Interesting. I've heard that too but I don't see a picture of that round HSS/carbide blank configuration on their website. Does the holder have a Vee notch & the rod gets clamped in there like the square bit? Happen to have a pic of that?
https://www.eccentricengineering.com.au/index.php?option=com_content&view=featured&Itemid=101
Also, just being picky - wouldn't the projected round section cut an ellipse due to the rod's back angle clearance & probably exaggerated a bit more with any positive rake on the upper surface?
The issue with any of these 'hold the cutter' type ideas (including mine) is that the cutting profile may well have penetrate half its diameter which means the supporting tool holder doesn't have very much meat to grab onto or support without interfering with the cut profile itself. Maybe one could braze the button on like a conventional turning tool, but now its getting even more complicated. A tool & cutter grinder would make an accurate radius shape with all the requisite relief geometry directly out of HSS blank, but not everyone has such a machine.
For larger grooves the weapon of choice is a Holdridge Radii principle, but I think the smallest arc is maybe .75" or so
Not sure, but I don't think it would cut an ellipse because the OD of the rod is still a circle (looking down along it's axis). Also, the tool is presented
to the work at an angle which cancels out some of the angle ground into the tool.
The way I see it, if the circular section cutting tool rod is positioned at an angle in the toolholder like post#29 pic to provide back clearance & fed into the work on the spindle rotation axis, that is equivalent to cutting a section across a cylinder & looking down on it from that cut plane...which yields an ellipse. This shows an example 15-deg angle & resultant effect on dimensions. To help visualize consider if the tool angle was greatly exaggerated like at 45 deg and now cut across the horizontal plane. The width would always equal the tool diameter but the profile would not be semi-circular, it grows into a longer aspect ratio elipse. And I suspect positive rake exaggerates this even more. Whether this is dimensionally significant or not is up to the machinist. Probably not on small diameter grooves. I'm just pointing out the geometry.
I certainly don't have that kind of experience in the trade but I do have an interest in HSS tool grinding. It has been sort of a pet focus for me for the last 30 years but I grant that I am not in a production environment. Still, I've ground a bunch of form tools over the years and have found that 12 degrees works well on a round nose tool like this (as opposed to the 8-10 degrees that is normally recommended). If the nose radius was narrower, I would increase the relief angles to 15 degrees to lower cutting forces even more to reduce chatter potential and improve finishes.
As I said, a matter of opinion.