My personal experiences:
Being forced away from win7 to adopt win10, because of Oculus and MS no longer supporting win7.
Edge is then very much favoured and pushed by MS, and the latest version is chromium based, and not an upgrade to IE/Old Edge. I am actually more willing to use Edge now than before, but Chrome is my browser of choice, on all OS’s
Old Edge had some serious problems with drag-drop and touch, exhibiting different behaviour from chrome and firefox, and breaking one of the projects I am working on. New Edge behaves exactly like Chrome.
So, as both a user and a developer, I will be supporting/using chrome and new Edge, and if I build something that doesn’t work in old edge/ie, i will just direct users to upgrade to latest edge. It feels very much like best of both worlds.
I know there are still a small core of legacy users who, often because of company policy have to use IE and can’t upgrade, I will be interested to see stats once Edge Chromium has been out for a year.
For my product configurator (https://designer.ormsbyguitars.com/), for the last 12 months, 36K sessions:
- IE traffic is 1.7%
- Edge is 2.1%
- Firefox is 6.5%.
- Safari is 22 %
- Chrome is 60%
So it probably depends what your product configurator is for. If it’s for office equipment you might want some IE support, otherwise probably not.
For office machines it’s rather necessary, as i mentioned some companies don’t allow installing other browsers for various reasons, as well as most users of non-tech companies either not being able or caring about installing another browser.
Especially any products relevant for b2b, viewers for presentation like architecture etc. At least the minimum IE version raised with the preinstalled browser version.
My rating:
MS Edge Chrome 
MS Edge 
IE 11 
IE 10 
IE 9 
IE 8 
This could possibly turn into an unnecessary discussion about IE in general - I’m super aware of the enterprise environment and how it runs, however choice of browser support is always up to the project, because they can actively choose not to have IE (etc) users succeed at using your product.
Having the conviction to have some (or all) of your users fail in their first attempt (i.e. you show them a message telling them which browser they should use and a link to further help) is always the right move in the end because it means actual technical progress for the entire organisation, it’s just potentially painful for everyone and takes some bravery. You know this yourself when you’re at the bank arranging a new mortgage and they fire up a remote desktop into Windows 2003 to do your loan application - they (the bank) lack the commitment necessary to progress.
IE10 and everything before is no longer supported in any way by Microsoft from the end of this month, so don’t support it unless you’re specifically being paid to. IE11 will be supported by MS for security fixes while Windows 10 is around, but it’s hugely unlikely to get any new or improved WebGL features.
If you actively support a legacy product you are actively extending it’s useful life. That’s a choice you make, good or bad.
[Edit: in case I haven’t been clear, three.js should not be expected to work in any version older than IE 11, and it’s probably unfair to even hold them to that.]
