The content of this topic has been archived between 16 Sep 2014 and 7 May 2018. Unfortunately there are posts – most likely complete pages – missing.
I highly doubt beamforming works with any open source drivers. These are all proprietary technologies.
Beamforming works with open source drivers. I can only verify that because I ran DD-WRT on my R6300, which had both implicit and explicit beamforming enabled (both were able to be enabled or disabled).
drawz wrote:I highly doubt beamforming works with any open source drivers. These are all proprietary technologies.
Beamforming works with open source drivers. I can only verify that because I ran DD-WRT on my R6300, which had both implicit and explicit beamforming enabled (both were able to be enabled or disabled).
Please point all of us to the open source driver DD-WRT uses for the Broadcom-based R6300. OpenWrt support for Broadcom will immediately improve exponentially. Oh wait, that's right, they don't exist.
JW0914 wrote:drawz wrote:I highly doubt beamforming works with any open source drivers. These are all proprietary technologies.
Beamforming works with open source drivers. I can only verify that because I ran DD-WRT on my R6300, which had both implicit and explicit beamforming enabled (both were able to be enabled or disabled).
Please point all of us to the open source driver DD-WRT uses for the Broadcom-based R6300. OpenWrt support for Broadcom will immediately improve exponentially. Oh wait, that's right, they don't exist.
I would assume they're open source, as Kong wasn't able to compile DD-WRT for the R8000 until the driver's became available (as of 6 months ago). Perhaps they're not open source and I misunderstood.
drawz wrote:JW0914 wrote:Beamforming works with open source drivers. I can only verify that because I ran DD-WRT on my R6300, which had both implicit and explicit beamforming enabled (both were able to be enabled or disabled).
Please point all of us to the open source driver DD-WRT uses for the Broadcom-based R6300. OpenWrt support for Broadcom will immediately improve exponentially. Oh wait, that's right, they don't exist.
I would assume they're open source, as Kong wasn't able to compile DD-WRT for the R8000 until the driver's became available (as of 6 months ago). Perhaps they're not open source and I misunderstood.
DD-WRT has some shady deal with Broadcom that gives them access to source, but it's not actually open to anyone else. That's why proprietary features like beamforming work on DD-WRT. It's too bad DD-WRT has so much cruft from the old days and no formal stable releases.
JW0914 wrote:drawz wrote:Please point all of us to the open source driver DD-WRT uses for the Broadcom-based R6300. OpenWrt support for Broadcom will immediately improve exponentially. Oh wait, that's right, they don't exist.
I would assume they're open source, as Kong wasn't able to compile DD-WRT for the R8000 until the driver's became available (as of 6 months ago). Perhaps they're not open source and I misunderstood.
DD-WRT has some shady deal with Broadcom that gives them access to source, but it's not actually open to anyone else. That's why proprietary features like beamforming work on DD-WRT. It's too bad DD-WRT has so much cruft from the old days and no formal stable releases.
I didn't realize that. It's fairly certain Kong either works for/with Broadcom or another router manufacturer as he was told by his employer he could no longer develop firmware for MyOpenRouter/DD-WRT and was required to pull almost all of his work from publicly available ftp server sites back in October (due to conflicts of interest with projects he was working on for his employer).
(Last edited by JW0914 on 20 Jun 2015, 03:43)
rich123321 wrote:After updating to RC2 build, wifi performance is terrible compared to chadsters AA build. 5ghz doesnt even get recognized from my xbox console upstairs, as before i was getting 44 mbps speed. Now on 2.4 ghz i get 14.5 mbps. Terrible! Any ways to boost the signal? Or is Kaoz release better
Out of curiosity, what channel are using for 5gHz? It works optimally set to something in the 150's, and I know 157 has always worked the best for me in two different locations on two entirely different routers. The reason I ask is 44mbps is extraordinarily low for a 5gHz N or AC signal (it's G speeds), however finding the channel that works best for you is trial and error, but it should always be in the 100's.
It helps to set the 5gHz to N or AC only, never legacy. 2.4 and 5 networks should never be not set to the same bandwidth frequency (20, 40, 80 mHz), and there's little reason for the 2.4 to be set above 20, with 5 set to 40 or 80.
You may also want to download and run wireshark, as WiFi works best when it's not competing in the same spectrum as other routers (much as a 2.4 gHz cordless will sound strange when you run the microwave, as they both operate in the 2.4gHz spectrum, as do many other household items). WireShark will allow you to see all named SSIDs around you and the channel they're running on.
It would also help, if you haven't already done so, if you stationed your router as close as possible to the center of the house... especially if you have more than one floor. I'd also give boosting the dB of your 5gHz channel, and you'll know the sweet spot when your connected strength on the XB1 is at least 4 bars, preferably 5. For me, I found it extraordinarily frustrating to play TitanFall while PC was also downloading/uploading files, and installing QOS, with the XB1 as #3 (Harmony Ultimate Home was 1 & 2) and set to express (priority is for small packets, not larger ones)
Thanks for the help. I set my 2.4 to 20 MHz and the 5ghz to channel 157. The frequency for 5ghz right now is set to MHz, from what you said, did you want the bandwith channels to be the same or different? Also I just moved the router upstairs next to my room. 5Ghz performance is stupid! I get 40 mbps when i have my xbox out from under my desk, but when i move it back under my desk it drops to 18. Yet the router is behind just one measly wall. also i set 2.4 transmit power to 30, and 5ghz to 17 which both seem to be the max
(Last edited by rich123321 on 20 Jun 2015, 03:51)
JW0914 wrote:rich123321 wrote:After updating to RC2 build, wifi performance is terrible compared to chadsters AA build. 5ghz doesnt even get recognized from my xbox console upstairs, as before i was getting 44 mbps speed. Now on 2.4 ghz i get 14.5 mbps. Terrible! Any ways to boost the signal? Or is Kaoz release better
Out of curiosity, what channel are using for 5gHz? It works optimally set to something in the 150's, and I know 157 has always worked the best for me in two different locations on two entirely different routers. The reason I ask is 44mbps is extraordinarily low for a 5gHz N or AC signal (it's G speeds), however finding the channel that works best for you is trial and error, but it should always be in the 100's.
It helps to set the 5gHz to N or AC only, never legacy. 2.4 and 5 networks should never be not set to the same bandwidth frequency (20, 40, 80 mHz), and there's little reason for the 2.4 to be set above 20, with 5 set to 40 or 80.
You may also want to download and run wireshark, as WiFi works best when it's not competing in the same spectrum as other routers (much as a 2.4 gHz cordless will sound strange when you run the microwave, as they both operate in the 2.4gHz spectrum, as do many other household items). WireShark will allow you to see all named SSIDs around you and the channel they're running on.
It would also help, if you haven't already done so, if you stationed your router as close as possible to the center of the house... especially if you have more than one floor. I'd also give boosting the dB of your 5gHz channel, and you'll know the sweet spot when your connected strength on the XB1 is at least 4 bars, preferably 5. For me, I found it extraordinarily frustrating to play TitanFall while PC was also downloading/uploading files, and installing QOS, with the XB1 as #3 (Harmony Ultimate Home was 1 & 2) and set to express (priority is for small packets, not larger ones)
Thanks for the help. I set my 2.4 to 20 MHz and the 5ghz to channel 157. The frequency for 5ghz right now is set to MHz, from what you said, did you want the bandwith channels to be the same or different? Also I just moved the router upstairs next to my room. 5Ghz performance is stupid! I get 40 mbps when i have my xbox out from under my desk, but when i move it back under my desk it drops to 18. Yet the router is behind just one measly wall. also i set 2.4 transmit power to 30, and 5ghz to 17 which both seem to be the max
The band width for 5gHz should be 80 (157mHz is the channel, while the 20, 40 or 80mHz options are the band width of the channel). Google will provide a much better explanation, and most articles will have a graphic (it's easier to understand electromagnetic radiation when it's visualized) to better explain why band width size matters (especially with 5gHz, as it degrades fairly quickly when hitting objects (channel 157 is 5.785gHz). The reason why 5gHz degrades far more rapidly than the 2.4gHz spectrum is physics... high frequency waves degrade faster than lower frequency ones (this is why Humpback whales can communicate with one another over thousands of miles; why elephants can literally hear a storm 100 miles away, as well as communicate via infrasonic waves with one another; and is also why you'll hear bass from a car that's blocks away, but not vocals). You sacrifice distance with 5gHz, but gain speed (although even 50 feet from my apt, I still get awesome WiFi speeds (intel 7260ac wifi card) exceeding 300mbp/s.
That's fairly odd behavior for an ac client (as the XB1 is ac)... I've been doing a bit of searching for XB1 slow WiFi and you're definitely not unique, and there's a myriad of causes. Ideally, you always want gaming systems hard lined, however that's not always possible for the end user for one reason or another. The only thing that springs to mind is possibly electromagnetic interference in that specific wall or in your office/around your desk. That's such a dramatic drop in speed, it seems to indicate interference of some kind.
Ok... i missed something in your post. I need the exact settings values you have for your 5gHz network, as something isn't configured correctly if your only seeing a max of 17dB (max should be 30dB, or 1000mw [1W]).
I'm having an issue with fw3 not applying select firewall rules, erroring with:
* Rule 'Block Inbound Traffic from NAS VPN'
! Skipping due to different family of ip address
* Rule 'Block Forwarded Traffic from NAS VPN'
! Skipping due to different family of ip address
* Rule 'Block All Other Traffic from NAS VPN'
! Skipping due to different family of zone
* Rule 'Block All Other Traffic to NAS VPN'
! Skipping due to different family of zone
* Forward 'lan' -> 'vpn'
! Skipping due to different family of zone
* Forward 'vpn' -> 'lan'
! Skipping due to different family of zoneThe rules that apply to the errors are as follows:
Rule 'Block Inbound Traffic from NAS VPN'
! Skipping due to different family of ip address
iptables -t filter -A delegate_input -s 10.10.1.0/255.255.255.0 -m comment --comment "Block Inbound Traffic from NAS VPN" -j DROPRule 'Block Forwarded Traffic from NAS VPN'
! Skipping due to different family of ip address
iptables -t filter -A delegate_forward -s 10.10.1.0/255.255.255.0 -m comment --comment "Block Forwarded Traffic from NAS VPN" -j DROPRule 'Block All Other Traffic from NAS VPN'
! Skipping due to different family of ip address
iptables -t filter -A zone_nas_input -m comment --comment "Block All Other Traffic from NAS VPN" -j zone_nas_src_DROPRule 'Block All Other Traffic to NAS VPN'
! Skipping due to different family of ip address
iptables -t filter -A delegate_forward -m comment --comment "Block All Other Traffic to NAS VPN" -j zone_nas_dest_DROPForward 'lan' -> 'vpn'
! Skipping due to different family of zone
This is set from LuCI under interzone firewall forwarding (not the same subnet or interface as NAS VPN above)
Forward 'vpn' -> 'lan'
! Skipping due to different family of zone
This is set from LuCI under interzone firewall forwarding (not the same subnet or interface as NAS VPN above)
I'm also receiving the following when issuing the command fw3 print (nvm about this one, as I'm an idiot and forgot to install ipt pkgs)
Warning: Unable to locate ipset utility, disabling ipset support(Last edited by JW0914 on 20 Jun 2015, 14:11)
A fan control flowchart created in Visio 2013 is available here:
https://drive.google.com/folderview?id= … sp=sharing
There are 2 files:
1. WRT1900ac Fan Control.png is a print screen graphic of the Visio file. Click the file to view it. Double click again will make it larger. Double clicking again will expand it one more time. This will print to an 8.5" x 14" sheet.
2. wrt1900ac fan control.vsdx is the Visio 2013 file which can be downloaded for editing if you have or access to Visio. Click the icon.
Either file can be downloaded by clicking the down arrow in the centre at the top of the window that opens.
No passwords required.
This addresses some issues with regard to load averages & higher than normal incremental temperature changes.
I am not able to program this.
Rick S
RickStep wrote:A fan control flowchart created in Visio 2013 is available here:
https://drive.google.com/folderview?id= … sp=sharing
There are 2 files:
1. WRT1900ac Fan Control.png is a print screen graphic of the Visio file. Click the file to view it. Double click again will make it larger. Double clicking again will expand it one more time. This will print to an 8.5" x 14" sheet.
2. wrt1900ac fan control.vsdx is the Visio 2013 file which can be downloaded for editing if you have or access to Visio. Click the icon.
Either file can be downloaded by clicking the down arrow in the centre at the top of the window that opens.
No passwords required.
This addresses some issues with regard to load averages & higher than normal incremental temperature changes.
I am not able to program this.
Rick S
Nice work I like the concept 
JW0914 wrote:rich123321 wrote:After updating to RC2 build, wifi performance is terrible compared to chadsters AA build. 5ghz doesnt even get recognized from my xbox console upstairs, as before i was getting 44 mbps speed. Now on 2.4 ghz i get 14.5 mbps. Terrible! Any ways to boost the signal? Or is Kaoz release better
Out of curiosity, what channel are using for 5gHz? It works optimally set to something in the 150's, and I know 157 has always worked the best for me in two different locations on two entirely different routers. The reason I ask is 44mbps is extraordinarily low for a 5gHz N or AC signal (it's G speeds), however finding the channel that works best for you is trial and error, but it should always be in the 100's.
It helps to set the 5gHz to N or AC only, never legacy. 2.4 and 5 networks should never be not set to the same bandwidth frequency (20, 40, 80 mHz), and there's little reason for the 2.4 to be set above 20, with 5 set to 40 or 80.
You may also want to download and run wireshark, as WiFi works best when it's not competing in the same spectrum as other routers (much as a 2.4 gHz cordless will sound strange when you run the microwave, as they both operate in the 2.4gHz spectrum, as do many other household items). WireShark will allow you to see all named SSIDs around you and the channel they're running on.
It would also help, if you haven't already done so, if you stationed your router as close as possible to the center of the house... especially if you have more than one floor. I'd also give boosting the dB of your 5gHz channel, and you'll know the sweet spot when your connected strength on the XB1 is at least 4 bars, preferably 5. For me, I found it extraordinarily frustrating to play TitanFall while PC was also downloading/uploading files, and installing QOS, with the XB1 as #3 (Harmony Ultimate Home was 1 & 2) and set to express (priority is for small packets, not larger ones)
Thanks for the help. I set my 2.4 to 20 MHz and the 5ghz to channel 157. The frequency for 5ghz right now is set to MHz, from what you said, did you want the bandwith channels to be the same or different? Also I just moved the router upstairs next to my room. 5Ghz performance is stupid! I get 40 mbps when i have my xbox out from under my desk, but when i move it back under my desk it drops to 18. Yet the router is behind just one measly wall. also i set 2.4 transmit power to 30, and 5ghz to 17 which both seem to be the max
Ok here we go. So far everything is working much better overall! For some reason my xbox 360 does not like this router but its ok I can use my macbook bridged connection. Every other client likes the new router placement, I even get a decent signal in my dads room. AKA if he's happy, he doesn't care what I do. I can finally use 5ghz in my living room as well! Here is my settings for the wifi.
2.4 ghz
http://i.imgur.com/CYZHIJf.png
5ghz
http://i.imgur.com/k58EDbt.png
As you can see, for some reason both on the 2.4 and 5ghz my transmit power was glitching out. At first even for the 2.4 it was set to only half. After a while messing with luci and adding upnp, i noticed the total power output let me select the full available. I guess it did the same with 5ghz so I will apply it right now. If there is anything you can think of that I can optimize from a fresh install of the RC2 build please let me know! You have been invaluable to my internet and my family thanks you haha
(Last edited by rich123321 on 20 Jun 2015, 17:40)
Hello,
I am a new user to openWrt and i have project which needs installation of node.js on linksys 1900AC. I tried many methods from the net but couldn't get much success. I am getting error that "Package uclibc version 0.2.4-1 has no valid architecture, ignoring.". Does anybody has a solution for this problem? any information would help a lot.
Ok here we go. So far everything is working much better overall! For some reason my xbox 360 does not like this router but its ok I can use my macbook bridged connection. Every other client likes the new router placement, I even get a decent signal in my dads room. AKA if he's happy, he doesn't care what I do. I can finally use 5ghz in my living room as well! Here is my settings for the wifi.
2.4 ghz
http://i.imgur.com/CYZHIJf.png5ghz
http://i.imgur.com/k58EDbt.pngAs you can see, for some reason both on the 2.4 and 5ghz my transmit power was glitching out. At first even for the 2.4 it was set to only half. After a while messing with luci and adding upnp, i noticed the total power output let me select the full available. I guess it did the same with 5ghz so I will apply it right now. If there is anything you can think of that I can optimize from a fresh install of the RC2 build please let me know! You have been invaluable to my internet and my family thanks you haha
Awesome =]
I personally would recommend (when you have time, as it's not of immediate concern) decreasing the power to both radios 1 dB at a time until the connection speed begins to drop and/or the range begins to degrade, then set it to the next highest dB when either happens. Pushing a full 1W of extra power to both radios doesn't seem like a lot, but it's an astronomical amount on the nanoscale of microprocessors (for example, the average quad core processor from Intel has ~300 billion transistors that operate on miniscule amounts of power at 1/1000 of a mw or less. That amount of power creates a significant amount of heat, and heat is the enemy of silicon based circuit boards.
You want to find the sweet spot where you're using the least amount of power possible to garnish the same result. It's not of immediate concern, as as long as the router isn't in direct sunlight and the ambient temperature where it's placed is below 80F, you probably won't have issues. The reason for the trial and error approach is you may be able to get the same speeds and range at 25 dB. You may experience degradation of speed and or range by only dropping 1 dB, and if that occurs, I'd recommend monitoring your router's temp at least once a day. That'll allow you to see where a problem is occurring and allow you to solve it before it becomes an actual issue.
There was a really good article I read a couple of months back that gave visual depictions and extremely detailed responses as to what increasing the dB range actually provides to the user. If I find it again, I'll post a link. You will almost always notice better improvement on the 5gHz spectrum by increasing the dB output, and will almost always notice a smaller improvement on the 2.4gHz spectrum. I wish I remembered what website the article was on, as it was aimed at people like us who prefer running custom firmware vs OEM, and was extremely well articulated.
trunk
openwrt r46069
trunk/toolchain/musl/patches/110-read_timezone_from_fs.patch
Is working fine, timezone is correct now.
Hello,
I am a new user to openWrt and i have project which needs installation of node.js on linksys 1900AC. I tried many methods from the net but couldn't get much success. I am getting error that "Package uclibc version 0.2.4-1 has no valid architecture, ignoring.". Does anybody has a solution for this problem? any information would help a lot.
another day, another musl error
They changed from uClibc to musl. Use RC builds which still uses uClibc.
Do you think it would help accelerate development if I put a bounty on the mwlwifi driver? I have some spare cash floating around. I'm no programer but I think the more people looking at the code the better.
Do you think it would help accelerate development if I put a bounty on the mwlwifi driver? I have some spare cash floating around. I'm no programer but I think the more people looking at the code the better.
I doubt it as there's only a handful of people who do code with OpenWRT for the WRT1900, and it's improbable more will join who haven't joined already.
Bounties are great for things like root and bootloader unlock on mobile devices, because you're working on the premise there are hundreds, thousands of people who code, have those devices, and are experienced with exploits. Not so with routers.
Even more than that, those of us who prefer custom firmware from OpenWRT or DD-WRT are an extremely small piece of the pie... much as those of us who want and need root access or an unlocked bootloader on a mobile device. 95%+ of consumers have no clue their router's OEM firmware cripples the router so severely, it's near worthless (IMO). (OEM firmware is the equivalent of having a 4k OLED and being required to view only 720p.)
(Last edited by JW0914 on 20 Jun 2015, 23:24)
RickStep wrote:RickStep wrote:A fan control flowchart created in Visio 2013 is available here:
https://drive.google.com/folderview?id= … sp=sharing
There are 2 files:
1. WRT1900ac Fan Control.png is a print screen graphic of the Visio file. Click the file to view it. Double click again will make it larger. Double clicking again will expand it one more time. This will print to an 8.5" x 14" sheet.
2. wrt1900ac fan control.vsdx is the Visio 2013 file which can be downloaded for editing if you have or access to Visio. Click the icon.
Either file can be downloaded by clicking the down arrow in the centre at the top of the window that opens.
No passwords required.
This addresses some issues with regard to load averages & higher than normal incremental temperature changes.
I am not able to program this.
Rick S
Nice work I like the concept
Thanks for the support.
I included values in the flow chart which are arbitrary.
In a coded application the fan speed values could be changed or the number of incremental changes increased/reduced. Currently the number is 8 and the value is 32. 0 - 255 (256/8 = 32); 0 - 31, 32 - 63, 64 - 95 etc.). The lowest value 31 is not used.
The time listed for "emergency cooling" (30 seconds) may not be appropriate.
I hope someone can code this and make the code available for testing.
Rick.
(Last edited by RickStep on 21 Jun 2015, 00:23)
Bummer no collectd ava.
Hostname OpenWrt
Model Linksys WRT1900AC
Firmware VersionOpenWrt Chaos Calmer r46069 / LuCI (git-15.168.50780-bae48b6)
Kernel Version 3.18.16Local TimeSat Jun 20 19:12:50 2015
Uptime 0h 19m 22s
Load Average0.03, 0.02, 0.05
Bummer no collectd ava.
Hostname OpenWrt
Model Linksys WRT1900AC
Firmware VersionOpenWrt Chaos Calmer r46069 / LuCI (git-15.168.50780-bae48b6)
Kernel Version 3.18.16Local TimeSat Jun 20 19:12:50 2015
Uptime 0h 19m 22s
Load Average0.03, 0.02, 0.05
https://dev.openwrt.org/ticket/19911
BTY, when I was running RC2 it was working.
(Last edited by gufus on 21 Jun 2015, 00:40)
northbound wrote:Bummer no collectd ava.
Hostname OpenWrt
Model Linksys WRT1900AC
Firmware VersionOpenWrt Chaos Calmer r46069 / LuCI (git-15.168.50780-bae48b6)
Kernel Version 3.18.16Local TimeSat Jun 20 19:12:50 2015
Uptime 0h 19m 22s
Load Average0.03, 0.02, 0.05https://dev.openwrt.org/ticket/19911
BTY, when I was running RC2 it was working.
I know.. It was here also. But it good practice to start over I usually flash a new ver on Sat after work to relax.
Do you think it would help accelerate development if I put a bounty on the mwlwifi driver? I have some spare cash floating around. I'm no programer but I think the more people looking at the code the better.
It's not much, but I'm willing to put 25 bucks in the pot.. If others are willing to contribute, then maybe that one developer might be willing to devote more effort to getting the job completed sooner.
Just a thought.
Also, what currently happens when you go to access one of the shares you can see under OpenWRT?
Thank you for the info I finally have samba working and even with the ntfs I still get 30 meg down but only 6 up on a wd passport. not as fast as the factory.img but acceptable for now.
Thanks again for your time.
A fan control flowchart created in Visio 2013 is available here:
...
I am not able to program this.
Rick S
Thank you for this, Rick! With it, I worked up something in a shell script that follows this pretty closely. You can download it here (apologies if the formatting/styling is weird, I'm not used to working in shell):
https://raw.githubusercontent.com/jjack … control.sh
It doesn't do any differential math for the 5 second temperature checks 'cause I'm not sure how you'd do that in a shell script, so I've got it just looking for a 2 degree temperature change instead as a simple hack (the temperatures come back as integers, so a 0.5 degree check there wasn't going to work and I noticed that my temps generally fluctuate by about a degree every 5 seconds anyway).
As another note, while I added it in, checking all 3 load averages might be a little superfluous and could result in the fans running up 15 minutes longer than potentially necessary. Checking the 1 minute value would likely be sufficient.
It's got a verbose toggle at the top of the script that you can use for a little more output for debugging and testing. I tried commenting enough to hopefully help the non-programmers figure things out.
To run it
- download the script or save it to your router and be sure to "chmod +x fancontrol.sh"
- for testing and debugging, run it by hand with "./fancontrol.sh"
- if you want to have it run all the time or in the background, run it with "./fancontrol.sh &" (make sure that VERBOSE is off!)
I believe that I have all of the logic in place, I just haven't had a chance to test it very much. Please try it out and let me know how well it works for you all.
DISCLAIMER: this is beta and might blow your router up and catch your house on fire and cause your cats to climb your curtains. But probably not. That would be really weird. I'm not responsible for any damages caused to you, your property, or your pets by this software. But I will be really sad if something does happen.
(Last edited by jeremyjack on 21 Jun 2015, 05:10)
Kaloz wrote:gufus wrote:Pushed to both CC and trunk.
Fantastic! I'm curious what the results will be when I put it on a couple of production WRT1900AC V1 units.
Crap it looks like they only fixed that issue on chip 8897.
Which router has chip 8897? Why wouldn't they fix 8864 first which is the chip most are concerned about?
Maybe I'm missing something here but I can't roll this change out in any case.
