RL317 Posted December 15, 2017 Share Posted December 15, 2017 I was recently looking to replace my BT Smart Hub as, despite using DMZ for my R1, I was double NATting and I recently read about that causing problems with packet loss and the like for the upstream router or something. I heard a lot of positive feedback on the HG612 3B for BT Infinity - not only here but as the general consensus on many forums across the internet - so I went for one of those on eBay, but the only listing I could find for a brand new HG612 disappeared and then reappeared at a much higher price so I went for my back up option instead. The W9970 is a modem router combo but acts primarily as a VDSL modem with its bridge mode setting, and it was £10 cheaper than the HG612 brand new, so I ordered that instead. Reviews online suggest it's arguably the next best thing to the HG612 in the VDSL world and performs identically much of the time. However I wasn't sure what to expect as I'd only ever used my ISP modem/router combos before, and I didn't think there'd be a positive performance change if at all. I plugged it in, got it set up as a modem only and connected it back to my R1, set up a thinkbroadband monitor just to see if changing routers would give me a base latency increase or add interleaving etc., then continued game downloads on my PS4 as my hard drive had failed on me for the umpteenth time. When the download was finished at around 11pm I changed back from 100% reactive to 70% preemptive and continued using the internet in order to watch a film on Amazon until I went to bed around 1:30am. When I got up the next morning I checked the monitor and I was a little confused over what I saw... I'm going to post two screenshots. Take a look at the first one and you'll see what a typical day using the Smart Hub with the R1 on 70/70 preemptive looked like. Notice the constant presence of small but otherwise unwanted yellow spikes which indicate maximum latency; this would occur regardless of whether or not I was using the internet. The second screenshot shows what happens after I got set up with the new modem, and what confused me is how there's a near total absence of yellow on the graph after I reduced my sliders following the game download. What on earth is going on here? LOL how could using some random third party modem improve stability? Are Smart Hubs just that bad? The weird part is those yellow spikes never really translated to poor performance because I'd never see anything matching those yellow spikes (see third screenshot as an example of how my line performed even with the Smart Hub) so I'm honestly a little lost. The extra weird part is my setup is a mess downstairs right now thanks to the TP Link coming with a short power cable - I couldn't use my usual power socket, so I had to resort to the crappy long ribbon DSL cable that came with my Smart Hub rather than my shielded twisted pair 0.5m cat6 VDSL cable, and I've got a bunch of power cables running alongside the data cables which can't be good haha Anyone seen this moving from an ISP freebie modem combo to a standalone modem before? Any idea why? Bonus points if you can tell me whether or not I can use my Smart Hub power cable with the TP Link despite the fact its label explicitly states "for use with Smart Hub 6A" - the other stats on the labels look close enough but I want to know if the voltage has to be an exact match rather than a rough range with the ~ symbol because I'm paranoid it might cause issues or fry the modem lol Link to comment Share on other sites More sharing options...
BIG__DOG Posted December 15, 2017 Share Posted December 15, 2017 from 12am to 12pm that connection looks sweet and it is the same as how mine normally is.........i noticed that evening peak time your connection goes to dog poop, that could be a 1 off and look at how it is tonight and see if you get massive ping spikes again. If it is the same tonight then it is likely isp traffic shaping or heavy usage in your household i imagine but not 100% sure! Link to comment Share on other sites More sharing options...
kinel Posted December 15, 2017 Share Posted December 15, 2017 Have you tried cod ww2 with your new modem setup Link to comment Share on other sites More sharing options...
N3CR0 Posted December 15, 2017 Share Posted December 15, 2017 I'm sure you saw my graphs before IIIRL of with BT HH and with Draytek. Its pretty much the same as what you have now. Trust me, in IW at least, it makes ALOT of difference Link to comment Share on other sites More sharing options...
RL317 Posted December 15, 2017 Author Share Posted December 15, 2017 from 12am to 12pm that connection looks sweet and it is the same as how mine normally is.........i noticed that evening peak time your connection goes to dog poop, that could be a 1 off and look at how it is tonight and see if you get massive ping spikes again. If it is the same tonight then it is likely isp traffic shaping or heavy usage in your household i imagine but not 100% sure! Oh that wasn't normal. I was maxing out the line by downloading several games and updates for my PS4 while watching a film on Amazon at the same time LOL. I had it on 100% reactive for that reason but my last download was winding down around half 10, so I dropped it to 70% and the crazy spikes dropped immediately Link to comment Share on other sites More sharing options...
RL317 Posted December 15, 2017 Author Share Posted December 15, 2017 Have you tried cod ww2 with your new modem setup Nah not yet mate. Not sure if I can bring myself to reinstall it because it was so depressingly laggy and broken. I've never NOT wanted to play a game like I have with WW2 lol Link to comment Share on other sites More sharing options...
RL317 Posted December 15, 2017 Author Share Posted December 15, 2017 I'm sure you saw my graphs before IIIRL of with BT HH and with Draytek. Its pretty much the same as what you have now. Trust me, in IW at least, it makes ALOT of difference Yeah I think I did mate. At least I think I remember you posting a BQM with a totally flat green band. Mine was never like that with the Smart Hub, but the reason I'm confused is I never saw those maximum latency (yellow) spikes on network diagnostics tests, PingPlotter or the geofilter ping graphs. It was always flat with maybe 0.2ms jitter max lol. Can't complain with the absence of yellow now though Link to comment Share on other sites More sharing options...
N3CR0 Posted December 15, 2017 Share Posted December 15, 2017 Yeah I think I did mate. At least I think I remember you posting a BQM with a totally flat green band. Mine was never like that with the Smart Hub, but the reason I'm confused is I never saw those maximum latency (yellow) spikes on network diagnostics tests, PingPlotter or the geofilter ping graphs. It was always flat with maybe 0.2ms jitter max lol. Can't complain with the absence of yellow now though Yeah man. You'll have to fiddle with ya Duma settings again though most likely. What worked for me with the BT HH, didn't work very well at all with the Draytek. Link to comment Share on other sites More sharing options...
RL317 Posted December 15, 2017 Author Share Posted December 15, 2017 Yeah man. You'll have to fiddle with ya Duma settings again though most likely. What worked for me with the BT HH, didn't work very well at all with the Draytek. I find nothing really works for CoD now unless some troll face wizard up there decides to flick the hit detection machine on for a game or two before he decides to troll me for a month solid again LOL Looking pretty sweet this evening even though I was using the internet again. On the Smart Hub I'd see those constant yellow bars even if I wasn't using it and had sliders reduced lol. I just really want to know why that didn't actually seem to affect my connection performance because I never saw spikes up to 25ms or whatever on tests. Link to comment Share on other sites More sharing options...
N3CR0 Posted December 15, 2017 Share Posted December 15, 2017 Screw WW2. Try IW again. It was almost night and day for me once I found the right settings. Link to comment Share on other sites More sharing options...
RL317 Posted December 16, 2017 Author Share Posted December 16, 2017 Screw WW2. Try IW again. It was almost night and day for me once I found the right settings. Just got off IW. It was more of the same... being "connectioned" Link to comment Share on other sites More sharing options...
ColonicBoom Posted December 16, 2017 Share Posted December 16, 2017 I saw a similar change when I switched from TalkTalk's provided all-in-one hub to a standalone modem. The TT hub caused constant little spikes that I thought were symptoms of a bad line, it would reset itself every couple of days too which meant that I was permanently interleaved doubling (at least) my ping. Within 12 hours of putting a standalone modem in, the spikes went away and the ping dropped back to fast-path speeds. Link to comment Share on other sites More sharing options...
RL317 Posted December 16, 2017 Author Share Posted December 16, 2017 I saw a similar change when I switched from TalkTalk's provided all-in-one hub to a standalone modem. The TT hub caused constant little spikes that I thought were symptoms of a bad line, it would reset itself every couple of days too which meant that I was permanently interleaved doubling (at least) my ping. Within 12 hours of putting a standalone modem in, the spikes went away and the ping dropped back to fast-path speeds. I've always been on fast path (I made a habit of checking my modem stats whenever I rebooted my modem/swapped cables/had a power cut lol) so my base latency has always been around the same, and my average always hovered within a few tenths of a millisecond around that, so I just can't work out the constant yellow up there on the first screenshot Link to comment Share on other sites More sharing options...
ColonicBoom Posted December 16, 2017 Share Posted December 16, 2017 I've always been on fast path (I made a habit of checking my modem stats whenever I rebooted my modem/swapped cables/had a power cut lol) so my base latency has always been around the same, and my average always hovered within a few tenths of a millisecond around that, so I just can't work out the constant yellow up there on the first screenshot I think my fast path issue was down to errors on my line or total drops / resets every couple of days. I checked the up-time and it rarely got beyond a week. It seemed to be resetting overnight, every few days. Because the yellow on BQM represents only the maximum ping within that 100 seconds, rather than an average, it means there was at least one spike (and maybe only one) reaching whatever level the yellow reaches. For 99 seconds, of that 100 seconds, it could be completely flat but if there's a single spike of 50ms then the yellow will show 50ms for that point on the graph. Averages can be misleading anyway, if you think about an average over 100 seconds... If your ping is only 1ms for each of 99 seconds, you have an average of 1ms. But if on the 100th second your ping hits a big 101ms spike, you will still only have an average of 2ms over the 100 seconds. It will be a nice average but it will contain a horrible spike. That it's now yellow-free on BQM, since dropping the hub, strongly suggests that your hub had some kind of regular processing hiccup that was causing you to see the yellow peaks before. It looks mint! Link to comment Share on other sites More sharing options...
RL317 Posted December 16, 2017 Author Share Posted December 16, 2017 I think my fast path issue was down to errors on my line or total drops / resets every couple of days. I checked the up-time and it rarely got beyond a week. It seemed to be resetting overnight, every few days. Because the yellow on BQM represents only the maximum ping within that 100 seconds, rather than an average, it means there was at least one spike (and maybe only one) reaching whatever level the yellow reaches. For 99 seconds, of that 100 seconds, it could be completely flat but if there's a single spike of 50ms then the yellow will show 50ms for that point on the graph. Averages can be misleading anyway, if you think about an average over 100 seconds... If your ping is only 1ms for each of 99 seconds, you have an average of 1ms. But if on the 100th second your ping hits a big 101ms spike, you will still only have an average of 2ms over the 100 seconds. It will be a nice average but it will contain a horrible spike. That it's now yellow-free on BQM, since dropping the hub, strongly suggests that your hub had some kind of regular processing hiccup that was causing you to see the yellow peaks before. It looks mint! That's the weird thing. If I ran a PingPlotter or did the test to 8.8.8.8, even repeatedly for an hour, I could see not one single spike past 10ms. I don't recall the diagnostic test EVER going above 10ms. I could be downloading something, reduce sliders only as far as 97% and not even see a 1ms increase in ping times lol BTW IW seems to run even worse for me now, with the standard 70% and ultimate profile setup. There is no way I'm trying WW2 again LMAO Link to comment Share on other sites More sharing options...
N3CR0 Posted December 16, 2017 Share Posted December 16, 2017 Just got off IW. It was more of the same... being "connectioned" Probably need to find ya new Duma settings then. When I first plugged in the Draytek, had crap games for like 2 weeks, had to experiment, find the right settings. Link to comment Share on other sites More sharing options...
ColonicBoom Posted December 16, 2017 Share Posted December 16, 2017 That's the weird thing. If I ran a PingPlotter or did the test to 8.8.8.8, even repeatedly for an hour, I could see not one single spike past 10ms. I suppose it could be the BQM server at the other end... Link to comment Share on other sites More sharing options...
N3CR0 Posted December 17, 2017 Share Posted December 17, 2017 I suppose it could be the BQM server at the other end... With the BT HH, my BQM looked exactly the same as his does using the BT HH. Just goes to show how rubbish the HH is I guess. Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.