Hi!

I use Mikrotik hardware for indoor WLAN networks. What I am missing here most is radiation patterns (https://en.wikipedia.org/wiki/Radiation_pattern) for each device.

That makes me sometimes really problems choosing right model. For example the cap models seem to not go too much to the side. so mounting them on the ceiling will not cover adjacent rooms.

Mikrotik please rectify this very important point in your documentation!

All indoor models have around 2-3 dBi which isn't too far from ideal sphere. Due to that it does not matter what device you put where, because walls will attenuate the signal anyway and within the room/hallway it will reflect no matter how you place the device.

To put some salt to injury: theoretical ideal omni-directional antenna (which would have to be electrical monopole and hence doesn't exist) would have gain of 0 dBi (because that's definition of antenna gain). A basic half-wavelength dipole antenna with toroidal radiation pattern has gain of 2.15 dBi, but this means it has low signal strength along the dipole axis. All antennae with gain higher than 0 dBm have lack of signal in some direction, gain higher than 0 dBi means that radiated energy is concentrated in certain directions and is missing in other directions (e.g. dish antennae which are one of highest.gain antennae focus energy in narrow beam in dish centre line).

Unfortunately low antenna gain doesn't mean it's a good omni-directional antenna ...it can as well be a bad directional antenna (e.g. a dipole with lenght other than a few multiples of wavelength is omnidirectional, but gain can be lower than 0 dBi). What I'm saying here is that without some charts (directional charts, frequency charts) we don't really know how omni-directional is a 2dBi-antenna. And whether the information about gain actually applies to whole frequency band used by device (even proffesional antennae have varying gain accross certain frequency band, e.g. a 12 dBi antenna can have gain as low as 9 dBi within operating frequency band).