what happen if in a point-point mimo 2x2 I connect at one side the vertical channel to chain 0 and horizontal channel to chain 1, and at the other side I connect the horizontal channel to chain 0 and the vertical channel to chain 1.
could the link works normally or it is a mistake ?
thanks
If you use N mode and have both tx/rx chains enabled then it should work ok.
ya, this correct as we have a lot of the back-haul is running on this setup.
but, you have to very careful on the wireless card selection, and N will get more interference than A_only.
thanks.
angboontiong wich card do you advice?
i use r52N
hi..
all my wireless card is from the dbii.
so far, i have no factory defect card from them, i have used more than 300 card from them (F520-Pro most).
for N, we use F52N-pro, so far, this card still working well.
but, remember that, the N will easy to get more interference, and i would advice you go for RB800 instead of RB4xx AH as it’s wasted your link.
why does rb 433ah wasted link??
you can see..
when in the real internet traffic flow.
the RB433Ah max just can carry 40Mbps (CPU utilization up to 40%).
when it reach this level, the will be high latency.
but if you using RB800, then you will having no issue to go up to 90Mbps.
WELL not joking but i am really amazed to see the talk for 40 mbps and 90 mbps.
In real terms i have never scored over 15/16mbps even for a 2km link and point to point and the best available gears, antenna and cables
i believe your result is with the nstreme instead of the nv2 with RB433Ah.
if with the RB433Ah and nv2, you should able to go up to 17~18Mbps.
with 40Mhz channel bandwidth, you should able to get 35Mbps full duplex in the TCP band-test mode only.
p/s: about is in the A-only mode.
if possible, try go for RB800 with the antenna (www.lanbowan.com), then you can get the good result.
I really feel the need to correct you here. I run a 9km link with 2x rb433AH with 802.11a/n and NV2 and 40Mhz channel and I run easy 40Mbps peak and 37 average over it (tcp, one way bandwidth test with some 1Mb return traffic from live network) and the conn rate is 108Mbps/108Mbps. (I have some Fresnel issue. Antenna needs to be elevated more.)
Both rb433AH’s cpu are showing around 12% (10-18% fluctuating).
I have some other 2km link with sort of same setup that I could push up to 80Mb real tcp.
So if your boards are running into high cpu usage with your low data throughputs it must be you have these card performing several processes on the data. Like firewall rules, mangle. QoS etc. and maybe lots of routing tables?
My only unit that has an rb800 is an AP with almost 40 connected CPE’s in ´a´ mode and NV2 enabled. Each customer is allowed to run 4Mb continuous (which never happens) and I can saturate my 20Mb symmetric capacity easy from this AP and it still shows only some 20-30% cpu peak usages.
So I don’t know about your setup but your figures are really disappointing…
http://routerboard.com/RB433AH will tell you that basically the throughput of the boards is more limited by the radio conn rate (max tcp traffic roughly 50% of conn. rate) than by the board.
So using rb800 for a link doing only 20Mb at times is a sheer waste of money… You buy 2 or 3 rb433’s for the same money.
Hi Rudy,
It seems you have configured Tx power mode selected manual. How come signal strength with only 12mb HT-20-3 and HT-40-3. which mini pci r u using
Hi bro..
when i mean full duplex, which mean the traffic is in (tx + rx concurrently), which mean, it match which what you shown in the print screen.
you shown is only on the half duplex mode btw.
Tx power mode is on default, H-power cards (R52Hn)
I only set the conn. rates to manual. So you can force the cards to ´communicate´ on the rates that works most stable.
For full duplex it is needed that radio can both send and receive at the same time. In 802.11 cards we normally use what is known as “semi-duplex”. In general when using only one card we always speak about semi-duplex. (Radio switches between send and receive very fast)
“Half duplex” is something that doesn’t exist. I think you mean “Simplex” (Only one way traffic)
And as you can see, while running a speedtest in one direction my link was also transporting more than 1Mb in the other direction. So nothing like “Simplex” Just a full dual way communication where in one direction much more traffic is transported than in the opposite direction.
If you state a link can only carry 17Mb than it is not clear for the reader it means in both directions at the same time.
And this is also not representing most real life scenario. Traffic stream is usually bigger towards client than from it. Special on network with many clients connected it would make it a rare event when uploads are same or bigger as downloads at the same time.