Mark Sharp vs Edge Crease?

I’ve seen it said that selecting edges and using Edge Crease to increase (no pun intended) the crease factor, then checking Use Creases in the Subdivision Surface modifier, is a superior method to using Mark Sharp on those edges and using an Edge Split modifier with Sharp Edges checked before the Subdivision Surface modifier. But when I tried it, I got (IMHO) better results with Mark Sharp.

screenshot.2021-09-06 15.47.35
EdgeCreaseTest001.blend (1.4 MB)

In the pic/file above the purple object is using Mark Sharp and doing exactly what I’m looking for, while the green object has Edge Crease set to 0.5 (also used more and less, didn’t help) and is distorted at the 6-poles.

Of course, 6-poles are bad topology, and I’d’ve done better there if I could’ve. Is Edge Crease really a superior method so long as it’s on good topology? Or is this just an opinion?

May be this video will answer your question. It affect the shading modifiers and topology a little bit differently.

Appreciate the thought, but I’m on the autism spectrum and have verbal processing issues – I found it difficult to follow, even the subtitle transcript didn’t help much.

Dunno if it’s an autism thing exactly. The guy in the video does mumble a lot, and talks very quickly.

You answered yourself already. If it works in given situation, it’s right for you.

Don’t follow. Adapt learning process to yourself.
First, don’t run DCC app while watching. Simply relax, observe to see and get the basic idea.
Then lay down foundations to understand “Idea”: Play - Pause - Write it down.
Finally, start up Blender & blend: practice, experience, decide on your values, judge in retrospect then optimize for self. Grow.

Alike fortune, it all comes down to personal preference, beiing able at given moment about what you really care about. While when working for an employee, it’s done when it’s good enough, so both ways would pass - yet the faster, the better.

Have to say, that his verbal explanation is hard to understand.

For my knowledge, I almost dont use hard edges, it affect the normals where you need them to be sharp as on your example. The geometry is ok but the edge doesnt look as sharp as it should be, you used hard edges and it looks cool.
Crease affect the subdivision modifier, so it actually can alter the geometry and you can choose how much it can affect it.

Thanks guys, appreciate the variations on “do what works for you”.

screenshot.2021-09-08 11.37.07
PoleFixCreaseTest002.blend (1.5 MB)

In this pic/file I’ve changed the 6-poles to 5-poles without messing up the rest of the topology too badly (ideally those poles shouldn’t be in the creases), and that’s improved the distortion a bit, but the edges Marked Sharp with the Edge Split modifier still look best here IMHO.

I’m trying to make a couple of material test objects using a shape similar to the Blender logo, so I’d like to use current best practice here if feasable. But I don’t have a decent understanding of what (and why) best practices are for this, nor do I seem to be implementing them effectively.