Jump to content

Has anyone been successful in achieving an A+ bufferbloat with any netduma?


Max2110
 Share

Recommended Posts

If so, can you please share your setup. Thanks 

FYI. I am never able to achieve even a b on bufferbloat. It's pointless lowering the sliders to less than 70% due to other devices on my network

Link to comment
Share on other sites

Just now, Max2110 said:

@Netduma Fraserwhat am I doing wrong? I literally tried almost all different percentages in the sliders. What else can I do?

@Netduma FraserI have ddwrt wrt on a tplink router and I get 0 bufferbloat there but I don't have any netduma features which I need for gaming. Is it possible to make the ddwrtwrt handle the bufferbloat problem while still getting the benefits from the netduma. If so how could this setup work? 

 

Link to comment
Share on other sites

  • Administrators

What router and firmware version are you currently on and how/where are you testing Bufferbloat? You could do that, you'd just need to do modem > router > DumaOS router with the DumaOS router WAN IP input in the DMZ of the other router so you can get an Open NAT.

Link to comment
Share on other sites

1 hour ago, Netduma Fraser said:

What router and firmware version are you currently on and how/where are you testing Bufferbloat? You could do that, you'd just need to do modem > router > DumaOS router with the DumaOS router WAN IP input in the DMZ of the other router so you can get an Open NAT.

Got it and which router should I connect my devices to? Ddwrt or netduma? 

Link to comment
Share on other sites

1 hour ago, Netduma Fraser said:

What router and firmware version are you currently on and how/where are you testing Bufferbloat? You could do that, you'd just need to do modem > router > DumaOS router with the DumaOS router WAN IP input in the DMZ of the other router so you can get an Open NAT.

I'm on the latest xr500 firmware

Link to comment
Share on other sites

  • Administrators

I'm not familiar with the QoS options it has so couldn't say for certain, I assume it has specific settings you can apply to devices so you could connect most to that router. 

It is highly likely that Traffic Prioritization is adversely affecting the test, a better test would be to saturate your connection with downloads, streams etc (which is actually when bufferbloat will occur), while following this guide and experimenting with percentages on Congestion Control (while set to Always) to see what brings down the ping results as much as possible.

Link to comment
Share on other sites

3 hours ago, Netduma Fraser said:

I'm not familiar with the QoS options it has so couldn't say for certain, I assume it has specific settings you can apply to devices so you could connect most to that router. 

It is highly likely that Traffic Prioritization is adversely affecting the test, a better test would be to saturate your connection with downloads, streams etc (which is actually when bufferbloat will occur), while following this guide and experimenting with percentages on Congestion Control (while set to Always) to see what brings down the ping results as much as possible.

IL give it a try. Thanks 

Link to comment
Share on other sites

Forgive me, I am probably an idiot here, but are you testing it correctly?

Moving the sliders makes no difference unless the Congestion Control is active. You can run the buffer bloat test without Congestion Control enabled just to get some base-line numbers but once you have those then you should turn Congestion Control to "on all the time" for the purposes of testing. You can switch it to "Auto-Enable" later on once you have completed the testing but do not have it set to that for testing.

The base-line figures are the worst case. Then with Congestion Control on, you can then start by say setting the sliders to 50%, 50% and see if there is improvement. Then taking them higher or lower from there. I would ignore the A B C rating rather just pay attention to improving it as much as you can.

In my case the download was great but the upload was horrific. I changed the sliders to 90% download and 70% upload. This kept the figures the same when idle and under load, which is the best you can hope for.

 

 

Link to comment
Share on other sites

  • 4 weeks later...

Hello, I have had it where I can get wave form bb test to always show a+ using xr500. It is mostly pc settings from registry edits, using tcp optimizer settings, and changing ethernet adapter settings. Good luck

20220104_002757.jpg

Link to comment
Share on other sites

21 hours ago, DMaR said:

Hello, I have had it where I can get wave form bb test to always show a+ using xr500. It is mostly pc settings from registry edits, using tcp optimizer settings, and changing ethernet adapter settings. Good luck

20220104_002757.jpg

hello do you have optimal settings or links to advise for the ethernet network card? thank you

Link to comment
Share on other sites

I have an Intel ethernet adapter, settings might be different. 

Adaptive interframe spacing- Disabled

Enable PME- Enabled

Energy Efficient Ethernet- Off

Flow Control- Disabled

Gigabit Master slave mode- Auto Detect

Interrupt Moderation- Disabled

Interrupt Moderation Rate- Adaptive

Ipv4 Checksum Offload- Disabled

Jumbo Packet- Disabled

Large Send Offload V2 ipv4- Disabled

Large Send Offload v2 ipv6- Disabled

Legacy Switch Compatibility Mode- Disabled

Link Speed Battery Saver- Disabled

Locally administered Address- Not present

Log Link State Event- Enabled

Maximum Number of RSS Queues- 2 Queues

Packet Priority & VLAN- Disabled

Protocol ARP Offload- Disabled

Protocol NS Offload- Disabled

PTP Hardware Timestamp- Disabled

Receive Buffers- 256

Receive Side Scaling- Enabled

Reduce Speed on Power down- Enabled

RSS load Balancing profile- Closest Processor

Selective Suspend- Enabled

Selective Suspend Idle Timeout- 20

Software Timestamp- Disabled

Speed & Duplex- Auto Negotiate 

System Idle Power Saver- Disabled

Tcp Checksum Offload ipv4 & ipv6- Disabled

Transmit Buffers- 512

UDP Checksum Offload ipv4 & ipv6- Disabled

Ultra Low Power Mode- Disabled

Wait for link- Auto Detect

All the wake on settings- Disabled

These were the settings I was using for my ethernet adapter at the time I took that picture. I've reformatted my Pc since then, So I could not tell you the exact registry edits I had in place to achieve exactly that in the picture (I Can tell you it was 100% consistent when testing), yet I will say try downloading the TCP optimizer from Speed Guide and play with the settings, load the optimal settings, then click custom there is a few to change around. Windows Auto tuning change to Highly Restricted. On the advanced tab, change the system responsiveness to gaming (0), Disable Nagle's algorithm, play around with some settings to test it if you want, but that should get you close enough to that picture without reducing your bandwidth from approximately 600 Mbps to 120 Mbps for an A+ score, like for instance if you were to fully disable the windows auto tuning setting.

I have had settings where I always got 5ms and + 1, + 1, always. But with what I have now it is not as consistent but should still give A + score. more in the range of 4-7ms and underload +1-5ms, +1-5ms. So hopefully that helps anyone enough who reads this, that extra trick to get it more consistent how I had previously I believe requires more manual Reg edits So attempt that at your own risk, if you want to delve down that rabbit hole. If you happen to stumble upon those settings report back because I would also want to have it back like I did but have not had the time to trial and error it back to that. 

Link to comment
Share on other sites

I'll add that settings like your dns will effect this test. Also, I am using the microsoft edge browser for the buffer bloat test, edge settings like hardware acceleration, efficiency mode, trackers, security settings can effect this test. You can try disabling microsoft defender settings, and Defender firewall settings temporarily (only for this test but do so at your own peril) I would not recommend leaving them like that. But I say this just to show you how many settings can and do have impacts on the results you see. Its up to you how you want to play around and trial and error all these variables to find the configuration that you are comfortable with, which can take a large amount of time to play with and learn what works best, anyways good luck and happy gaming 

Link to comment
Share on other sites

3 hours ago, DMaR said:

I have an Intel ethernet adapter, settings might be different. 

Adaptive interframe spacing- Disabled

Enable PME- Enabled

Energy Efficient Ethernet- Off

Flow Control- Disabled

Gigabit Master slave mode- Auto Detect

Interrupt Moderation- Disabled

Interrupt Moderation Rate- Adaptive

Ipv4 Checksum Offload- Disabled

Jumbo Packet- Disabled

Large Send Offload V2 ipv4- Disabled

Large Send Offload v2 ipv6- Disabled

Legacy Switch Compatibility Mode- Disabled

Link Speed Battery Saver- Disabled

Locally administered Address- Not present

Log Link State Event- Enabled

Maximum Number of RSS Queues- 2 Queues

Packet Priority & VLAN- Disabled

Protocol ARP Offload- Disabled

Protocol NS Offload- Disabled

PTP Hardware Timestamp- Disabled

Receive Buffers- 256

Receive Side Scaling- Enabled

Reduce Speed on Power down- Enabled

RSS load Balancing profile- Closest Processor

Selective Suspend- Enabled

Selective Suspend Idle Timeout- 20

Software Timestamp- Disabled

Speed & Duplex- Auto Negotiate 

System Idle Power Saver- Disabled

Tcp Checksum Offload ipv4 & ipv6- Disabled

Transmit Buffers- 512

UDP Checksum Offload ipv4 & ipv6- Disabled

Ultra Low Power Mode- Disabled

Wait for link- Auto Detect

All the wake on settings- Disabled

These were the settings I was using for my ethernet adapter at the time I took that picture. I've reformatted my Pc since then, So I could not tell you the exact registry edits I had in place to achieve exactly that in the picture (I Can tell you it was 100% consistent when testing), yet I will say try downloading the TCP optimizer from Speed Guide and play with the settings, load the optimal settings, then click custom there is a few to change around. Windows Auto tuning change to Highly Restricted. On the advanced tab, change the system responsiveness to gaming (0), Disable Nagle's algorithm, play around with some settings to test it if you want, but that should get you close enough to that picture without reducing your bandwidth from approximately 600 Mbps to 120 Mbps for an A+ score, like for instance if you were to fully disable the windows auto tuning setting.

I have had settings where I always got 5ms and + 1, + 1, always. But with what I have now it is not as consistent but should still give A + score. more in the range of 4-7ms and underload +1-5ms, +1-5ms. So hopefully that helps anyone enough who reads this, that extra trick to get it more consistent how I had previously I believe requires more manual Reg edits So attempt that at your own risk, if you want to delve down that rabbit hole. If you happen to stumble upon those settings report back because I would also want to have it back like I did but have not had the time to trial and error it back to that. 

Thank you very much for all this information I will try on my side and I will give you a report after several days of testing I have on a realtek 2.5g card and I feel like I have pretty much the same settings

Thank you

Link to comment
Share on other sites

1 hour ago, dmar613 said:

I'll add that settings like your dns will effect this test. Also, I am using the microsoft edge browser for the buffer bloat test, edge settings like hardware acceleration, efficiency mode, trackers, security settings can effect this test. You can try disabling microsoft defender settings, and Defender firewall settings temporarily (only for this test but do so at your own peril) I would not recommend leaving them like that. But I say this just to show you how many settings can and do have impacts on the results you see. Its up to you how you want to play around and trial and error all these variables to find the configuration that you are comfortable with, which can take a large amount of time to play with and learn what works best, anyways good luck and happy gaming 

DNS have that much impact? on the other hand for windows defender I prefer to leave it on Thanks for your advice ;)

Link to comment
Share on other sites

 Share

×
×
  • Create New...