Jump to content
Reminder, starting today you will no longer be able to login to the forum using your display name, to login you must now use your email address. ×

Question about qos algorithm and hit detection constistency issues


Recommended Posts

With the r1 there was two options, reactive or preemptive.

Since I don't see the option in DumaOs on the XR500, does that mean that the router automatically decides based on speeds or does DumaOs only have one algorithm?

 

A pay for 50 down and 10 up but my actual speeds are 60 down and 12 up (Comcast Business internet).

I noticed on the r1 that I had better games with preemptive selected but that I would get disconnected from the activision servers after matches in WW2 (work around provided by sim).

 

Since I don't get disconnected from the activision servers anymore after matches I would guess that DumaOs either uses a different algorithm than preemptive or that the XR500 is choosing another algorithm due to the speeds that I have.

 

I also noticed with the r1 that if I chose reactive that my ping would typically be higher (as shown on the xbox with 3 bars and the in-game ping monitor > 100 ms). I notice that this issue comes and goes with the XR500. Rebooting usually fixes this problem temporarily but it always seems to return eventually. 

 

I upgraded to the latest firmware that just came out so maybe that will help but I haven't had enough time to properly test the new firmware.

 

I have left everything with default settings except for the qos settings which I set anti bufferbloat sliders to 70%. (same as I used on the r1)

Anti bufferbloat is set to apply "when high priority traffic is detected" and duma classified games is enabled.

When I'm gaming I see the light for high priority traffic detected and see the packet count increasing. When I turn the xbox off then the light goes out and the packet count stops increasing. 

Doing speed tests show that the anti bufferbloat is working as it is supposed to.

Xbox and COD both say open NAT.

 

In general the XR500 feels more consistent overall. But with the r1 it felt like the entire match was either good or bad. With the XR500 it feels like some parts of the match are good and other parts are bad.

 

Any suggestions would be appreciated to help with the hit detection consistency.

I understand that games will vary based on factors outside of my home but the hit detection was much more on point and consistent game after game with the r1 and preemptive algorithm (the only problem was having to restart the whole game after every match).

P.S. I did try using the geo filter on the r1 and the XR500 and it never seemed to make a difference for me (I'm located in the US state of Illinois)

 

 

Thank you for making COD playable for me with both the r1 and now the XR500. 

You guys truly do an amazing job with both router firmware and customer support.

 

 

 

Link to comment
Share on other sites

  • Administrators

Hi Pete,

 

In DumaOS on the XR500 there is no need for two algorithms because the XR500 has a very powerful chip (quadcore). Two of the cores are for general purpose (anything not network related) and the other two cores are dedicated to hardware accelerated packet processing. This means our Anti-Bufferbloat can be applied at the highest speeds, which the R1 was unable to do via hardware so we had to introduce two solutions.

 

Based on the description of your settings it sounds like your QoS is setup correctly. So any variance in performance is far more likely to be caused by the CoD dedicated servers you are playing on.

 

What are your Geo-Filter settings? Does your radius cover more than one server location? I'd recommend playing with these settings to get the best connection for you. Many users like to set their home location to their favourite server base so they can guarantee a consistent gaming experience.

 

Thank you for your kind words about the firmware and for ordering both the R1 and the XR500!

Link to comment
Share on other sites

  • Administrators

I thought the XR500 had a Dual core not a Quadcore? 

 

It's actually a misprint on the Netgear spec list. We have asked them to update that. Our Developers have worked very intimately with the hardware and they have said it's a quad core.

Link to comment
Share on other sites

  • Administrators

Im assuming the CPU usage app then will also be updated to reflect this? 

 

As far as I am aware it will only show two (the one's for general purpose) because those are the ones where the CPU usage is likely to fluctuate. I will consult the Development team to get clarification on this.

Link to comment
Share on other sites

  • Administrators

As far as I am aware it will only show two (the one's for general purpose) because those are the ones where the CPU usage is likely to fluctuate. I will consult the Development team to get clarification on this.

 

That is incorrect, the 2 graphs you see are general-purpose cores. The two hidden ones are dedicated to network processing. You can confirm this is true because the CPU graph doesn't rise when maxing out the connection :) 

Link to comment
Share on other sites

Hi pete, I'm just checking in to see if you're still having inconsistencies in gameplay?

 

Hi Fraser, thank you for checking. Unfortunately, the problem still persists. I wanted to wait to update until I had time to adequately test.

Using geo filter does not improve the quality of my games but it doesn't hurt either so I can leave that on.

 

There are times in a match where the hit detection is on point and then all of a sudden it will feel like I'm shooting an air soft gun.

The hit detection seems very similar to what I experienced with the r1 using reactive algorithm.

I wish it was more like the preemptive algorithm because that always provided much better hit detection. The only problem I had with the r1 and preemptive was the issue of losing connection to the activision servers after almost every match. That issue is gone with the XR500 just like it would go away with the r1 using reactive but I wish the hit detection felt more like the r1 with preemptive (the hit detection was usually on point the entire match rather than just some of the time).

 

The only difference in my setup now is that I only have one router hooked up (modem to XR500 with xbox hardwired to XR500).

Before I had Modem to r1 with xbox hardwired to r1 and then a hardwire from the r1 to another router to handle the wifi (wifi on the r1 disabled).

Link to comment
Share on other sites

  • Administrators

Are you actually experiencing any lag with the XR500 or is it just hit detection? What is the average ping you're getting as shown by the Geo-Filter?

 

QoS uses hardware acceleration now so there is no need for algorithms, so theres much more power going towards eliminating any local congestion.

Link to comment
Share on other sites

Are you actually experiencing any lag with the XR500 or is it just hit detection? What is the average ping you're getting as shown by the Geo-Filter?

 

QoS uses hardware acceleration now so there is no need for algorithms, so theres much more power going towards eliminating any local congestion.

 

Usually its just the hit detection but occasionally I get actual bad lag (ping in geo filter over 100 ms and yellow bars in game, this happens with or without geo filter).

Most of the time my ping is under 20 ms with an occasional spike less than 30 ms and green bars in game.

Regardless of the ping, the hit detection varies a lot for me. One thing I tried that seemed to help a little bit was in the WAN settings where I changed Nat Filtering from secured to open but that only seemed to make a small difference. 

I'm not sure how the algorithms worked or how the nat acceleration works but there was a clear difference for me with hit detection when I used the preemptive on the r1 vs reactive. 

On the XR500 the hit detection is great at some times and then horrible at other times.

Another thing I noticed was that the hit detection seemed to improve a little bit when I enabled ipv6 (although I'm not sure if I set it up correctly, i just selected the auto detect, since ipv6-test gave me a 5 out 10 but the xbox did say that I was connected via ipv6 and ipv4).

I read in a different post that ipv6 bypasses the download qos (which I confirmed doing speed tests) so I have it disabled now. I sent you a PM to get the fix once it's available so that I can try that out.

 

My wild guesses on the issue after switching from the r1 are:

1) with the r1 and a different router for wifi, perhaps if it didn't have the fancy MU MIMO and stuff then the wireless devices on my network were only allowed to access the internet one at a time which may have helped my game packets from the xbox that was hardwired to the r1 go through faster.

Or

2) maybe the XR500 is giving me so much better of a connection vs the r1 with preemptive that lag comp is penalizing me more than it was before.

Or

3) Most likely, I just need to find the right settings.

 

Thanks for all the help, I will be happy to try any settings that you guys suggest. I usually try to play at least 5 to 20 games with a specific setting in order to make sure it's not placebo or just a coincidence that I got a good lobby or a bad lobby at the same time that I changed the setting. So keeping that in mind it usually takes me a few days to test out specific settings.

Link to comment
Share on other sites

Great thank you for the information, it sounds more likely to be an ISP issue. If you follow this guide: http://forum.netduma.com/topic/23881-ping-plotter-quick-guide/and provide us with the screenshots we can determine if that's the case or not.

 

All of these tests were done with my computer hardwired to either the modem or the XR500.

 

Below is the ping plotter run connected directly to the modem:

http://forum.netduma.com/uploads/images/1518666505-U5405.png

 

Below is connected to the XR500 with qos sliders at 70% and youtube streaming on another device 

http://forum.netduma.com/uploads/images/1518666944-U5405.png

 

Below is connected to the XR500 with no qos and youtube streaming on another device.

http://forum.netduma.com/uploads/images/1518667071-U5405.png

 

Below is connected to the XR500 with an idle line and qos sliders at 70%.

http://forum.netduma.com/uploads/images/1518667240-U5405.png

 

Below is connected to the XR500 with an idle line and no qos.

http://forum.netduma.com/uploads/images/1518667330-U5405.png

 

Let me know if you need me to run any other tests.

Thank you

Link to comment
Share on other sites

Starting at hop four you have ISP packet loss this is only an issue if it ends up showing at the destination hop, which it does.

 

Add this to 10ms average jitter, you have an ISP issue.

 

Use the ping plotter results from the modem to show your ISP the problem.

Link to comment
Share on other sites

Starting at hop four you have ISP packet loss this is only an issue if it ends up showing at the destination hop, which it does.

 

Add this to 10ms average jitter, you have an ISP issue.

 

Use the ping plotter results from the modem to show your ISP the problem.

 

Yes as Zennon has said it's an ISP issue, hopefully with the modem evidence you've gathered (don't show them the XR500 plots) they'll be able to improve it.

 

Thanks for the help.

I will try to get my ISP to work on this issue but I'm not sure how helpful they will be (it's comcast but its business internet so I might have a small chance of getting some help).

 

If they try to blame it on my personal modem and recommend that I use one of theirs, should I entertain the idea? 

I currently have a motorola arris SB 6183 modem.

 

Also, if they say they need to send a technician out to check the problem, is there any chance that it could be do the lines inside my house? (Comcast does not cover them and would therefore charge me to come out and tell me its nothing they are responsible for.

Link to comment
Share on other sites

  • Administrators

Thanks for the help.

I will try to get my ISP to work on this issue but I'm not sure how helpful they will be (it's comcast but its business internet so I might have a small chance of getting some help).

 

If they try to blame it on my personal modem and recommend that I use one of theirs, should I entertain the idea? 

I currently have a motorola arris SB 6183 modem.

 

Also, if they say they need to send a technician out to check the problem, is there any chance that it could be do the lines inside my house? (Comcast does not cover them and would therefore charge me to come out and tell me its nothing they are responsible for.

 

That is good news that it's business internet, if you stress it's affecting your ability to do work then they'll be more inclined to help. I wouldn't have thought the modem would be the issue but no harm in trying that. Yes there is that chance, I don't know how you'd be able to check that though. Zennon may have some more advice on that for you.

Link to comment
Share on other sites

That is good news that it's business internet, if you stress it's affecting your ability to do work then they'll be more inclined to help. I wouldn't have thought the modem would be the issue but no harm in trying that. Yes there is that chance, I don't know how you'd be able to check that though. Zennon may have some more advice on that for you.

 

Update: I called my ISP today and they ran a remote signal test and identified a signal issue so they are sending a technician out tomorrow to try and resolve it.

I can't wait to see how the XR500 performs when there are no ISP issues.

 

 

 

 

 

I used an sb 6183 with the R1 and it was crap. Not sure if it has the puma chipset but things got better when I switched to Netgear cm600.

 

Thanks for the tip Ed.

I will look into getting a new modem if I can convince the wife that we need it (I just bought a brand new router so I already owe her a new tattoo).

I'll wait and see what the technician says tomorrow and possibly try out the comcast modem first.

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...