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:

Author Topic: Modem testing question  (Read 1439 times)

cliver

  • Reg Member
  • ***
  • Posts: 169
Modem testing question
« on: September 23, 2016, 12:15:52 PM »

I'm looking to test my modems to see which gives me the best speed and stability on my line.
If I put a new/different modem online and it immediately syncs at a lower speed than its predecessor is that enough to rule it out straight away? Or should you give it time to settle (if so how long)?
I ask because I have a VMG8924 online syncing at 53462kbps down 8681kbps up and staying stable at that. I put an HG612 online just to factory reset it and it synced at 49658kbps Down 9224kbps up.
I left 30 mind between swaps and when I put the Zyxel back online it immediately synced at the higher speeds again.
Any thoughts appreciated.

Sent from my Nexus 6P using Tapatalk

Logged

Weaver

  • Senior Kitizen
  • ******
  • Posts: 11459
  • Retd s/w dev; A&A; 4x7km ADSL2 lines; Firebrick
Re: Modem testing question
« Reply #1 on: September 23, 2016, 02:00:40 PM »

You need to be very careful that DLM or changing interference and crosstalk levels are not mucking up the result. Suggestions at
    http://forum.kitz.co.uk/index.php/topic,15905.msg307641.html#msg307641
Logged

ejs

  • Kitizen
  • ****
  • Posts: 2078
Re: Modem testing question
« Reply #2 on: September 23, 2016, 02:01:54 PM »

I think it depends on which you are more concerned about, speed or stability. Everyone wants the best speed and stability, but the trouble is, higher speeds tend to be inherently less stable.

Your testing has already discovered the difference in speed, but I think the testing has not been long enough to say much about the stability of either. I wouldn't expect the speed to change much, unless the modem can't hold the connection, or the error rates are high enough to trigger the DLM to act.

For stability, I would monitor it for at least a day, maybe a few days or even a whole week, to cover day/night and weekday/weekend conditions.
Logged

cliver

  • Reg Member
  • ***
  • Posts: 169
Re: Modem testing question
« Reply #3 on: September 23, 2016, 03:25:26 PM »

Thanks guys, good link Weaver. I would happily trade a bit of speed for stability. To be fair my line is pretty stable as it is now and I should apply the "if it ain't broke don't fix it" rule.
Trouble is I can't help fiddling and trying things out 😉.
My only noticeable operational moan is that I sometimes get a glitch in my TV decoder where the picture freezes for a few seconds. The TV isn't streaming but it pops to an internet server for decoding, this is constant but only a few bytes. Funnily streaming TV services seem to work fine. So i can't be sure if it's my network or the servers at the other end causing the problem. The provider says all is on their end and no one else is reporting the same problem.
On the modern test my main thoughts were, that as I measured my existing modem, then changed it over to the HG 612 and it synced lower then put the old one back and it re synced at the higher rate straight away then the DLM hasn't changed my profile at all. So notwithstanding stability, which I would have to test over a longer period for my existing modem appears to be the fastest option, given that it's pretty stable anyway (I have been on the same profile for months and the only re syncs I get are when I instigate one for some reason).
I am changing from Talk Talk to BT shortly only because my contract is up and I'm on their 76Mbps service getting 50Mbps downloads. I'm trying BT's 54Mbps service as it's cheaper and I should in theory get close to maximum for my line. Therefore I'm not paying for bandwidth I can't get.
I think once I'm on BT I'll give each one a soak test and measure stability and take some stats from each over a week or so just to be sure I'm getting the best from my line.
Thanks again guys.
Logged

Weaver

  • Senior Kitizen
  • ******
  • Posts: 11459
  • Retd s/w dev; A&A; 4x7km ADSL2 lines; Firebrick
Re: Modem testing question
« Reply #4 on: September 23, 2016, 05:10:13 PM »

I've found that it's always worth turning things on at least twice, with a generous separation and down-time as suggested in that link, which will prove you from angering DLM, based on Kitz’ 15-min time bands.

Perhaps you could perhaps take the highest sync speed reported after any power-up?
Logged

cliver

  • Reg Member
  • ***
  • Posts: 169
Re: Modem testing question
« Reply #5 on: September 23, 2016, 05:14:44 PM »

I'll build that into my test schedule thanks Weaver.

Sent from my Nexus 6P using Tapatalk
Logged
 

anything