I have proven that edge width is a main component in possible sharpness. I also know that assumptions based on theory and what happens with practical application do not always line up. I have reduced bevel angle on too many blades to count. Reducing the angle of the bevel also reduces the edge width. In general, overly-stout beveled razors don’t shave as well as they do with more acute bevel angles. A reduction in edge-width can result in sharper edges sometimes.
This is not an absolute 100% of the time thing. .
But, in general - that’s how the cookie crumbles.
It’s not a matter of taking 1 degree off a bevel angle. And, honing skills and steel capabilities factor in. Soft steels will not take killer edges, bad carbide structures, poorly formed pearlite bands, etc - all contribute to impeding the possible edge creation and edge retention qualities. Same for embryonic honing skills, low-grade stones, etc.
Where edge width vs cutting efficacy is concerned, one has to make a MAJOR reduction in that physical dimension before any sort of real difference in sharpness becomes possible.
I have reduced a 20˚ bevel to 19˚ without gaining any sharpness. I found that when starting at 20˚, I needed to drop to 17˚ before any change in cutting efficacy was possible. And, reducing the bevel angle by a full 3 degrees did not produce a physical indication of any change in bevel flexibility, not by handling and not by shaving.
Is it possible to create a honing surface that reduces edge-width to a greater degree than reducing the bevel angle by 3 degrees is capable of?
The math - starting with a 17˚ bevel, the edge-width 3 µm back from the apex (peak of vertex angle of acute isosceles triangle) is 0.89 µm. If I reduce that bevel angle to 14 degrees the edge width is 0.73 µm. That is a change of 0.1 µm.
I have done this many times with a variety of blades. After the work, it is impossible to tell if the bevel is more flexible; not by handling it or by shaving with it. Some steels will not take an edge at 14˚, while some will take an edge but it won’t cut better than it did at 17˚. Some blades will show a tendency to chip. A good number of quality blades will show the ability to get ‘sharper’ though.
If I start with a 20˚ bevel, the edge-width is 1.1 µm. Reducing that angle to 17˚ I now have the edge-width at 0.89 µm. That is a 0.2 µm difference, and that is why fat-bevel blades show significant cutting efficacy improvement after their spines are ground down.
So, I have seen edge performance improve, sometimes, with a geometry change of 0.1 - 0.2 µm. That’s the math I have seen turn actual tangible and documentable results.
I have honed on stones with all sorts of curves ground into them. I spent a lot of time trying to exploit different topologies actually. Truthfully, I have not seen a convex stone that imparted significant changes in edge-width or flexibility. Concave stones have been a mixed bag of tricks but none of those tricks have proven to be of worth to me when honing razors.
The simplicity of a reliable and repeatable flat surface seems, to me, to be a no brainer and anything else just seems like a waste of time.
The exception being this… if I was a razor manufacturer that was constantly losing money on razors that were warped, then I might want to use curved stones to get them honed. Reason being this…selling a new razor with uneven hone wear isn’t going to work well. Putting a warped blade on a flat stone is going to bring out some sort of visible indicator of the badly executed geometry.
Having a stone that allowed me to navigate the warp in a way that leaves typical hone wear on the spine would probably provide the means to recover what would, otherwise, be a financial loss.
