Netonix Forums

I read this one: http://muniwireless.com/2016/03/21/galactus-vs-guerilla-wireless-boy-grown-up/Is Mimosa really performing that much better than UBNT ?

There will be no answer as this gear is not available. Only beta testers use it. This story is made by the only beta tester I have heard of and might be not that neutral ...

Rory is well known and respected... I wouldn't call him out as biased...

His numbers are...well...to high to be a 'fluke'... If you got 50% of those, it would be a significant improvement.

rebelwireless wrote:Rory is well known and respected... I wouldn't call him out as biased...

His numbers are...well...to high to be a 'fluke'... If you got 50% of those, it would be a significant improvement.

Then call it "euphoric". We'll see if SW has its edges as it had with B5. The numbers are normal for AC gear using 80MHz channels.
Beta testers who are elected by the vendor tend to be euphoric.

ste wrote:

rebelwireless wrote:Rory is well known and respected... I wouldn't call him out as biased...

His numbers are...well...to high to be a 'fluke'... If you got 50% of those, it would be a significant improvement.

Then call it "euphoric". We'll see if SW has its edges as it had with B5. The numbers are normal for AC gear using 80MHz channels.
Beta testers who are elected by the vendor tend to be euphoric.

Hey, not all of us, when I was a superuser over at UBNT I was very BIAS and pointed out the good and the BAD. If they did something I liked (which was most of time) I was the Banana waving the flag but when they did something I hated I spoke very loudly about it.

sirhc wrote:

ste wrote:

rebelwireless wrote:Rory is well known and respected... I wouldn't call him out as biased...

His numbers are...well...to high to be a 'fluke'... If you got 50% of those, it would be a significant improvement.

Then call it "euphoric". We'll see if SW has its edges as it had with B5. The numbers are normal for AC gear using 80MHz channels.
Beta testers who are elected by the vendor tend to be euphoric.

Hey, not all of us, when I was a superuser over at UBNT I was very BIAS and pointed out the good and the BAD. If they did something I liked (which was most of time) I was the Banana waving the flag but when they did something I hated I spoke very loudly about it.

I did not say there is bad bahavior. The vendor selects testers. So he would not select the big critics. And the beta testers get something without paying so there is a feeling of obligation. So a beta tester will see the chances of a gear and is not pissed off when he had to reboot on a daily basis to keep things going in beta phase.
The same discussion is with customer tests on amazon where e.g. samsung gives away some stuff for free. This tests are more positive than the average tests.

I am often surprised buying gear soon after appearance which does not meet the enthusiastic beta stories. e.g. with B5 this praised autoeverything feature starts working with FW 1.3.1 for me. All releases before very bad channel selections where made. This is a main feature which makes a difference between this and other gear.

So I guess A5 will need it's time to mature as B5 did.

I don't think the beta A5/C5 gear even has TDMA enabled, pure 802.11ac.

The numbers about improved SNR is what I was referring to, not 80Mhz channels. You can see a similar thing in Cambium. The numbers on output power and antenna gain on ubiquiti don't compare directly with Cambium, a -5dBm disadvantage per published numbers ends up being the exact same reported signal levels once deployed, and then you get a 5-10dBm improvement in SNR with identical signal levels. If Mimosa gear ends up being similar to cambium in this respect, it will speak very poorly to ubiquiti's hardware...

I don't get this:

MASSIVE SECTOR CAPACITY
A5c is the first outdoor 4x4:4 solution leveraging advanced antenna beamforming and Multi-User MIMO technology. With 4 integrated collaborative antennas, the A5c creates separate "beamforms" to target each client uniquely. Meanwhile a second unique beamform can target a second client simultaneously, getting twice the performance out of the channel.
What does the 4x4:4 actually mean ? It it just that one AP able to handle more antennas ?

For those APs that can be mounted on an antenna it means that you need to have a special antenna sector type that have 4 or 8 inputs (depending on AP Model).
But i wonder about the performance (loose or gain) if you use those APs with 2x120 degree or 4x90 degree normal Sectors.

mu-mimo and 8 spatial streams, and a lot of fancy code on a very high end 802.11ac radio.. It's nothing all that special on the surface, but obviously is a very difficult thing to actually pull off. This is what all the hype about mu-mimo really is, being able to transmit 2 or more streams to multiple clients simultaneously. In a so called 4x4:4 configuration, 4 clients can be transmitted to at the same time with beam-stearing and some azimuth isolation (it's 4x dual polarity sectors)

What kills me is that ubiquiti's current solution, and only solution for the forseable future, is JUST an 802.11ac radio with 2 channels, wave1 ac , absolutely nothing remarkable other than higher modulations and more cpu cycles :/ You might argue that there are other improvements, but they are there ONLY because they re-wrote airmax.