The test results using bandwidth test in Mbps can be found here: http://www.mikrotik.com/download/rb112test.pdf
Could you post the link to the old 532 test? I can’t find it.
Yes, Please do!
It would be nice if the same test was done on a 533Mhz VIA, 1.2Ghz PentiumM and a 3Ghz P4.
I’m about to build a ptp link with a CeleronM 1.5Ghz CPU and I sure hope I’m not pissing away the money on this if it doesn’t go any faster.
I wanted to say that there where 532a tests around, but I lost the link.
P.S. Can you share the confs, soo if we make tests they will be comparealbe.
P.P.S. I found it: http://www.mikrotik.com/Documentation/wireless_test.pdf , but the test is with 2.9rc7 
.
Does anyone know why the performance is better with Framing Policy equal to None vs Framing Policy set to Dynamic Size = 3200 bytes? What is the recommended Nstreme configuration for PtP applications? And for PtMP?
Dave
Probably because it’s CPU limited. One setting uses more CPU cycle than the other.
Would be nice to hear from other users which the “preferred” nstreme framer setting is -
Regards
It depends on many things, on average packet size, packets per second number, e.t.c.
You must try each policy and then, and only then you will be able to make a decision.
I have tested bandwith with FTP download from a local server with two CM9 cards installed on RB112.
The total bandwith of the single card is 18 Mbit.
But if I download at the same time from two clients connected one to first CM9 and one to the second CM9 the total bandwith is about 20 Mbit instead of 36 Mbi.
did you ever make any tests with sr9 and RB 112
No only 2.4G e 5Ghz with CM9 and Senao ETSI