Petelunn wrote:
"We are all angry but Virgin and Intel really don't seem to care
Upgraded to Vivid 200 Gamer as my son was doing well in an overwatch team but since upgrading he has had to stop playing with them as the latency is soooo bad. I've also had to stop raiding with my Warcraft guild as I've gone from one of the top DPS in the guild to one of the lowest. People with lower gear then my character do better dps.
I have four sons at home all who game and so do I, they also stream a lot of content so the 200mbps is great but I'm seriously thinking of leaving if I see a price increase. Only been on Gamer and therefore changed to the Hub 3 for a few months so unless they increase prices I can't leave and am stuck with this crappy hub.
Can get Sky Fibre at 76mbps in my area and its looking very attractive."
Like you I have a son who's gaming, like my own has been ruined by this fiasco.
I make fun of it but I'm actually really really ****** off.
So I know where you are coming from.
We've all been sold a lemon.,,,that includes Virgin and the rest.
By the way, I reboot this pile of **** every day to try and improve it's pathetic ******** performance. Apparently it gets worse, wonky after 22 hours of use. This may or may not be true but I do it anyway.
Anyway, November will be interesting.
Fwiw, the Puma 7 modem with or without Docsis 3.1 is greatly improved over the Puma 6 modems. The various protocol latencies that are observed in the Puma 6 modems are not present in the Puma 7 modem. That is for the Hitron CODA-4582 on the Rogers network in Canada. Having said that, the latest firmware version has introduced a modem to CMTS ICMP latency that wasn't present in previous firmware versions. This appear to be restricted to the modem to CMTS ICMP ping only, with no observable effect (at least not detected) on any other protocols or destinations beyond the CMTS. None the less, this requires attention from Rogers, Hitron and Intel to sort out. Running Docsis 3.1 downstream, there is an absolute improvement in downstream and upstream data rates. I didn't expect to see the upstream improvement but the data rates went from roughly 30 Mb/s during the mid day hours, 50 Mb/s at night, to 55 to 58 Mb/s throughout the day. UDP throughput performance (data rate) without losses has also improved to the point where the modem is probably better than many routers, so in terms of straight throughput consideration, running the modem might be a better idea. The exception will be for high end routers such as a PfSense, OpenSense, Sophos type of router. Users with routers will have to run their own tests to determine what is better on their ISP, considering the modem and router models ..... router or no router, that is the question. Routers will have much better user controls than you would ever see in a modem, but, that might not matter if you run a simple network and are interested in throughput rather than router functions.
On my list of things to do is to post some screen shots, Puma 6 versus Puma 7 on the DSLReports site, just haven't found the time lately to do that.
Here are a a few screen shots that show the Puma 6 IPV6 latency to the CMTS, taken in June 2016, followed by the Puma 7 switchover from Docsis 3.0 to 3.1 taken in March 2017. The Puma 7 shot still shows high evening latency running Docsis 3.0, which basically disappeared when 3.1 was enabled.
grrr, short timeframes to edit a post. That needs to be corrected.... hello mods ...
Ok, continuing with the task at hand.
First plot below: Puma 7 prior to switching to Docsis 3.1. On the left of the center latency spike, the day before my CMTS starts to switch to 3.1, so, still runnnig Docsis 3.0 with its high latency during the evening. On the right of the center latency spike, probably the day that some users switch over to Docsis 3.1. I wasn't one of them, but, there is still a remarkable drop in evening latency.
Second plot below: the day of the change. On the left of the central packet loss indication, Docsis 3.0, on the right of the packet loss spike, Docsis 3.1. Flat line except for one spike during the entire day.
Lets see a 1hr plot setup like this one. 1 hour width. TCP, 10hz at least, same scale settings.. It can be port 80 to google. Then we can compare a Puma 6 VS Puma 7
