Redflame

I see. Maybe my test was glitching out I’ll try the ping. my line as previously mentioned is 900 down and 115 up. Duma OS router speed test sometimes will give me 888 or 907. Upload is 107-105. So I gone in and manually added 900 and 115 Hopefully this will help lock in results. I know 80 percent is a good value for download so I’ll do some more bb testing and focus on upload turning.

Thanks for response. So I did that. I found 75 added no bloat, going up to 80 then offer same result. So 85 and up seems to be bloat city. So for download I stuck on 80. If I had 85 down and 95 up my download values and jitter was below 1ms. If it lower my upload lower, my download ping values increase and jitter could hit 6-15ms. So it seems strongly shaped can add more impact to download. Do you think this is just noise or can capping upload on fttp cause issues?

Thanks. I’ve manage to get back onto my R3. My mother still had Fritzbox in storage so it’s my edge device now. For testing CC - is it best to go onto waveform bufferbloat test move down from 95 and down 10 percent and then up by 5 percent? My ping optimiser is telling me my ping 99-100ms is stable even though my London ping is usually 8-12, this sounds like it may be picking up wrong info? If I do test ping option instead of optimise it does show 8-9ms.

Hello, I ran an R3 from late 2025 to May 2026 on a 900/120 Openreach UK PPPoE line and got excellent results (A+ bufferbloat, only +0-2 ms added latency). Before that I was on a C grade with constant lag. A disabled family member who had two small strokes moved downstairs into a hospital bed for safety. Previous renovations that included knocking down walls and adding ceiling beams etc, caused WiFi coverage became poor on the far side of the house. The R3 was the only router that could reach part of the room with a weak but workable signal. We tried a TP-Link WiFi extender, but it didn’t work well with their streaming setup. So we move towards eero mesh. Because I didn’t have safe desk space to run the full ISP router → R3 → Eero chain, I replaced both the ISP router and the R3 with a full Eero mesh (Eero gateway + Eero 6 nodes). This solved the coverage problem, but has re-introduced noticeable bufferbloat and latency spikes during gaming, even with eero’s SQM enabled. My family member is getting their own extension with dedicated FTTP line soon, so this setup is temporary. I’m planning to go back to the R3 for its much better latency performance. Since I no longer have an ISP router, I need a replacement PPPoE router to sit in front of the R3 without bottlenecking it. Any recommendations or advice? (previous router was fritzbox). Thanks

Hi Fraser, I hope you and your team are having a good week so far. I manage to set it up successfully and I’ve kept my open Nat type. I found 80 down and 90 up was sweet soft, giving a+ rating with zero up or down added lat on bufferbloat test (Fritzbox was a B). On my old isp I had to go down to 60 percent to get rid of bufferbloat. When boosting ps5 device.should I set it to level 2 or level 5?

Hi So it arrived on time. So didn’t have to do pppoe. I decided not to use R3 for at least a week just so I could have a baseline of my ISP without and with R3. So tomorrow night - Saturday - I will doing the R3 to fritzbox just to ensure I do the most performant way. Plug R3 into a lan port on Fritz > go to Fritz give it static ip > select expose host > enjoy gaming. Is that the right flow? I hope you and your team have a great weekend.

Thanks for clearing it up for me. Static IP is part of the package, but thanks for clarifying what you mean about wan. My service is going live in 4 days time and router is coming in 4 days, so I may try to plug r3 in if it doesn’t arrive when I go live. As a temp solution, if it works it works if it doesn’t, it doesn’t. Hopefully it will be less than 24 hours just in case. I am not sure if you recall, but I did struggle with whole r3 in DMZ, just about manage to do it, haha. I’m a technical noob, so thankful for all your support. I am getting a Fritzbox from my isp. I understand you have to do an ‘expose host’ for the R3 to use it to do pppoe. Are you aware if setting this up is hard or straightforward?

Hi Fraser Thanks for information. So from what I read, you have to go back to older firmware and do a certain approach to get it to not disconnect, but you also noted you may not have an issue if you have a static ip? I have order a static one with my isp due to having nat issues with my current isp.

I am moving from my current ISP to a newer ISP. There may be a chance that my service could go live without the new router arriving yet therefore I may need to plug my R3 directly into ONT until I get new router for DMZ. I was told not so long back, that there was a disconnect bug with direct PPPoe.

Hi Fraser Before getting back you I wanted to check lots of stuff. I can’t do the windows/pc test. We have a laptop but bizarrely my mom’s laptop doesn’t have an Ethernet port. R3 and EE hub = the R3 wan ip address is same as DMZ. So I did port forwarding for cod. And the game still reported moderate Nat type with port forwarding. My Ps5 connection test says I have the type 2 Nat type. I asked EE myself for clarification on this and I know you can get conflicting info from isp based on agent, but the agent was adamant they use CG-Nat on all products fibre and mobile (some customers on forums argue they don’t use it for fibre). Although when I was with BT before my migration opening up a port to cod services give me open Nat. So I’m guessing it’s highly plausible that I’m sitting behind CG-Nat. My contract is over in May so I’m going to PlusNet (as BT backbone does give me a good below 1ms jitter connection to most game servers if my local network is kept in check with QOS) Or I’m going to try Sky but option 61 hurdle worries me lol. Thanks for all your support, I am very grateful.

Hi So I double checked. Wan IP > 192.xx.xx.xx is the same as the static ip that is set on ee hub - the one I used since day one. public ip > is the same one that call of duty, whatismyipaddress dot com and ip info see. I don’t know if this is a sign of double Nat or just the way things work in DMZ. But I noticed in trace routes that after dumaos.lan I hit ee router default gateway. I’ll show a screenshot,

Hi Fraser I hope you and your team had a good Christmas and 2026 is treating you all well so far. I was wondering if i could ask your advice. Due to my setup (Using EE hub with the R3 inside its DMZ). I have double Nat. My ps5 doesn’t seem to care, but my Xbox is not happy about it. So if I were to plug my R3 into my EE’s FTTP ONT, would I need to setup R3 again or can I just go to wan settings on Duma os app and put sign in details? Also, if this approach doesn’t go well and go back to ee DMZ. Should I screenshot my current wan settings or will the router automatically revert to previous settings once reconnect to EE hub? Just to reinform you of my current setup. EE Hub WiFi is off, UPnP is off on ee hub and R3 is assigned a static ip and in the EE DMZ. R3 has UPnp on.

Thanks for information. Yes I found the sweet spot now on cc now. My pings are usually rather low but there’s a lot jitter on my path so that can cause negative feedback from game especially in PvP. So Steady Ping helps me. It makes hit markers and incoming landing with good frame by frame accuracy rather than massive bursts traffic like the famous one hit/10 bullets all in one. I will prioritise it as it’s very I download during gameplay. It was on off. I feel having backup it’s better than not BTW I can fully BF6 without having filter turned off in geo filter. Plus all servers I get seems to be on their rightful location e.g London or Frankfurt. So thanks to you and your team for your hard working locking that down

Server ID - IKUGH4 Geo filter map location - Oklahoma City, USA Real location - EU I was playing BO7 co op end game mode and this popped up in USA. But ping was nowhere near USA, so I strongly believe it was misplaced. My ping to server from uk was 30-35ms. Usually those are pings I get to Cod Madrid, Milan and Warsaw servers.

Thanks very much, Fraser. It wasn’t as hard as I thought it would, yay. I’m familiar with interface a bit due to EE WiFi Enhancer, But I’ve been trying to get most out of my R3 so a few things: 1. BufferBloat and Ping Optimiser So I ran Ping Optimiser and it kept reporting n/a on new ping. I thought it might be glitching out as I know there had been issues with it before. But I don’t think it couldn’t do any meaningful change to my connection, not a glitch. When I first step up R3 it gave me an 95 cap up and down. So I decided to do a waveform bufferbloat test to ensure R3 is doing its job. We have a laptop but no Ethernet port, ps5 browser is inaccessible you can get around it to open but it doesn’t function properly. So I got my series x and plug it into Ethernet to R3 to do the test. 95% cap down/up on internet speed of 915/111 Grade A+ Idle Latency 9ms Download Latency 5+ms > total 14ms Upload Latency 0ms > total 9ms Jitter - Idle 0.5ms, download 2.4ms, upload 0.7ms I tried to get rid of that 5ms by lowing down to 90.5 percent of 915, and 87.5 percent of 111 Test 1: Grade A+ Idle Latency 9ms download latency 4ms > total 13ms upload latency 0ms > total 9ms Jitter - Idle 0.8ms, download 4ms, upload 3.5ms Test 2: Grade A+ idle Latency 9ms Download Latency +3ms > Total 12ms Upload latency 0ms > 9ms Jitter - Idle 0.9ms, download 2.5ms, upload 0.7ms I think it’s great R3 is able to completely kill off upload latency, before it I was getting 22-45ms when uploading. As a ps5 user I use the auto upload feature on PSN when I get a good sniper shot. So this doesn’t affect gameplay anymore! 2. Steady Ping I like the concept of steady ping. It does seem to help in call of duty multiplayer a lot. I understand it adds a buffer and it’s slightly delaying communication. Is that both upstream my input to the server, or the downstream snapshots to my ps5 or both? Also what does the tick rate synchronisation o 1.2 actually mean itself? In Battlefield it’s like 1.9-2.5 but in cod multiplayer it’s jumping all over the shop, 8 > 12 > 17. My Steady ping usage This is a mixed bag. If I use it when setting my CC to 95 or 70 it feels good, but if I use it with my current 90.5/87.5 it feels not good. Without steady ping using 90.5/87.5 it feels okayish, but I feel the slowdown/speedup/normalish pacing - I guess due to irregular packets arrival. At 95 without steady ping, again I can feel different pacing more intensely At 70 percent I don’t feel the irregularities. So I would say it works well at 95 or 70 I’m debating what should I hit. I have a streaming household, usually either 3 tv streams or 2 and one game - so at my current speed I’ll be dropping down to 632mpbs download and 77.7 upload, which is still plenty I feel? But 95, which ping optimiser set, seems to work well. Smart boost When it comes to smart boost, should I prioritise my ps5 on device list or is gaming as no1 priority enough? I have noticed when playing cod if another game is updating from patch/dlc ping can spike with ps5 as a favourite. Double Nat Nothing on PS5 has screamed double Nat. Like all gameplay is superior to before. I can play gta online, which uses peer to peer just fine and play server hosted games like cod/bf fine no issues. Series x though is complaining about double Nat. I’m not going to use much as family/friends mainly play on PS5 and PS games now not Xbox anymore, so I use my ps5 more for that reason. Is it normal with my setup with ee and DMZ to have double Nat or did I do something wrong?