link=topic=101682.msg843983#msg843983 date=1439322325]
http://www.sns-hdr.com/forum/viewtopic.php?f=7&t=4138
"The gradients are less precise when the step is large. I do not recommend step greater than 2ev."
Sebastian, SNS-HDR's author, is surely right just by adding "when using SNS-HDR" to his claim. This also happens with Photomatix and other blending algorithms like Enfuse's pyramids. But this is a fault of those pieces of software, which are sub-optimum at the blending stage doing quite extensive progressive blending, it is not a general truth about bracketing fusion.
Sub-optimum progressive blending algorithms have two consequences: they fail to get the best available information from the input RAW files so they need more shots for the same output blending quality, and they loose sharpness when a minimum missalignment between shots persists.
With modern sensors (and not so modern as my Canon 350D's), gradients are fine at 4 stops intervals, although I prefer not to go that far and bracket at 3 stops intervals.
Regards