New Netduma R3 with low internet speeds, high ping and hard to get connection

[byTrunK]

Hi all,

This afternoon, I received my R3, but I'm encountering more issues than expected.

Firstly, after I connected for the first time and updated to the latest firmware, the router stopped getting a connection. I have a VLAN IPv4 connection, but I had to restart and perform several factory resets to regain the connection. If I turn on my ONT and the router at the same time, I cannot get a connection and have to restart it again.

Secondly, I'm experiencing low internet speeds. With my internet provider's router, I get 950Mbps for both download and upload speeds, but with this router, I'm losing 300Mbps.

With SmartBoost enabled:

With SmartBoost disabled (ridiculously low upload speed):

 

And these are the advanced ping settings I have:

Also, take a look at the high ping during both the download and upload processes... It doesn't make sense.

What am I doing wrong? This is disappointing compared to what I was expecting from this router, and I'm seriously considering returning it since it doesn't improve upon what I had before.

Hopefully, you can help me resolve this.

Thank you!

[byTrunK]

And more things that are not working...

EAFC24 ping increased from 16ms to 30ms to the same servers (in Madrid).

Geofilter is not working either... In strict mode, it's connecting to Brussels servers all the time instead of Madrid's ones

[byTrunK]

Updates:

I installed the 4.0.236 version, did a factory reset and exactly the same... 

I had to restart it 4 times to get connection, high ping (with and without steady ping), very low speeds and geofilter being ignored... 

At least, the router should be able to provide same speeds and ping than a normal router, with any config needed. But it doesn't... 

Another thing that annoyed me was that I had to change all my devices wifi passwords because the router doesn't accept spaces as part of the password... 

Faulty router? Help me, please! 

[Netduma Fraser]

I've given you Early Access so you can try that firmware and see if anything improves: https://forum.netduma.com/forum/151-netduma-r3-early-access/

At least from the Geo-Filter aspect I can tell you that your ping is higher due to Steady Ping being enabled, disabling this will lower it again. You're getting a game outside your radius due to Geo-Latency being enabled, if you disable that it will prevent it happening. 

[byTrunK]

I installed the v4.0.273 and it improves a little bit, but still far away from what's expected.

 

With SmartBoost enabled:

With SmartBoost disabled:

 

Ping spikes are amazing... I've tested several routers and I didn't see such as bad results as those...

For the higher ping, as I said earlier, it's happening with and without steady ping. Actually, I don't see any change between having it enabled or not.

 

With Steady Ping enabled:

With Steady Ping disabled:

 

And in-game I can confirm I'm getting exactly the same ping. And this ping is the double I can get with my ISP router which, believe me, it's close to be a potato.

 

This is what I get from that potato router:

 

I don't know what's hapenning. I want to think that this is faulty router or the it's not prepared for the connections we have in Spain (that it doesn't mean that they are good). But, unfortunatelly, I have no reasons to keep the R3.

[Fuzy]

Just to be sure: What exactly are you trying to do?

Reduce ping to game servers...? That's not possible...

Steady ping is a tool that creates a buffer to overcome the latency spike problem on the server you're playing on.

If you're trying to stabilize your jitter, the tool to use is ping optimizer!

[Fuzy]

You live in Spain ?

The Ookla ping test server (Madrid) does not necessarily reflect the ping to the game server (Madrid)

Your ISP seems to have a bad peering to this game's servers!

 

[byTrunK]

I’m trying to optimize the settings on the R3 to improve my gaming experience while maintaining a balance for other uses.

I understand that I cannot reduce the ping to game servers, but at the same time, I don’t want it to double either.

As I mentioned, steady ping isn’t making any difference. With or without it, I get the same results, which makes sense since my closest servers are stable.

I've run the ping optimizer about 10 times in the past 24 hours, and it seems like a placebo tool because I haven’t noticed any difference.

Just compare the results between the two routers: the ping is higher with less speed capacity. Do I really need to pay for double the ISP speed just to configure the router at 50% to get the results I want? I hope not...

 

Yes, I live in Spain, and I know the Ookla servers are not the same as the game’s servers. I'm using Ookla tests to the same servers in Madrid to show you that the difference they get there is the same difference I find in games and everything else: bad ping and low speeds.

You cannot say that my ISP seems to have bad peering to these game servers when I get different results between a stock router and the R3. The results with the stock router are actually better than the R3. My ISP is the best here in Spain, with the best peering.

Don't take my words the wrong way, I'm just feeling frustrated and disappointed to see results far below what I expected from a high-end router.

[Fuzy]

Welcome ! 😉

If you want to play exclusively on the Madrid server, disable geolatency!

There is one thing I don't understand is your ISP ping and the ping on the Madrid game server. If you don't have peering problems, you have set up a VPN upstream of the R3 maybe?

Testing from France, I have a better ping than you! (Heatmap ping)

[byTrunK]

No, I don't have any VPN configured.

That's one of the weird things the R3 is doing. I have 29ms to Madrid EAFC servers, with or without Steady Ping (and same in game). And with any other router I play between 14-16ms.

[Fuzy]

Ok! It's normal, steady ping adds 10ms of buffer (simple mode), but if you have a stable connection given your ISP bandwidth, you don't need it!

By the way, Steady ping has no influence on ping heatmap!

[byTrunK]

No, it's not normal! 😂 I said that I'm getting the same results with or without Steady Ping, so why is it adding more than 10ms to my ping?

It also seems like we're focusing on the ping, but the speeds are even worse... There's significant bufferbloat with 600Mbps (which is 350Mbps less than the maximum capacity where bufferbloat should start occurring...).

[Netduma Fraser]

Could you do a speed test from the actual website rather than the CLI instead please, what results do you get then? I'd suggest testing with downloads while following: https://bit.ly/bufferbloat-test start with a high value for Congestion Control e.g 95%, test, decrease 10%, test, decrease 10% etc, until you get to a value that is pretty good, then try 5% either side of that value to see if it can be improved. Important to note that Download/Upload on CC don't have to be the same value & you may have a better experience with differing values. This is more of a real world test as you're actually using traffic you would normally use.

[Fuzy]

Ok !  Glad it makes you laugh!😉

 

[byTrunK]

  

14 minutes ago, Netduma Fraser said:

Could you do a speed test from the actual website rather than the CLI instead please, what results do you get then? I'd suggest testing with downloads while following: https://bit.ly/bufferbloat-test start with a high value for Congestion Control e.g 95%, test, decrease 10%, test, decrease 10% etc, until you get to a value that is pretty good, then try 5% either side of that value to see if it can be improved. Important to note that Download/Upload on CC don't have to be the same value & you may have a better experience with differing values. This is more of a real world test as you're actually using traffic you would normally use.

I've just started running the test, and right from the beginning, it's not going well. I cannot get more than 600 Mbps, with or without SmartBoost, and no congestion control, which is the same result shown by Speedtest. This starting point doesn't encourage me to continue.

Decreasing the congestion control would imply admitting that the router is not capable of handling high speeds. It’s like paying my ISP for 1 Gbps speeds and using a non-gigabit network card.

I paid €200 for a router that should, at the very least, provide me with all the speed I'm paying for. From there, I should be able to adjust what I need, not what the router forces on me.

[Fuzy]

Really !

[byTrunK]

8 minutes ago, Fuzy said:

Really !

 

Exactly! That should be the starting point to see if the router is working properly. That's what I'm trying to determine before thinking about configurations and other stuffs.

 

Once again... Am I doing something wrong? Is the router faulty? Or is it simply not designed to work well with certain types of connections, like VLANs?

[Fuzy]

I'm in "VLan" so forget it!

I would love to help you, believe me!

 

[Fuzy]

Your R3 works very well, just take the time to tame it! 😉

[Netduma Fraser]

Reducing Congestion Control is specifically for the bufferbloat aspect, it will help with that. Could you provide me with an overview of your exact physical setup please including the model for the ONT?

[byTrunK]

2 minutes ago, Netduma Fraser said:

Reducing Congestion Control is specifically for the bufferbloat aspect, it will help with that. Could you provide me with an overview of your exact physical setup please including the model for the ONT?

Agree. But, again, bufferbloat should appear when reaching the maximum speeds, no 350Mbps less... What it's happening. 

I only have an Ufiber Loco ONT connected behind to the R3. And the R3 directly to my computer and PS5 via Ethernet. 

Same physical set-up used before with an Asus Ax86u pro with no speed/ping issues. 

[Netduma Fraser]

Is the ONT using a power adapter or purely PoE? If the latter if you try with the adapter do you notice any difference? 

Otherwise as a test do you have another router to handle VLAN instead of the R3 then see what the speeds are like?

[byTrunK]

4 minutes ago, Netduma Fraser said:

Is the ONT using a power adapter or purely PoE? If the latter if you try with the adapter do you notice any difference? 

Otherwise as a test do you have another router to handle VLAN instead of the R3 then see what the speeds are like?

It's using a power adapter. 

I shared a screenshot this afternoon with the speeds I'm getting with my ISP router (with an own integrated ONT) 

And pretty similar I was getting with the Asus Ax86u pro I mentioned earlier (but I don't have it since a couple of days ago) and using te same Ufiber Loco ONT. 

[Netduma Fraser]

I understand you're not having the issues with other routers but what speeds do you get through the R3 when trying the above scenario with the asus handling VLAN instead so ONT > Asus (VLAN) > R3 (DHCP)

[byTrunK]

33 minutes ago, Netduma Fraser said:

I understand you're not having the issues with other routers but what speeds do you get through the R3 when trying the above scenario with the asus handling VLAN instead so ONT > Asus (VLAN) > R3 (DHCP)

Sorry, but as I said in my previous message, I don't have the Asus router anymore. I sold it to buy the R3, so I cannot test what you suggested 🙃