UBNT / Mimosa / Mikrotik comparision

Hi cambium guys.

What do you think about this ? do you have something like this ?

https://community.ubnt.com/t5/airMAX-Stories/Radio-Shootout-Pt-2-let-s-try-a-whole-bunch-of-them/cnc-p/1232426#M3006

is interesting.

1 Like

eejimm in these  tests simply measured  Adjacent Channel Rejection  of various radios  . ACR is important parameter, but it does not give any estimation of interference resistance  of ePMP , ubnt  and Mikrotik radio. So these tests looks   like  ubnt  «expert»  trick with false conlusions.

Really  these tests showed following :

- ACR  of Atheros  802.11n  and ubnt/mikrotik  АС  radio  at 2.5  adjacent channel - 30 MHz guard interval (GI)  in 20MHz channel bandwidth , roughtly  are  about  15-17 dB. It is not good, but is not so  bad.

- Mimosa  radio ACR  is much worse – less than  12 dB at 30 MHz guard interval. It is  not  good.

- ACR of ubnt AF5X   it seems due to Airprizm filter is about 30 dB at  (GI i >30 MHz). But we see that at GI= 20 MHz  and less  ACR is the same as  other systems  without filter. So airprizm does not works at 1st  and probably at 2nd  adjacent channels, so  it makes this type of filter  almost useless.

In regards to measure of thoughput.

 ePMP aggregate throughput  60-70 mbps without interference ( 70 mHz GI) is not correct. It should have been 110-120Mbps. Something   wrong   in configuration, may be   DL/UL ratio  is 50/50 and  simplex ( one way) speed  is measured .   Also   ePMP works at  64QAM  and  comparison    with  other system at   256QAM  is not correct.  

So these tests  are   usual  throughput  comparison  at  64QAM 5/6 and 256QAM modulation and ACR measurement  various systems and nothing more.

eejimm  should provide tests  with   interference  at  the same frequency channels  (co-channel interference)   with   C/I = 10, 20,30 dB. These results will be more interesting and usefull.

3 Likes

It's the test from a major UBNT fanboy and the tests doesn't relate well at all to real world performance.  He obviously doesn't have the epmp configured correctly.  

In nearly every deployment I have, the epmp has outperformed the Ubiquiti AC equipment.  Pretty impressive since the ubiquiti is AC and epmp in N.   And I have no bias, I just want to use the equipment that is going to work the best and make my life the easiest.

2 Likes

@Vyacheslav wrote:

 ePMP aggregate throughput  60-70 mbps without interference ( 70 mHz GI) is not correct. It should have been 110-120Mbps. Something   wrong   in configuration, may be   DL/UL ratio  is 50/50 and  simplex ( one way) speed  is measured .   


The "ePMP Capacity Planner Guide" produced by Cambium states a maximum total throughput on a 20 Mhz channel is 71.4 Mbps. Is that incorrect?

On a 20Mhz channel with 75/25 split I'm able to get 85Mbps down and ~20Mbps up consistently.

1 Like

The truth is in the field. They do not measure equipment failure and what happens when there is a problem with their firmware. At sometime s.th. does not work and you're messed up. There is still no ATPC with UBNT. This is a regulations requirement in ETSI so all these fanboys operate illegal.

Nevertheless ePMP is (only) 11n based so we do net get the higher modulation rates with good conditions. Even if this gives us only a 20-30 percent more bandwidth to some users this is something we want. We need high speedtest values as this is the best marketing.

I would like to see a statement that there is an upgrade path anytime soon.

1 Like

i agree ste, 

here in my area....there is a company that only uses airfiber for their backhauls....(they do 2 links for the same ptp, one is in production...the second ..backup..)....i hope ETSI takes care of this problem... my area is getting impossible to operate... i have order a pmp450 because of this... hope it works

regards

such update on PMP450 behaviour ?

Try ePMP2000 it realy works great.

PMP450i is the best there is, we use it for dedicated custumers...

1 Like

We use all the AF products except the 4x.  Here are some observations from the field.

The AF24 is impressive. And consumes 51 watts.  If it rains, link will fail.

The AF5 iworks well but it does degrade with any Cambium product nearby.  Affects all Cambium products as well.

The AF11FX is better than the Mimosa B11 in terms of latency but not throughput.  It is more spectrally efficient. It is not comparable to the 820c or s in any way other than they require an 11Ghz license.

AF2, 3, and 5x radios are comical.  ePMP stomps on the AF2x and AF5x in moderate to high interferance environments.  Where the 2x and 5x are barely operational (<10Mbps), ePMP will come through with MCS14 and MCS15.  Non-GPS model as well but you are constrained by the FE interfaces. Seal the Cat5 caps on the AFx with something or rain will blow up into the radio and it will fail.  They are cheesy.  NxN?  Good luck with that.

AF3x offers a 40Mhx channel width in NA.  So it is usefull to near 200Mbps FD currently if everything is perfect.

Other considerations are PPS limits and how PPS affects latency.  This is where Mimosa falls apart.  AF11FX too.  500Mbps indicated capacity = 400Mbps throughput at 40k PPS.  Pathetic.  820s sub ms RTT ramps to 8ms on B11.

Lack of SFP cage is going to be an issue for Ubiquiti BH with fs.com selling BiDi lasers for $9.

Were MikroTik to pick up the code from Bitlomat (Fluidmesh with Nokia engineers), they'd walk on water.  No halfduplex radio beat Bitlomat in terms of throughput, multipath or interference robustness.

Conclusion - 

For economical PtMP, ePMP is the only choice.  For many reasons.

4 Likes

Can you give me more information on Bitlomat ?