"to small step 6-8" for example.

What about the situations were profiles set both of those settings to default to "12 px. Average 2" and "By two with global refinement" but I cannot use both? Which one should be my first choice?

oviano,

There are explanatory tooltips if you hover your mouse on each of those settings. However, a more in-depth analysis would be:

"Motion vector grid sets values of three svpflow parameters: block.w, block.h, block.overlap.
The smallest value give more smoothness, more fluidity, but more artifacts.
Advises: Change it to smaller to increase local smoothness. Change it to bigger to decrease artifacts.

Decrease grid step setting manages amount of calls to recalculation function with block size dividing (refine.[thsad1, thsad2, ...]).
Using of it can give more smoothness and decrease waves-artifacts, but makes more visible halo around moved objects.
Advises: Change it to smaller to increase local smoothness. Disable it to decrease halo-effect."

These are the best explanations available online in English, found here: http://www.svp-team.com/forum/viewtopic.php?id=652

1st one
cause it's the default option

I suspect the answer will be to experiment and see what is best for you, but personally I'd like to see a description of what each of these settings means to aid the experimenting process.

So with my current set up and processing power, I can do either:

1.

Motion vectors grid: 12 px. Average 2
Decrease grid step: Disabled

or

2.

Motion vectors grid: 16 px. Average 0
Decrease grid step: By two with global refinement

-

If you could do either, which one would you choose and why?

Thanks in advance!