If the floor is wobbly, springs can work, if the speaker is on solid ground, 3 spikes is preferred.If one would consider a solid reinforced concrete floor to be "not wobbly" it has been my personal experience that sprung isolation works significantly better than on a so called "wobbly floor".
The main reason I concluded was that instead of the floor being excited by the energies created in a loudspeaker enclosure deforming, that the higher percentage of energies are deforming the springs. The floor no matter what it is, has a resonant frequency, however the more it can withstand deformation the higher the percentage the spring will be required to deform in order to effectively hold the loudspeaker as motionless as possible.
The concept that energy is stored in the springs used to uphold a loudspeaker is not entirely accurate, it deforms and returns to status quo with the load, and will move above and below the status quo holding the mass placed upon it. The ideal is to have an engineered spring loading for the mass it’s to isolate a broad range of frequencies, all within the audible range.
Such an engineering precedence is used in correctly assembling cars with different rated springs, depending on the result desired. One spring does not suit all applications, and so, it is the same with different loudspeakers.
I will suggest that the more rigid the car structure, the body, the more effective the suspension system can be engineered to support it. The results are much more predictable.
Last time I checked, I was not driving my speakers down a bumpy roadyou have it upside down, your loudspeakers ARE the bumpy road!