cancel
Showing results for 
Search instead for 
Did you mean: 

Recently upgraded to 1gb

JamboGee92
Joining in

Hi, I'm a gamer so latency is very important to me, it was too high on 100mb so I recently upgraded to 1gb in the hope my latency would decrease, the maximum latency has decreased but the minimum is the same, and is double of most BQM's I've seen. 

Here is my BQM: 

43d5f5b8b3f49eb1b5b67b2b5661fd4103b137ee-14-04-2022.png

I'm hoping to reduce it, the green bar seems way too high?

11 REPLIES 11

Andrew-G
Alessandro Volta

The red bars suggest loss of connection or hub restarts, and the 15 minute outage at around 11pm was a regional or national loss of core network connectivity, and shows on my BQM (well done VM, keeping up the dismal reliability).  

Putting that aside your BQM looks good for a VM DOCSIS 3.1 connection.  At the best of times DOCSIS always has slightly worse latency than a good Openreach connection, although so long as the yellow fringing remains minimal it shouldn't affect your gaming.  Your reaction to to an on-screen event is around 250ms, factor in the latency from server to your screen, your screen refresh interval, and then your input back to the server, and the circa 10ms minimum latency penalty for DOCSIS is neither here nor there.

Obviously if you see worse ping spiking (to which DOCSIS is prone under non-ideal conditions) then you might conclude that you've made a bad move, but on the BQM posted I'd not worry.  In terms of the worse baseline latency you might observe compared to your 100 Mbps connection, that seems to be a feature comparing a Hub 2 or 3 to a Hub 4 or 5.  I guess that either VM are using a different CMTS architecture for Hub 4 and 5 that adds a few milliseconds, or that it's a simple artefact of having to process the greater number of channels these newer hubs use.

There's nothing you can do about this - it is inherent in the DOCSIS technology that VM use.

EDIT: Didn't read the post properly!

 


@Andrew-G wrote:

The red bars suggest loss of connection or hub restarts, and the 15 minute outage at around 11pm was a regional or national loss of core network connectivity, and shows on my BQM (well done VM, keeping up the dismal reliability).  

 


To add to that, the routing has changed, I'm close to london and all my traffic is now going via manchester. 

Thanks. It was just the fact my online teammate has Virgin 1gb and his minimum latency is down at 10ms, which is half of mine. I was hoping for it to be around the that, but maybe it's unnecessary anyway. He is based in London though, I'm based in Edinburgh, so probably pointless comparing. 


@JamboGee92 wrote:

Thanks. It was just the fact my online teammate has Virgin 1gb and his minimum latency is down at 10ms, which is half of mine. I was hoping for it to be around the that, but maybe it's unnecessary anyway. He is based in London though, I'm based in Edinburgh, so probably pointless comparing. 


10ms sounds like they have FTTP, whereas you probably have FTTC, so copper coax from the local cabinet to your home.

20 - 30ms is about right for FTTC.

VM provide both types depending on the area you are in (They are upgrading to FTTP everywhere over the next 4 (Ish) years).


@carl_pearce wrote:

@JamboGee92 wrote:

Thanks. It was just the fact my online teammate has Virgin 1gb and his minimum latency is down at 10ms, which is half of mine. I was hoping for it to be around the that, but maybe it's unnecessary anyway. He is based in London though, I'm based in Edinburgh, so probably pointless comparing. 


10ms sounds like they have FTTP, whereas you probably have FTTC, so copper coax from the local cabinet to your home.

20 - 30ms is about right for FTTC.

VM provide both types depending on the area you are in (They are upgrading to FTTP everywhere over the next 4 (Ish) years).


He told me he has FTTC, so maybe it could be that my local exchange is further away?


@JamboGee92 wrote:

@carl_pearce wrote:

@JamboGee92 wrote:

Thanks. It was just the fact my online teammate has Virgin 1gb and his minimum latency is down at 10ms, which is half of mine. I was hoping for it to be around the that, but maybe it's unnecessary anyway. He is based in London though, I'm based in Edinburgh, so probably pointless comparing. 


10ms sounds like they have FTTP, whereas you probably have FTTC, so copper coax from the local cabinet to your home.

20 - 30ms is about right for FTTC.

VM provide both types depending on the area you are in (They are upgrading to FTTP everywhere over the next 4 (Ish) years).


He told me he has FTTC, so maybe it could be that my local exchange is further away?


I've see 10's of BQMs on here and most are around 20ms average, unless they are FTTP.

Maybe there is an exception to that rule.


@carl_pearce wrote:


10ms sounds like they have FTTP, whereas you probably have FTTC, so copper coax from the local cabinet to your home.

20 - 30ms is about right for FTTC.

VM provide both types depending on the area you are in (They are upgrading to FTTP everywhere over the next 4 (Ish) years).


VM do not have any FTTC (Fibre to the cab) they have a hybrid coax network (FTTC or VDSL 2 is a different technology mostly used by Openreach, (copper twisted pairs), hence why it is slow at a distance, as the signal drops off). HFC (hybrid fibre coax) isn't restricted to these limitations (although it has its own, one of which is high latency and higher power consumption (topical at the moment))

FTTC with a provider lets say BT, can obtain pings of 8MS and below, I know this since I manage a FTTC line which has a ping of 9 during peek and 8MS offpeak. FTTC has considerably lower pings than HFC however comes at a cost (slower if far from the cab) all the FTTC lines i have ever used have been under 30M from the CAB so I have always had decent performance, even on copper lines that are 40 years old.

FTTP is capable of 1-3MS on speedtest.net if your in the right location, I know the lowest I have had in the past over FTTP with Orange was 1MS and the highest I ever saw was a massive 3MS.

From the same location cable was 26MS ping.

 

 


@JamboGee92 wrote:


He told me he has FTTC, so maybe it could be that my local exchange is further away?


FTTC (VDSL 2) mostly used by openreach in the UK does not require an exchange at all, they are routed via an aggregation node to a parent FTTP or fibre headend. This could be a datacenter. 

Openreach has two types of exchanges a parent exchange and a child, parent exchanges are far superior in technologies, capability, network capacity. Child exchanges are being phased out as they are no longer needed with FTTP as fibre is pulled (or blown mostly) from the parent or headend.