Irrlicht Engine

You need a custom vertex shader (and tessellated/dense geometry, because paraboloid projection is non-linear and properly computed the triangle edges should "curve"), stefbuet had a post about it on the forum.

The projection is not linear so its neither Perspective or Orthographic.

Shadows are raytraced, one algorithm for everything, area light, skylight, line light, point light, etc. no RTX needed, just OpenCL, runs on everything IrrBaW always ran (OpenCL 1.2 + OpenGL 4.4) even Intel HD Graphics. HD Graphics 620 and Nvidia 1050 Mobile give the same perf, simple scenes are ins...

Raytracing, real-time. https://scontent.fzrh1-1.fna.fbcdn.net/v/t1.0-9/79275442_1439673522853387_3541940837589975040_n.png?_nc_cat=108&_nc_eui2=AeEFdkmeWW9UBKVnr8J38rK-t7Oz2X0Yv6qG4texil4S116nh0KvjyA3Iv9U0IWPPmZ0o69v9BVzVG6p_vGab2qatx167cbL60A4MKhwr3j1uw&_nc_oc=AQnSfWgCie2vk4bQQAt5T0LJXGikVR...

I'd be down for it, I have a Open Source fork of Irrlicht that does quasi-real-time raytracing

We have a raytracing branch now

Should be correct if mul(matrix, vector) means a transpose of front multiply (since A^T.B^T = (AB)^T )

Here's a modelviewer and PBR demo from IrrlichtBAW https://drive.google.com/file/u/1/d/1x_RKIehnaztkfJHY2SpVBN0rK0KhE_Wj/view?usp=sharing&fbclid=IwAR1rRSRym3HvnzU6yl8o86LCvR1Y4GfH5JzrrsxvHImneccTwdxwvhNRHM4 Load your model into that (it will only display the first meshbuffer), see what performan...

Its because now you're switching between 6 different textures and before you only had 1.

Most reasonably optimized games keep the number of meshes < 2000

You should merge all your mesh data that uses the same texture together, if you can use a texture atlas then you can have all data in one mesh.

Working with a new person to bring IrrlichtBaW to Qt5

Vulkan Discord:
https://discord.gg/e26rduK

Graphics Programming Discord:
https://discord.gg/NCsppvc

There are some more links to gamedev discords too

PBR Book
http://www.pbr-book.org/

Have Fun!

Final Result with all Texture Maps that were Available https://scontent-waw1-1.xx.fbcdn.net/v/t1.0-9/62561264_1294836000670474_1465247918599438336_o.jpg?_nc_cat=108&_nc_oc=AQnsR-zITOun2_43a93D-DMpX04We_dzdapjNFrHx_xiccYeUlkGBKJZShscBBxlwRc&_nc_ht=scontent-waw1-1.xx&oh=40eb3c52deb0bf716aa...

Anisotropic GGX with Height Correlated Smith https://scontent-waw1-1.xx.fbcdn.net/v/t1.0-9/65644304_1294788130675261_6414625727789924352_o.png?_nc_cat=105&_nc_oc=AQnnKjfhba9Rj4qeHVWQVuNJ_DDXuGMg_Lns5-vluppI7tbeAk-vzfETFIpLpjVgVZw&_nc_ht=scontent-waw1-1.xx&oh=e3a0dc8f95a5ddcab9dc775999988...

Discord official Server:
https://discord.gg/4MTCVaN

BRDF Explorer (viewer) 80% done https://scontent-waw1-1.xx.fbcdn.net/v/t1.0-9/65052360_1289937344493673_3054651625055453184_o.png?_nc_cat=110&_nc_oc=AQlKAE3aXng3JEPtUve707VS2ytBES6QuNoLX-kEW0OOZdxOMOmu-5pCjfKx4aDEF70&_nc_ht=scontent-waw1-1.xx&oh=4b35521fb15976d546c771f50bf44f64&oe=5D...

Tried Derivative Mapping, Made a mistake.. this is what INFs and NaNs look like. https://scontent-waw1-1.xx.fbcdn.net/v/t1.0-9/65195262_1289414167879324_6743392593691803648_o.png?_nc_cat=102&_nc_oc=AQk45dD85UFrCrlJ-Q32oF3a5bvSqNLeHYedolubXx8cvwc1uDqtkk4OlmQsKRrk5JU&_nc_ht=scontent-waw1-1.xx&...