Discussion in 'Tomato Firmware' started by shibby20, Feb 26, 2011.
Works on MIPS builds. Don't have an ARM router to test.
how do u go about doing it on your router, inside or outside dhcp range? Did u find that it has to be outside dhcp?
Edit: Actually, I got moderate alert on one xbox again using static dhcp out of dhcp range too.
Try with CTF disabled. If that doesn't work, then there may be an issue with UPnP in Tomato ARM when multiple devices try to use the same ports.
please read changelog first!
Flashing now, I'll report QoS and ping issue. Also Samba improvement if I notice one. Give me 10-15 minutes my friend!
LED's don't work but I remember you saying it wasn't a priority. Ping seems to have gone higher too. Also, Samba is slower.
Nevermind, didn't notice that QoS on ARM isn't the same as MIPSR2. Any specific reason that QoS on ARM doesn't do Inbound limit anymore? I loved it. Bandwidth Limiter is too much work for me honestly. I constantly upgrade my router to the newest version and setting it up everytime is too much. Anyway to have QoS do both Inbound and Outbound limit?
Well I haven't yet enabled CTF, but ping are still currently bouncing around. After first flashing, and rebooting the router pings were pretty stable. But then I started applying settings, and watched as pings starts to bounce around. I will be enabling CTF in next couple minutes, and I will see how the pings look after that.
Well just flashed Tomato 121 VPN to my new R7000. So far everything is amazingly good.
As I have a 100/50mbps Optical internet connection, I've never been able to reach 100mbps on my N16, it always capped at ~92mbps DS/50mbps US.
Now I'm getting ~110/50mbps @ 16ms!!!!!
Runnings tests with CTF on and Jumbo On. So far so good!!
Thanks Shibby for this amazing piece of software!!!
As NoobWRT mention, Samba is slower. As for me I use a USB3 NTFS drive, and writing to the drive max I see on 121 is 14MB's, avg is around 10-11MB's. On your last release I am pretty sure writing speed's were right around 30MB's for me, on this same drive.
The QoS method used on MIPS requires lowlevel structure changes in the kernel's networking stack (to the skbuf structure, to be more accurate). These changes aren't part of the kernel Asus (and Broadcom in general) use for ARM. Applying those changes results in the wireless driver no longer working since it was compiled for a kernel without those structure changes. And since the driver is closed source... That means a different QoS technique must be used for ARM devices.
Thanks for the update Shibby, however VLAN for RT-N66U is not fixed.
When you say "vlan order" do you mean the port order or the VID order? Because the port order still appears to be wrong on the VLAN page. Since before 121, If I assign "Port 1" to a VLAN, I have to physically plug into port 4 on the RT-N66U. This has not changed in 121.
VID Offset is still broken in 121 on the K26RT-AC build. If I change my VID offset in the K26RT-N (SDK5) build, all is well and it appears to function normally, as it has in the past. However if I change the VID Offset in K26RT-AC (SDK6) build, all ethernet ports stop working. I don't know if this happens on any other router, but it happens on RT-N66U still.
I also tried resetting NVRAM to no avail.
Can anyone else confirm?
Well not to keep spamming or anything, But I went ahead and went back to DD WRT for now on my R7000. As I tried many of things, as the ping issue is still an issue, and Samba performance is about half of what it was for me in your 120 build.
I plan to continue to test all the new tomato build's, and give my feedback on what I see. I really appreciate the time, and effort, you guy's are putting into the ARM router build's.
I've got something weird and i do need some assitance.
I've got a Cisco Linksys E3200 and i've just uploaded to the latest version : Tomato Firmware 1.28.0000 MIPSR2-120 K26 USB AIO
I've configured two wifi network. One mixed on the 20Ghz and One N-Only on the 40ghz.
If i connect my iphone on the 'N-Only', all is fine and i can get access to my appletv (wired connection) via Airplay.
If i connect my iphone on the 'Mixed' mode, my iphone seems not detecting the airplay function of Appletv. The bonjour protocol seems working as i can connect my laptop to my appletv. So this is limited to iphone but i still do not understand why on 'n-only' this is working and not on the mixed network.
So pretty much it'll never work like MIPS?
I'm having a difficult time getting Shibby mod to detect ethernet ports on the WRTSL54GS correctly. It wont detect the WAN port at all, thinks port 1 is WAN, port 2 is 3, and so on. Also can't renew ip from the modem. This happened on build 120 and 121, but wasn't an issue on RAF. NVram has also been cleared many times. I am using tomato-K26USB-1.28.RT-MIPSR1-121-Big-VPN.
Think maybe if we asked nicely we could get them to make a blob driver compatible with said kernel modules? No harm in asking anyway.
Unless they re-implement it using a different packet classifier that could implement the same functionality without requiring patches to the network stack.
Not so simple. In addition to having to re-ask again every time the wireless driver is updated, it would also require them to contact Broadcom to request a recompiled version of CTF. So, quite unlikely to ever happen IMHO.
Can you check in Syslog that the drive does get mounted by xhci (USB3 driver) and not ehci (USB2)?
In theory, Tomato should be able to reach 50 MB/s read speed and 30 MB/s write speed with ext3. No idea what performance could be expected out of ntfs, it will depends (amongst other things) on whether you're using UFSD or NTFS-3G.
I would but I already switched back to DD WRT on my R7000. As the ping issue is still a problem, and that's a big issue to me to be fixed. I would play with tomato more, as I expect minor issue's. But this R7000 is being a PITA with this tricky unstable ping issue.
I game competitively on my router and it's causing me to get killed constantly from high ping. I'm not complaining, this is done with Shibby's free-time but I'm just putting it out there. RT-AC66R ran wonders with it's QoS, I like the R7000 because it's DLNA and Samba is MUCH faster than the RT-AC66R's.
still waiting for implicit/explocit beamforming function next build 122
Bought a Tenda N60 to see what its like, waited for Shibby's 121 release to even unbox it. I'm impressed! After an hour online and configuring it the way I like, it's working GREAT on Comcast under ipv6. Much snappier UI than the Linksys e2500v3 it has now replaced. Thank you Shibby for another great release.
I see that someone else mentioned an issue earlier in the thread with their Chromecast. I seem to be having a different issue though. Maybe someone else can confirm if they are having a similar issue.
This issue only happens on the R7000. I have no issues on my RT-N16 running the same version of Shibby build.
After the Chromecast sits for awhile without being used, I can no longer connect to it. I have to reboot the router in order to stream to it again. I see it listed in the devices list in the router GUI. I just don't see it in the device i'm trying to use to stream to it. After I reboot the router it shows up and I can stream again.
Both the RT-N16 and the R7000 are set up with the exact same settings because the R7000 was purchased to replace the RT-N16.
I have tried re-flashing the router. I even went back the to original firmware and re-flashed again. I followed the instructions to the tee. I still have this issue.
I'm currently back to using the RT-N16 with no issues.
How long do you have to let it sit before noticing this? I have a chromecast that I haven't tried yet. I just flashed 121 tonight, so I can test and see if I have the same issue on my R7000.
Whats the difference between the tomato-K26USB-1.28.RT-N5x-MIPSR2-121-AIO-64K & tomato-RT-N66U_RT-AC6x--121-AIO-64K for the RT-N66U.
Seems both will work but which is better?
Tomato Firmware 1.28.0000 MIPSR2-120 K26 USB AIO-64K
Linux kernel 126.96.36.199 and Broadcom Wireless Driver 188.8.131.5212 updates
Tomato Firmware 1.28.0000 MIPSR2-121 K26AC USB AIO-64K
Linux kernel 184.108.40.206 and Broadcom Wireless Driver 220.127.116.11 (r341183) updates
Nice, new version with QoS. I will flash my R7000 tomorrow and then see if I can replicate the USB 3 performance issue compared to v120. What's a good way (free tool) to measure this from PC/MAC so I can compare before (120) and after (121)?
Also, I have a problem with a new RT-N10P. I wanted to flash it with Tomato. I checked the Tomato Anon list and there are 50+ boxes listed using v120 K26 RT MIPRS2 MAX. It's only 6MB or so. I tried 5 times to flash via the Asus recovery utility always got the "success" message at the end, but when the router boots up I cannot get anywhere. It shows a power light and a LAN cable light but I cannot telnet or reach it via regular 192.168.1.1 from my static .2 IP. I waited quite some time in between and then tried to restart and booting with WPS pressed etc. No web portal. Is there a trick to get it on the N10P or am I using the wrong firmware? Thx.
I have this issue sometimes after flash tomato. Hardwire from router to pc, go to your Ethernet card properties and change your ipv4 properties settings to manual , set your ip 192.168.1.2 , subnet 255.255.255.0 , gateway 192.168.1.1 then apply. Wait a min and try to access tomato interface from browser again. you then should be able to access tomato GUI login
@kamaaina: The "success" message from the firmware restoration tool might be misleading. Give the router some time to flash the firmware after it is loaded (I had to leave my RT-N66Us for more than 25 min., which seemed like an eternity to me ). The best way to know when it is done flashing is to run a ping 192.168.1.1 /t from your PC and see when it starts responding.
Also, make sure to reset the NVRAM before and after flashing the firmware. I don't know about RT-N10P, but NVRAM reset and 30-30-30 reset on RT-N66U are done using the WPS button, not the reset button.
Hope this helps.
I am working on correcting/verifying vlan port order for almost all routers
you can't "get 110/50 @16ms". that makes about 0 sense. try running a ping to google when you do a speedtest, rather then looking at the ping speedtest takes BEFORE it starts saturatiing your connection....
just saying.. u cant have 110/50 @ 16ms, that statement does not make sense
you can get 110/50 and have 16ms as your lowest ping to a server, but your ping will never stay 100% the same under full load........
I've never timed it. It's always just been the next day. I let my daughter watch stuff streamed from Plex to her Chromecast at night. Then the next night l go to do it again and so far every night i've had to reboot the router. I may just try another full re-flash just for the heck of it. It's not that difficult.
Let me know though if you have the same issue please and thank you.
Which version is the most bare bones version for Linksys E4200? Including 5GHz support.
What's a good Transmit Power for R7000 on 121? It seems whenever I use TomatoUSB my iPhone 5 or iPad are soooo slow downstairs. Current Transmit Power is 80 on 2.4 band and 115 on 5.8 band. Anything else I can tweak that'll give me more range? YouTube is impossible on even 360p, 240p is just too bad to watch any videos on honestly.
More transmit power != more range.
Set the transmit power to 0 and let the defaults handle it. I find that I have a HUGE range with the default power setting on my R7000.
More power = more noise and that can actually SHORTEN the range.
So 0 works best?
for iphone and ipad best performance, use 40MHz (if you don't have an AC devices and can afford to) then use 50mw transmit power, and use the HIGHEST channel possible. i think on 40MHz the highest channel u can get is 161 or 165.
if you dont get improved performance and throughput with these settings i will be suprised. i certainly noticed a massive throghput jump with 50mw power on higher channels, also setting location to singapore could help. i do it
Also, I'm running 40MHz on both bands because I don't have any AC device that can support up to 1300 MBPS anyway so I use 40MHz because I know 40MHz gets more range than 80MHz, I'd use 20MHz but I need the extra speed.
ive seen 95/150mbit link on iphone 5 with 40MHz channel 161 at 50mw i believe.
I'm getting 150/24 link right now and my iPhone 5 is a few inches away from the R7000, Transmit at 0. Using 40MHz. RSSI is at -51 and quality is at 47 but I'm literally on my desk right next to it...
does 40MHz get better range then 80MHz ? i would think so but* maybe if the 80MHz extension channels arent being used range will be close to the same?
ive been curious. im currently running 80MHz with no AC devices just cause
when i said 95/150mbit, i mean i got 95mbit of real throughput on a 150mbit link, using speedtest on iphone 5, and i got consistent results
what kinda throughput are you seeing with good signal close to the router?
I run DD-WRT and all the older guru members say the lower the MHz the better as it gives more range. I would think 80MHz is half of 40MHz, or am I wrong? I don't believe I'm wrong but I'm sure someone like Shibby or Toastman or Merlin can help.
Speedtest on my iPhone is limited to 50/5 because that's what my ISP provides.
i have only heard people say 20MHz is better then 40MHz on 2.4GHz. NEVER HAVE i heard anyone mention anything about htis on 5GHz...........
that sucks u only get 50/5 i get 107/5.5 so i can max out iphone 5 link and still have breathing room
Well 5GHz obviously has worse range than 2.4 but has better speed, I'm sure the same applies to it, if anything it makes a HUGE difference you would think. Yeah, Time Warner sucks. They just rolled out 8 downstream bonded channels and 4 upstream bonded channels in my area so I expect 100/10 soon, hopefully it comes with 300/20 and 200/20. I'll probably get 200/20 at most because I will never use 300 down and I only want the 20 up
0 is the hardware default. Is it the best? I cannot say.
What I can tell you is that it is a good starting point.
There are MANY factors that affect wireless range and throughput. Simply cranking up the power is NOT the way to fix it. In fact it can may times make it far worse.
Search the forums. There have been several cases were people LOWERED the power and got better throughput and signals.
I just plugged mine in and my phone wouldn't find the Chromecast... I turned off 5ghz wireless on the router and it started working.. Try that next time and see what happens.. I just started mine so I'll see what happens. I'm streaming from Plex as well.
They are referring to the radio frequency, not to the channel width. Higher frequencies are more affected by obstacles, which is why 2.4 GHz has more range than 5 GHz.
i dont believe saying 2.4 has more range actually makes sense, but only that it has better penetration. 5GHz actually has better range as a fact, because the waves are smaller....
2100, 2600 LTE = RANGE,
850,700 LTE = PENETRATION, 850 is almost what every phone in canada uses indoors. atleast with bell where we live, the phone will prefer 850 3G rather then any higher LTE signal in my house, rogers will stay on LTE
rogers LTE = 6-7mbit
bell 3g = 14mbit because there using 850....... better indoor penetration... higher frequency, better range!
think about it, if someone screams, high frequency, if its outdoors u will hear it really far away.
bass penetrates, but it doesnt have as much range....... this is why indoors u can usually only hear the bass when someone is listening to music loud
Semantics. To the eyes of an end user with a router in his house, it still boils down to range.
My explanation remains accurate - I did specify this had to do with penetration, not raw distance.
merlin, im curious how much u think 20MHz vs 40MHz vs 80MHz makes a difference for example on RT-AC66U, assuming we have a device that can support all 3 modes (an AC device) will any width have superior range?
for 5GHz only obviously. its clear 20MHz is superior to 40MHz on 2.4GHz
fair arguement, i just think its clarified 5GHz actually does have more range technically its just that the common use of AP's are indoors making 2.4GHz and its superior penetration appear to have longer range....
That's the channel width, and it shouldn't have any impact on range/coverage. It only increases the likelihood of interference (which is often not a big issue on the 5 GHz band).
Would increasing the Channel Width reduce the likelihood of interference (because the router can choose from more channels) or increase the likelihood?
hehe exactly, i agree with this logic but if you look at the posts by "noobWRT" he was stating obviously since 2.4GHz has better range on smaller channel widths then 5GHz will aswell, when this is simply not the case.
thanks for clarifying.
"I run DD-WRT and all the older guru members say the lower the MHz the better as it gives more range. I would think 80MHz is half of 40MHz, or am I wrong? I don't believe I'm wrong but I'm sure someone like Shibby or Toastman or Merlin can help."
"the lower the MHz" the better as it gives more range" is not true.
Thanks Shibby. But what is the point of having the latest for E1000 if we can't install it cause it doesn't fit?
This has been discussed before, 104 is the last version that will fit
TVLZ makes a mod that does fit.
See post #112 on this page: http://www.linksysinfo.org/index.ph...for-linksys-e1000-v2-1-will-help.35630/page-2
I have the 20 Mini build on my secondary E1000 v2.1 right now.
Hello shibby... it appears webmonitor is working only a little bit on R7000, but not fully. It is not logging as much as my e900 and it seems to crash after a while and stop monitoring altogether until a reboot. my last reboot it log one thing "Mon Jul 14 2014, 1:02:50 AM liveupdate.symantecliveupdate.com" and then nothing more after 3 hours of surfing web.
Thanks. I'll try TVLS's mod. I hope it supports MultiSSID and QOS.
Shibby, would you please elaborate on what fix has been implemented to get QoS running on the ARM platform? I know before we were using the imq kernel module to create the QoS queue's but I can't seem to figure out what mechanism is being used now.
It seems that not all my old QoS rules are being matched with the new config and I would like to investigate further. I also want to create a new QoS queue for VPN traffic.
QoS outbound only was fixed. Inbound isn`t (imq module needed). I`m still "fighting" with it.
Shibby, thank you for your quick response. I'm reading up on the HTB queuing discipline to see what went wrong with my configuration.
Can I assist you in any way on getting the imq working?
Increase. You are using a larger portion of the spectrum, so you can potentially overlap with a lot more things.
I didn't see it in the changelog; was SNMP re-included for ARM devices!
You may be confusing channel width and channel selection. Wider "channel width" means a single signal spanning several channels in the wireless spectrum licensing sense of the term. A device listening to an 80MHz wide part of the spectrum will pick up more noise than one using 20MHz. More noise can mean unreliable connection, which one might perceive as loss of usable range.
Shibby, I'm glad you're so active here.
I've got a problem with IPv6 DNS though, I would love if you could take a look at it. So after some trouble, I got IPv6 working, and everything works great, with DNS servers entered at Basic->IPv6.
But as soon as I start the PPTP VPN, the IPv6 dns-nameserver entries are removed, and only the IPv4 DNS nameservers appear in the log. (The IPv6 connection still works after starting the VPN via direct address, but the dns-lookup fails)
More information, logs etc here: http://tomatousb.org/forum/t-931736/bug-dnsmasq-dhcpv6-not-working-with-native-ipv6
If anybody has an idea, i would be thankful.
There is some erroneous information being posted here.
The range of radio frequencies is determined by the attenuation with distance, the path loss for signals of the same power level vary with frequency as shown in the graph here. For other frequencies, you can easily interpolate.
Looking at the graph, you can see that 2.4GHz will give you approximately 6dB+ better signal than 5GHz. Forget about what your RSSI indicator says, it isn't accurate. The loss in a typical "real-life" scenario at 5GHz is typically worse than this, due to attenuation of the medium, moisture, the higher scattering effect at higher frequencies, and more severe multipath problems. For in-depth discussions please refer to the links below.
2.4GHz also penetrates solid materials and especially buildings better than 5GHz. Much better. That is why there is so much interference on that band these days. 5GHz is pretty bad at penetrating brick, concrete, etc. Therefore you can use 5GHz deliberately to restrict range and localize the signal to within one room. It is also why using 2100MHz v 850MHz allows cellular operators to put in approximately three times the number of base stations in a given area, and thus increase the number of simultaneous calls.
The question of what range 20MHz and 40MHz the bandwidth setting gives - this is absolutely nothing to do with the radio frequency in use, we are talking here about the bandwidth. And for the purposes of this article, we'll say they are pretty similar when there is no co-channel interference.
also of interest:
Thank you all, I didn't mean to cause confusion. I was just stating what I thought. Once again, I have stated that I'm not a Guru and just restating what I heard and saying "I THINK" multiple times. Just trying to figure out what will give me more range between floors of my home. If I go downstairs to the other corner of the home i'll have internet connection but it's painfully slow.
What do you mean fighting it?
With working imq module.
I checked shibby's source code in repocz and found some missing in basic-network.asp file . I think shibby will repair it next version.
There is a small bug that exists for a while, Asus RT-N12B1 running your mod (I use RT-N5X MIPSR2 MAX), LAN1-4 are in wrong order. Anyway, this doesn't matter. If you have some time, please fix it.
I think the real question is, why does shibby keep building them if they cannot be flashed?
@Toastman i asked my friend what would be the best proof to show you
his reply was
"Standard RF theory ?"
The rule is: 2.4Ghz vs 5.0Ghz
If you double frequency signal drop on same distance 6dB, same as half power transmitted.
@Connor, Indeed, if you punch the figures into that calculator that you linked to, the "Free Space Loss" program, you will see that attenuation of 2.4GHz over 1km is 100.026dB and that of 5GHz is 106.401dB, approximately 6dB worse. So your own link confirms everything I have said.
Listening to "knowledgeable friends" is not a good idea sometimes.
Now, I think that this is shibby's thread, and it would be courtesy to him not to fill it with anything further on this subject. I have nothing more to add to it. It is particularly irritating when some know-all states "standard RF theory, when standard RF theory is exactly the opposite of what he is stating.
EDIT: Sorry I'm tired power is 3dB and voltage is 6dB nothing else.
here ya go
i agree but in my defense i did message you privately and you chose to respond on here.......
however i dont disagree you provide a good argument.
it would be nice if someone could explain why cellphone companies choose high frequencies for outdoors and low for indoors if lower has better range.........
FYI Shibby: previously added IPv6 DNS servers (In Basic->IPv6) aren't being removed from dnsmasq's config when disabling IPv6.
Causes the connected devices to receive IPv6 internet addresses even though they can't connect.
To the great reviews I bought the R7000.
Now I'm not sure which version Tomato I should take for this router and how to install it.
What router should do besides its router function is:
tunnel IP4/IP6 to "Hurricane Electric"
I found RAF, but is that the right firmware for me ?
Must I first flash "netgear2tomato RAF" and then "netgear RAF" version?
So 30/30/30 -> flash netgear2tomato -> 30/30/30 -> flash netgeare ?
This is the first time i flash an expensive router ( for me )
Shibby's version should do all that:
Initial flash: http://tomato.groov.pl/download/K26ARM/118/R7000/
Then you can update within GUI to latest: http://tomato.groov.pl/download/K26ARM/121/
Erase NVRAM in between flashes. Good luck and all at your own risk of course.
Never reach my inbox.
I agree, it is confusing for people who don't follow the threads.
I've asked that question before and no answer was given.
I was just pointing out that there is at least an alternative that gets people to a newer version.
Wow. My RT-N10P box somehow does not want to run Tomato. I tried a few more flashes, waited 60 min. No luck. Flashed back to latest Asus firmware, router is back. I tried the Asus utility, I tried tftp via command line. No luck. I wonder if anything else is wrong. Maybe I should try an older version first. This is a MIPS2 router, right?
Here is what I see in the pings now, two machines connected via cable, one PC static, one Mac DHCP:
1) When I replug after the flash, I see 4-6 pings returned (PC) and then everything times out again. No address for the Mac.
2) When I boot with WPS button pressed for 10 sec (erase NVRAM), then I get 4-6 pings returned (PC), then they time out again, then they come back for 4-6 pings (PC), the Mac gets an IP address from the router, and then everything disappears again on both machines and turns to destination host unreachable.
Edit (solved): Let's book this under user error and forget about it. Shame on me. I used the wrong firmware file. I must have ended up in the Exxxx folder and used on of the files that did not say Exxxx but obviously they are still wrong. Went back as puzzled to double check to Shibby's site and took a second look and there it stared right at me, another folder with special RT-N builds. Another flash, router is up and running Tomato. Thanks to the router gods for the Asus recovery mechanism, I would have killed this box multiple times otherwise. ;-)
thx shibby.I have already flash v121(R7000 edition) to my R6300 V2.All functions work well.But we all want a real shibby tomato for R6300 V2.That's would be great!
Is the max bandwidth limit for both upload and download broke in the original asus firmware and merlin build? I cannot get it to limit a devices bandwidth to specified percentage?
Source IP or MAC field - using device mac
destination port - <65535
priority - Lowest
maximum bandwidth for both upload and download limit is set to 10% as a test, not working properly in the slightest. Any ideas? Thanks.
Asus's QoS does not work as a bandwidth limiter. It will only reduce the bandwidth to the specified percentage if you don't have any other traffic with a higher priority. That's why, for example, P2P will run at your full speed, until you start web browsing, at which point P2P traffic will drop to the 5% throughput defined in its class.
Ohh, thanks again for explanation.
I guess that won't help me if people are reaching my internet limit streaming and downloading ect while I'm playing xbox 360... Xbox uses like 6kbps down on some games and it's too small of an amount for QOS to keep it solid no?
Any chance of a bwl being added to your merlin builds? I assume highly unlikely, but yeah... Any chance?
The reason I can't just use the WRT or Tomato firmware is because I play a Diva game in GOW2 (Gears of War 2). Connections were really great at first, but then things fell off terribly (used upnp). Original firmware was more consistent in connection quality for GOW2. I am using your merlin build now, and expect it to be the same with some bug fixes with upnp ect...
But either way, thanks for the work you guys do.
Bandwidth Limitation is a poor way to manage bandwidth, and it's quite wasteful. No reason in not using the full WAN bandwidth when no one else is contending with you for that bandwidth.
Give your XBox's MAC address a higher priority in the QoS rules, and that way it should always overrule torrent traffic coming from other clients on your network.
alright, I input both my xbox's, selected their ip (mac) and tcp/udp as protocol. I left service name and destination port empty because I have multiple xbox consoles; and upnp may want different ports I assume. Selected highest for xbox and set to the highest reserve to 100%.
100 up and down limit for everything.
0% reserve for all other priorities. Now my xbox's are the only rules, I don't have any other rules, is that okay?
Also, I have 6000kbps down and 800kbps up from my isp; prolly normally at 5mbps down and 670kbps up. I tried both 6000kbps/800kbps and 4500kbps/600kbps
But yeah, you think this should stop people from lagging me as they stream and download? Would lag spikes be an issue? Thanks for the help.
Shibby is there any chance to put DNSSEC into K26 MIPSR2 w/o USB MAX and K26-RT MIPSR2 w/o USB MAX? These two image are only 5.8MB and 6.2MB, I don't know if you put DNSSEC module in, they still fit for 8MB flash device.
I am from China and as you may know, our government use DNS cache pollution, and DNSSEC may be very helpful. Thanks.
@ierwin yes, i can do that - in next release.