Menu
Reply
  • 82
  • 0
  • 13
CJS23
Up to speed
922 Views
Message 3701 of 4,479
Flag for a moderator

Re: Hub 3 / Compal CH7465-LG (TG2492LG) and CGNV4 Latency Cause


@Boygroucho wrote:

"I was a big complainer but to be honest the latest trial firmware looks pretty rosy compared to before!"

 

Its interesting that as expectations lower, how much easier it is to celebrate mediocrity.

Nowtv.png


To be fair, the quote at the top does not imply to me that mediocrity has been accepted, just that the latest Firmware has empirically and evidentially resolved the majority of the issues they were experiencing. (Apologies, however, if I've misunderstood what you were actually saying, btw)

That's all I'm after too, to be honest. I just want the same performance from the "updated" Hub 3.0 as I have always had from my trusty SH2AC.

0 Kudos
  • 47
  • 1
  • 8
Germ
Dialled in
851 Views
Message 3702 of 4,479
Flag for a moderator

Re: Hub 3 / Compal CH7465-LG (TG2492LG) and CGNV4 Latency Cause

So 9.1.116V is the new firmware?
0 Kudos
  • 883
  • 123
  • 389
Andruser
Well-informed
841 Views
Message 3703 of 4,479
Flag for a moderator

Re: Hub 3 / Compal CH7465-LG (TG2492LG) and CGNV4 Latency Cause

No, 9.1.116V is the previous "general issue" firmware with all its faults. 

Because Virgin Media management (to judge by what we see) live in abject fear of communicating with customers, no customer actually knows as a matter of proven fact* what the latest general issue firmware is. Most of us believe that the latest firmware is 9.1.116.603 that appears to be being deployed to users on 350 Mbps connections, but absent any official word it is possible that's an extended trial.  

* When I say "proven fact", I don't mean the mutterings of an installation technician, a statement by the APPALLING offshore call centre, or even promises from UK second line support or retentions, I mean a proper official statement about what is the latest firmware, what the company they believe they've fixed, whether there is any further firmware development going on for the other broken bits of the Hub 3 software, and what the roll out plan is.  This being Vermin Media, I'm expecting that statement on the 12th of Never.

  • 47
  • 1
  • 8
Germ
Dialled in
822 Views
Message 3704 of 4,479
Flag for a moderator

Re: Hub 3 / Compal CH7465-LG (TG2492LG) and CGNV4 Latency Cause

When gaming everything was relatively smooth for a day and a half but went back to normal that's why I was asking. I'm starting to think that there's just to many people on VM where I live that's making it worse. It still wasn't close to perfect but it was a major difference from what i have experienced. 

0 Kudos
  • 82
  • 0
  • 13
CJS23
Up to speed
818 Views
Message 3705 of 4,479
Flag for a moderator

Re: Hub 3 / Compal CH7465-LG (TG2492LG) and CGNV4 Latency Cause

Aside from those on 350mbps and the Triallists, does anyone on 200mbps have the .603 FW does anyone know?

0 Kudos
Highlighted
  • 102
  • 2
  • 91
Datalink
Up to speed
812 Views
Message 3706 of 4,479
Flag for a moderator

Re: Hub 3 / Compal CH7465-LG (TG2492LG) and CGNV4 Latency Cause

@CJS23's comments are rather interesting. As far as I'm aware, no one, and no company, including Intel has publically proven that the latency issues have been resolved for all protocols, ie: ICMP, TCP/IP and UDP, and the various offshoots of those protocols. If thats the case, I'd love to see the test results that prove that this issue has finally been resolved, some 6 years after Intel initially developed the Puma 6 chipset.


"4. Intel argues that “[i]f Intel’s internal improvement efforts
were to become public, third-party competitors would be able to use Intel’s own internal analysis,
which they ordinarily would not be able to access, to disparage Intel’s Puma 6 chipset and spread
fear, uncertainty, and doubt in the marketplace relating to the Puma 6 chipset and latency issues
that Intel identified and already resolved.”

Show me the proof !  Its pretty easy to disparage a product that has shown terrible performance since day one, when the design company itself hasn't shown any test data that refutes the observed problems.  Intel could easy put this behind them by releasing independent verified test results.  Until that happens, there will always be doubt about the Puma 6 performance, and that doubt will carry on to Puma 7 modems, which is unfortunate.

0 Kudos
  • 279
  • 19
  • 94
philjohn
Fibre optic
790 Views
Message 3707 of 4,479
Flag for a moderator

Re: Hub 3 / Compal CH7465-LG (TG2492LG) and CGNV4 Latency Cause

Yeah, when you're in the middle of a class action lawsuit and there is a lot of information out there there will be a good reason that legal is overriding PR - Judges tend to look down on trying to make an end-run around the process by releasing things to the press.

Anyway, there's empirical data that typical UDP flows or around 1Mbps are free from jitter spikes for extended periods of time, whereas before they weren't. Testing UDP at much more than that isn't particularly useful as you'll get more packet loss (and that's the same with any connection - UDP packets aren't guaranteed to be received, much less in order - they are fire and forget).

I doubt any game is using more than 1Mbps anyway, and VOIP uses considerably less than that.

  • 82
  • 0
  • 13
CJS23
Up to speed
781 Views
Message 3708 of 4,479
Flag for a moderator

Re: Hub 3 / Compal CH7465-LG (TG2492LG) and CGNV4 Latency Cause


@philjohn wrote:

Yeah, when you're in the middle of a class action lawsuit and there is a lot of information out there there will be a good reason that legal is overriding PR - Judges tend to look down on trying to make an end-run around the process by releasing things to the press.

Anyway, there's empirical data that typical UDP flows or around 1Mbps are free from jitter spikes for extended periods of time, whereas before they weren't. Testing UDP at much more than that isn't particularly useful as you'll get more packet loss (and that's the same with any connection - UDP packets aren't guaranteed to be received, much less in order - they are fire and forget).

I doubt any game is using more than 1Mbps anyway, and VOIP uses considerably less than that.


That pretty much sums up my view also. Anyone any ideas on my firmware question? I'm having a nightmare after just 2 days and am SO tempted to go back to the SH2AC which I retained with VM's permission. This means I would be turning down the possibility of 350mbps though as they will flat out refuse to provision on the SH2AC Smiley Frustrated

From my public FB Post to VM here 

 [MOD EDIT: Personal and private information has been removed from this post. Please do not post personal or private information in your public posts. Please review the Forum Guidelines]

2 x AC.PNG

  • 279
  • 19
  • 94
philjohn
Fibre optic
748 Views
Message 3709 of 4,479
Flag for a moderator

Re: Hub 3 / Compal CH7465-LG (TG2492LG) and CGNV4 Latency Cause

You have three options, go back to the SH2, keep the SH3 and Vivid 200 or upgrade to Vivid 350 for an extra £5 a month and be in with a good chance of getting the new firmware.

Remember, you can always downgrade, with 30 days notice, back to Vivid 200 and start using your SH2 again if you don't get the updated firmware, or if the update doesn't fix your problems.

  • 883
  • 123
  • 389
Andruser
Well-informed
722 Views
Message 3710 of 4,479
Flag for a moderator

Re: Hub 3 / Compal CH7465-LG (TG2492LG) and CGNV4 Latency Cause

CJS23: "This means I would be turning down the possibility of 350mbps though as they will flat out refuse to provision on the SH2AC "

If they re-enable the SH2ac, chances are they'd do that remotely and you'd still have the Hub 3.  If at some future date you wanted and were able to order 350, by that time you should have a definitive view on performance reported by other users with the 603 firmware, VM would be able to reactivate the Hub 3 that you'd have at the back of a cupboard, so I don't think you're turning down the possibility of anything.