Intel and AMD going to the mattresses, and the crowd sheers

[Julio Jerez]

http://www.businessinsider.com/amd-intel-release-32-core-cpu-2018-6
this mean that very soon 16 cores will become the main stream CPU just like 4 cores are now entry level. Now a fancy toaster come with a quad core CPU.

If the CPU war make reasonable high core count the main stream, then my prediction that we do not need GPU for physic simulation is becoming true, even if it was much later than I anticipated.

This is one for the reason I resume work on the parallel solver. with high core count we can relax solver quality in favor of more iteration and have a unified solver that can do everything.
Rigid bodies, cloth, particles, soft bodies, and even fluids.
That is my goal and it seems it mat be possible.

on another line of news, it appears Apple gave a death blow to OpenCL.
https://apple.slashdot.org/story/18/06/05/1719205/apple-deprecates-opengl-and-opencl-in-macos-1014-Mojave
I am now glad I did not waste my time on that.

[JoeJ]

If the CPU war make reasonable high core count the main stream, then my prediction that we do not need GPU for physic simulation is becoming true, even if it was much later than I anticipated.

I like this. It will take a decade until many core can be minimum spec, but until then just do more with less
So, i'm fine with 'no GPU support' with Newton. Good decision i think.

(But i want softbodies, cloth and particles nevertheless, )

[Leadwerks]

This might be a good time for you to contact both of those company's developer relations managers.