on someone's line (probably on TBB), where the line had an uptime of over 70 days, and their ES count in that time was still below 2800. That ought to have been a recipe for DLM de-intervention, right?

However, their FEC count was increasing at a hell of a rate - I don't recall exactly, but it might have been multiple-tens per second, possibly low-hundreds per second. This was enough to show that the current DLM settings were doing the perfect job.

I recall that SIN 498 requires that modems be able to report the FECS counter (FEC-seconds), which is something we never see in telnet or GUI output. I assume it works in the same way that ES's accumulate WRT the CRC counter. The fact it was required makes me suspicious that DLM does indeed use it for some purpose.
Logged

Chrysalis

  • Content Team
  • Addicted Kitizen
  • *
  • Posts: 6684
Re: ZyXel VMG8324-B10A Preliminary findings.
« Reply #221 on: February 12, 2015, 03:00:37 PM »

no for DLM to remove interleaving the ES has to be below 10% of the normal limit, so e.g. 280.

Also if they on the standard profile its just 140 allowed.
Logged
AAISP - Billion 8800NL bridge & PFSense BOX running PFSense 2.4 - ECI Cab - LINE STATISTICS CLICK HERE

burakkucat

  • Global Moderator
  • Senior Kitizen
  • *
  • Posts: 34441
  • Over the Rainbow Bridge
    • The ELRepo Project
Re: ZyXel VMG8324-B10A Preliminary findings.
« Reply #222 on: February 12, 2015, 03:19:38 PM »

Following on from Wombat's posting . . .

So you believe that the rate of change of the FEC count has a significance to the DLM process. That is something I have never previously considered but after a quick think, I would agree that it does seem to be feasible.

In the terms of, say, DSLstats it is the delta FEC count per minute.
Logged
:cat:  100% Linux and, previously, Unix. Co-founder of the ELRepo Project.

Please consider making a donation to support the running of this site.

les-70

  • Kitizen
  • ****
  • Posts: 1255
Re: ZyXel VMG8324-B10A Preliminary findings.
« Reply #223 on: February 12, 2015, 04:12:48 PM »

no for DLM to remove interleaving the ES has to be below 10% of the normal limit, so e.g. 280.

Also if they on the standard profile its just 140 allowed.

   We are going off topic here but MDWS shows many people who have kept below 40ES/day and are still interleaved after at least 60 days below 40ES/day.  From MDWS it does not look like the 140 or 280 figures can be correct or simply apply. The number of people interleaved does not get really small until you get below 30ES/day.  As I said earlier some lines may be subject to a history which cautions the DLM but who knows.

In this day and age it is strange how you can buy a service with secret characteristics.  ???
Logged

Chrysalis

  • Content Team
  • Addicted Kitizen
  • *
  • Posts: 6684
Re: ZyXel VMG8324-B10A Preliminary findings.
« Reply #224 on: February 12, 2015, 05:33:08 PM »

Yeah I am not disputing that bit we dont know if the line he is on about is also below the 10% rate as well.
Logged
AAISP - Billion 8800NL bridge & PFSense BOX running PFSense 2.4 - ECI Cab - LINE STATISTICS CLICK HERE
Pages: 1 ... 13 14 [15] 16
 

anything