lighting.rin
src/6.pbr/1.1.lighting/lighting.rin · 497 lines
// 1.1 PBR lighting
//
// Physically based rendering replaces Phong's hand-tuned terms with a model of
// how light actually reflects off a rough surface: Cook-Torrance. A material is
// no longer "diffuse colour, specular colour, shininess" but:
//
// albedo the base colour
// metallic 0 for dielectrics (plastic, wood), 1 for metals
// roughness 0 a mirror, 1 completely matte
// ao how much ambient light reaches it, baked in
//
// Energy is conserved: whatever a surface reflects as specular it cannot also
// scatter as diffuse. The lights are given as radiance in real units, far above
// 1, so the result is tone mapped (Reinhard) and gamma encoded on the way out.
//
// The grid runs metallic 0 -> 1 from bottom to top, roughness 0 -> 1 from left
// to right. Tab swaps in a textured material, where every one of those numbers
// comes from a map per texel: rusted iron (LearnOpenGL's 1.2 lighting_textured).
// The normal map is read without stored tangents: the tangent frame is rebuilt
// per pixel from screen-space derivatives.
//
// Tab uniform grid / textured spheres
import "../../../includes/win32.rin"
import "../../../includes/math.rin"
import "../../../includes/image.rin"
import "../../../includes/learn/shader.rin"
import "../../../includes/learn/camera.rin"
import "std/print.rin"
import "std/memops.rin"
// settings
scr_width: i32 = 1280;
scr_height: i32 = 720;
vertex_layout: [3]D3D11_INPUT_ELEMENT_DESC = {
{ .SemanticName = "POSITION", .Format = DXGI_FORMAT.DXGI_FORMAT_R32G32B32_FLOAT, .AlignedByteOffset = 0 },
{ .SemanticName = "NORMAL", .Format = DXGI_FORMAT.DXGI_FORMAT_R32G32B32_FLOAT, .AlignedByteOffset = 12 },
{ .SemanticName = "TEXCOORD", .Format = DXGI_FORMAT.DXGI_FORMAT_R32G32_FLOAT, .AlignedByteOffset = 24 },
};
material_constants: struct = {
albedo: vec3;
metallic: f32;
roughness: f32;
ao: f32;
textured: i32;
pad: f32;
}
lights_constants: struct = {
positions: [4]vec4;
colors: [4]vec4;
}
// Mirrors `cbuffer transforms` in the shaders, register for register.
transforms: struct = {
model: mat4;
normal_matrix: mat4; // inverse transpose of model, for normals
view: mat4;
projection: mat4;
view_position: vec3; // the eye, in world space
pad: f32;
}
// camera
cam: camera = {};
last_cursor: POINT = {};
scroll: f32 = 0.0;
// keys pressed since they were last read, for toggles
keys_pressed: [256]b32 = {};
key_pressed: proc(key: i32) -> b32 = {
if (!keys_pressed[key]) { return 0; }
keys_pressed[key] = 0;
return 1;
}
// timing
delta_time: f32 = 0.0;
last_frame: f32 = 0.0;
perf_frequency: i64 = 0;
perf_start: i64 = 0;
// Seconds since the first call: what glfwGetTime is to LearnOpenGL.
get_time: proc() -> f32 = {
now: LARGE_INTEGER = {};
if (perf_frequency == 0) {
QueryPerformanceFrequency(now&);
perf_frequency = now.QuadPart;
QueryPerformanceCounter(now&);
perf_start = now.QuadPart;
}
QueryPerformanceCounter(now&);
return cast(now.QuadPart - perf_start, f32) / cast(perf_frequency, f32);
}
// win32: every event for the window (keys, mouse, closing) arrives here
// ---------------------------------------------------------------------
window_proc: proc[callconv(CALLBACK)](hwnd: HWND, msg: UINT, wparam: WPARAM, lparam: LPARAM) -> LRESULT = {
if (msg == WM_DESTROY) {
PostQuitMessage(0);
return 0;
}
// Bit 30 of lparam is set when the key was already down: a repeat.
if (msg == WM_KEYDOWN && wparam < 256 && (lparam & 0x40000000) == 0) {
keys_pressed[wparam] = 1;
return 0;
}
if (msg == WM_MOUSEWHEEL) {
// The high word of wparam is the distance turned, 120 per notch.
scroll += cast(cast((wparam shr 16) & 0xffff, i16), f32) / 120.0;
return 0;
}
return DefWindowProcA(hwnd, msg, wparam, lparam);
}
// process all input: ask Windows whether relevant keys are down this frame and react accordingly
// ----------------------------------------------------------------------------------------------
// The top bit of GetAsyncKeyState is set while a key is held, which makes the SHORT negative.
key_down: proc(key: i32) -> b32 = {
return GetAsyncKeyState(key) < 0;
}
process_input: proc(window: HWND) -> void = {
cursor: POINT = {};
GetCursorPos(cursor&);
// GetAsyncKeyState reads the whole keyboard, so ignore it while another
// window has the focus; glfwGetKey only ever sees its own window's keys.
if (GetForegroundWindow() != window) {
last_cursor = cursor;
scroll = 0.0;
return;
}
if (key_down(VK_ESCAPE)) { DestroyWindow(window); }
if (key_down('W')) { camera_process_keyboard(cam&, camera_movement.forward, delta_time); }
if (key_down('S')) { camera_process_keyboard(cam&, camera_movement.backward, delta_time); }
if (key_down('A')) { camera_process_keyboard(cam&, camera_movement.left, delta_time); }
if (key_down('D')) { camera_process_keyboard(cam&, camera_movement.right, delta_time); }
if (key_down(VK_SPACE)) { camera_process_keyboard(cam&, camera_movement.up, delta_time); }
if (key_down(VK_SHIFT)) { camera_process_keyboard(cam&, camera_movement.down, delta_time); }
// Hold the right mouse button to look around.
if (key_down(VK_RBUTTON)) {
// Screen y grows downwards, so moving the mouse up is a negative dy: pitch up.
camera_process_mouse(cam&, cast(cursor.x - last_cursor.x, f32), cast(last_cursor.y - cursor.y, f32));
}
last_cursor = cursor;
camera_process_scroll(cam&, scroll);
scroll = 0.0;
}
// An immutable buffer: filled once at creation, never written by the CPU again.
create_buffer: proc(device: *ID3D11Device, bind: UINT, data: *const void, size: u64) -> *ID3D11Buffer = {
desc: D3D11_BUFFER_DESC = {
.ByteWidth = cast(size, UINT),
.Usage = D3D11_USAGE.D3D11_USAGE_IMMUTABLE,
.BindFlags = bind,
};
init: D3D11_SUBRESOURCE_DATA = { .pSysMem = data };
buffer: *ID3D11Buffer = null;
ID3D11Device_CreateBuffer(device, desc&, init&, buffer&);
return buffer;
}
// A constant buffer the CPU rewrites, usually every frame. Its size must be a
// multiple of 16 bytes, the size of one HLSL register.
create_constant_buffer: proc(device: *ID3D11Device, size: u64) -> *ID3D11Buffer = {
desc: D3D11_BUFFER_DESC = {
.ByteWidth = cast(size, UINT),
.Usage = D3D11_USAGE.D3D11_USAGE_DYNAMIC,
.BindFlags = D3D11_BIND_CONSTANT_BUFFER,
.CPUAccessFlags = D3D11_CPU_ACCESS_WRITE,
};
buffer: *ID3D11Buffer = null;
ID3D11Device_CreateBuffer(device, desc&, null, buffer&);
return buffer;
}
// WRITE_DISCARD hands back fresh memory, so the GPU never waits on the CPU.
update_buffer: proc(context: *ID3D11DeviceContext, buffer: *ID3D11Buffer, data: *const void, size: u64) -> void = {
mapped: D3D11_MAPPED_SUBRESOURCE = {};
ID3D11DeviceContext_Map(context, cast(buffer, *ID3D11Resource), 0, D3D11_MAP.D3D11_MAP_WRITE_DISCARD, 0, mapped&);
memcpy(mapped.pData, data, size);
ID3D11DeviceContext_Unmap(context, cast(buffer, *ID3D11Resource), 0);
}
// The same buffer in the same register for the vertex, geometry and pixel stages.
bind_constants: proc(context: *ID3D11DeviceContext, slot: UINT, buffer: *ID3D11Buffer) -> void = {
buffers: [1]*ID3D11Buffer = { buffer };
ID3D11DeviceContext_VSSetConstantBuffers(context, slot, 1, buffers[0]&);
ID3D11DeviceContext_GSSetConstantBuffers(context, slot, 1, buffers[0]&);
ID3D11DeviceContext_PSSetConstantBuffers(context, slot, 1, buffers[0]&);
}
// utility function for loading a 2D texture from file, with a full mip chain
// --------------------------------------------------------------------------
// `srgb`: the file stores colours, gamma encoded, and sampling should decode
// them to linear. Data such as normals or roughness is linear already.
load_texture: proc(device: *ID3D11Device, context: *ID3D11DeviceContext, path: [*]const char, srgb: b32) -> *ID3D11ShaderResourceView = {
img: image = image_load(path);
if (img.pixels == null) {
printf("Texture failed to load at path: %s\n", path);
exit(1);
}
desc: D3D11_TEXTURE2D_DESC = {
.Width = cast(img.width, UINT),
.Height = cast(img.height, UINT),
.MipLevels = 0, // the whole chain
.ArraySize = 1,
.Format = srgb ? DXGI_FORMAT.DXGI_FORMAT_R8G8B8A8_UNORM_SRGB : DXGI_FORMAT.DXGI_FORMAT_R8G8B8A8_UNORM,
.SampleDesc = { .Count = 1 },
// Generating mips draws into the texture, so it is a render target too.
.BindFlags = D3D11_BIND_SHADER_RESOURCE | D3D11_BIND_RENDER_TARGET,
.MiscFlags = D3D11_RESOURCE_MISC_GENERATE_MIPS,
};
texture: *ID3D11Texture2D = null;
srv: *ID3D11ShaderResourceView = null;
ID3D11Device_CreateTexture2D(device, desc&, null, texture&);
ID3D11Device_CreateShaderResourceView(device, cast(texture, *ID3D11Resource), null, srv&);
ID3D11DeviceContext_UpdateSubresource(context, cast(texture, *ID3D11Resource), 0, null, img.pixels, cast(img.width * 4, UINT), 0);
ID3D11DeviceContext_GenerateMips(context, srv);
return srv;
}
// A UV sphere of radius 1, indexed: `segments` slices around and as many
// stacks from pole to pole. u runs around the equator and v from the top
// pole down, the way an equirectangular texture wraps it.
sphere: struct = {
vertices: *ID3D11Buffer;
indices: *ID3D11Buffer;
index_count: UINT;
}
create_sphere: proc(device: *ID3D11Device, arena: *memops_arena, segments: u32) -> sphere = {
ring: u32 = segments + 1;
vertices: [*]f32 = cast(memops_arena_push(arena, cast(ring * ring * 8, u64) * 4, 16), [*]f32);
indices: [*]u32 = cast(memops_arena_push(arena, cast(segments * segments * 6, u64) * 4, 16), [*]u32);
for (y: u32 = 0; y <= segments; y += 1) {
for (x: u32 = 0; x <= segments; x += 1) {
u: f32 = cast(x, f32) / cast(segments, f32);
v: f32 = cast(y, f32) / cast(segments, f32);
p: vec3 = v3(cosf(u * 2.0 * pi) * sinf(v * pi), cosf(v * pi), sinf(u * 2.0 * pi) * sinf(v * pi));
values: [8]f32 = { p.x, p.y, p.z, p.x, p.y, p.z, u, v };
for (k: u32 = 0; k < 8; k += 1) { vertices[(y * ring + x) * 8 + k] = values[k]; }
}
}
n: u32 = 0;
for (y: u32 = 0; y < segments; y += 1) {
for (x: u32 = 0; x < segments; x += 1) {
a: u32 = y * ring + x; // this stack, this slice
b: u32 = a + ring; // the stack below
quad: [6]u32 = { a, a + 1, b + 1, a, b + 1, b }; // counter-clockwise from outside
for (i: u32 = 0; i < 6; i += 1) {
indices[n] = quad[i];
n += 1;
}
}
}
s: sphere = {
.vertices = create_buffer(device, D3D11_BIND_VERTEX_BUFFER, vertices, cast(ring * ring * 8, u64) * 4),
.indices = create_buffer(device, D3D11_BIND_INDEX_BUFFER, indices, cast(n, u64) * 4),
.index_count = n,
};
return s;
}
draw_sphere: proc(context: *ID3D11DeviceContext, constants: *ID3D11Buffer, t: *transforms, s: *const sphere, model: mat4) -> void = {
stride: UINT = 32;
offset: UINT = 0;
ID3D11DeviceContext_IASetVertexBuffers(context, 0, 1, s[0].vertices&, stride&, offset&);
ID3D11DeviceContext_IASetIndexBuffer(context, s[0].indices, DXGI_FORMAT.DXGI_FORMAT_R32_UINT, 0);
t[0].model = model;
t[0].normal_matrix = m4_normal_matrix(model);
update_buffer(context, constants, t, sizeof(transforms));
ID3D11DeviceContext_DrawIndexed(context, s[0].index_count, 0, 0);
}
// loads one PBR material's five maps from a folder; only albedo is a colour
load_material: proc(device: *ID3D11Device, context: *ID3D11DeviceContext, folder: [*]const char, out: [*]*ID3D11ShaderResourceView) -> void = {
names: [5][*]const char = { "albedo.png", "normal.png", "metallic.png", "roughness.png", "ao.png" };
for (i: u32 = 0; i < 5; i += 1) {
path: [260]char = {};
cstr_copy(path, filesystem_path(folder));
cstr_append(path, names[i]);
out[i] = load_texture(device, context, path[0]&, i == 0);
}
}
main: proc() -> i32 = {
// win32: register a window class and create the window
// ----------------------------------------------------
SetProcessDPIAware();
instance: HMODULE = GetModuleHandleA(null);
wc: WNDCLASSA = {
.lpfnWndProc = window_proc,
.hInstance = instance,
.hCursor = LoadCursorA(null, IDC_ARROW),
.lpszClassName = "LearnD3D11",
};
RegisterClassA(wc&);
// A fixed size: the swap chain and every render target keep the size they
// were created with, so nothing needs rebuilding on resize.
style: DWORD = WS_OVERLAPPEDWINDOW & ~(WS_THICKFRAME | WS_MAXIMIZEBOX);
rect: RECT = { .right = scr_width, .bottom = scr_height };
AdjustWindowRect(rect&, style, 0);
window: HWND = CreateWindowExA(0, wc.lpszClassName, "LearnD3D11 (Tab: uniform grid / textured)", style | WS_VISIBLE, CW_USEDEFAULT, CW_USEDEFAULT,
rect.right - rect.left, rect.bottom - rect.top, null, null, instance, null);
if (window == null) {
printf("Failed to create the window\n");
return -1;
}
// d3d11: create the device, its immediate context and a swap chain
// ----------------------------------------------------------------
swap_desc: DXGI_SWAP_CHAIN_DESC = {
.BufferDesc = {
.Width = cast(scr_width, UINT),
.Height = cast(scr_height, UINT),
.Format = DXGI_FORMAT.DXGI_FORMAT_R8G8B8A8_UNORM,
},
.SampleDesc = { .Count = 1 },
.BufferUsage = DXGI_USAGE_RENDER_TARGET_OUTPUT,
.BufferCount = 2,
.OutputWindow = window,
.Windowed = 1,
.SwapEffect = DXGI_SWAP_EFFECT.DXGI_SWAP_EFFECT_FLIP_DISCARD,
};
feature_level: D3D_FEATURE_LEVEL = D3D_FEATURE_LEVEL.D3D_FEATURE_LEVEL_11_0;
swap_chain: *IDXGISwapChain = null;
device: *ID3D11Device = null;
context: *ID3D11DeviceContext = null;
if (D3D11CreateDeviceAndSwapChain(null, D3D_DRIVER_TYPE.D3D_DRIVER_TYPE_HARDWARE, null, 0, feature_level&, 1, D3D11_SDK_VERSION,
swap_desc&, swap_chain&, device&, null, context&) < 0) {
printf("Failed to create the D3D11 device\n");
return -1;
}
backbuffer: *ID3D11Texture2D = null;
IDXGISwapChain_GetBuffer(swap_chain, 0, IID_ID3D11Texture2D&, cast(backbuffer&, **void));
// The back buffer is viewed as sRGB: what the shaders write is linear, and
// the view gamma-encodes it on the way in (5.2).
backbuffer_view: D3D11_RENDER_TARGET_VIEW_DESC = {
.Format = DXGI_FORMAT.DXGI_FORMAT_R8G8B8A8_UNORM_SRGB,
.ViewDimension = D3D11_RTV_DIMENSION.D3D11_RTV_DIMENSION_TEXTURE2D,
};
backbuffer_rtv: *ID3D11RenderTargetView = null;
ID3D11Device_CreateRenderTargetView(device, cast(backbuffer, *ID3D11Resource), backbuffer_view&, backbuffer_rtv&);
// A depth buffer the size of the window: 24 bits of depth, 8 of stencil.
depth_desc: D3D11_TEXTURE2D_DESC = {
.Width = cast(scr_width, UINT),
.Height = cast(scr_height, UINT),
.MipLevels = 1,
.ArraySize = 1,
.Format = DXGI_FORMAT.DXGI_FORMAT_D24_UNORM_S8_UINT,
.SampleDesc = { .Count = 1 },
.BindFlags = D3D11_BIND_DEPTH_STENCIL,
};
depth_texture: *ID3D11Texture2D = null;
depth_dsv: *ID3D11DepthStencilView = null;
ID3D11Device_CreateTexture2D(device, depth_desc&, null, depth_texture&);
ID3D11Device_CreateDepthStencilView(device, cast(depth_texture, *ID3D11Resource), null, depth_dsv&);
// configure global d3d11 state
// ----------------------------
// GL's conventions rather than D3D's: counter-clockwise triangles face the
// viewer, and nothing is culled.
raster_desc: D3D11_RASTERIZER_DESC = {
.FillMode = D3D11_FILL_MODE.D3D11_FILL_SOLID,
.CullMode = D3D11_CULL_MODE.D3D11_CULL_NONE,
.FrontCounterClockwise = 1,
.DepthClipEnable = 1,
};
raster_state: *ID3D11RasterizerState = null;
ID3D11Device_CreateRasterizerState(device, raster_desc&, raster_state&);
ID3D11DeviceContext_RSSetState(context, raster_state);
viewport: D3D11_VIEWPORT = {
.Width = cast(scr_width, f32),
.Height = cast(scr_height, f32),
.MaxDepth = 1.0,
};
// A sampler: how textures are filtered and what happens past their edges.
sampler_desc: D3D11_SAMPLER_DESC = {
.Filter = D3D11_FILTER.D3D11_FILTER_ANISOTROPIC,
.AddressU = D3D11_TEXTURE_ADDRESS_MODE.D3D11_TEXTURE_ADDRESS_WRAP,
.AddressV = D3D11_TEXTURE_ADDRESS_MODE.D3D11_TEXTURE_ADDRESS_WRAP,
.AddressW = D3D11_TEXTURE_ADDRESS_MODE.D3D11_TEXTURE_ADDRESS_WRAP,
.MaxAnisotropy = 8,
.MaxLOD = D3D11_FLOAT32_MAX,
};
sampler: *ID3D11SamplerState = null;
ID3D11Device_CreateSamplerState(device, sampler_desc&, sampler&);
ID3D11DeviceContext_PSSetSamplers(context, 0, 1, sampler&);
// build and compile shaders
// -------------------------
s: shader = shader_create(device, filesystem_path("src/6.pbr/1.1.lighting/pbr.hlsl"), vertex_layout);
// set up vertex data and textures
// -------------------------------
arena: memops_arena = {};
memops_arena_initialize(arena&);
ball: sphere = create_sphere(device, arena&, 64);
transforms_buffer: *ID3D11Buffer = create_constant_buffer(device, sizeof(transforms));
material_buffer: *ID3D11Buffer = create_constant_buffer(device, sizeof(material_constants));
lights_buffer: *ID3D11Buffer = create_constant_buffer(device, sizeof(lights_constants));
rusted_iron: [5]*ID3D11ShaderResourceView = {};
load_material(device, context, "resources/textures/pbr/rusted_iron/", rusted_iron[0]&);
cam = camera_create(v3(0.0, 0.0, 22.0), -90.0, 0.0);
textured: i32 = 0;
// lights
// ------
lights: lights_constants = {
.positions = {
v4(-10.0, 10.0, 10.0, 0.0),
v4(10.0, 10.0, 10.0, 0.0),
v4(-10.0, -10.0, 10.0, 0.0),
v4(10.0, -10.0, 10.0, 0.0),
},
.colors = {
v4(300.0, 300.0, 300.0, 0.0),
v4(300.0, 300.0, 300.0, 0.0),
v4(300.0, 300.0, 300.0, 0.0),
v4(300.0, 300.0, 300.0, 0.0),
},
};
// render loop
// -----------
running: b32 = 1;
while (running) {
msg: MSG = {};
while (PeekMessageA(msg&, null, 0, 0, PM_REMOVE) != 0) {
if (msg.message == WM_QUIT) { running = 0; }
TranslateMessage(msg&);
DispatchMessageA(msg&);
}
// per-frame time logic
// --------------------
current_frame: f32 = get_time();
delta_time = current_frame - last_frame;
last_frame = current_frame;
// input
// -----
process_input(window);
if (key_pressed(VK_TAB)) { textured = 1 - textured; }
// render
// ------
clear_color: [4]f32 = { 0.1, 0.1, 0.1, 1.0 };
ID3D11DeviceContext_OMSetRenderTargets(context, 1, backbuffer_rtv&, depth_dsv);
ID3D11DeviceContext_RSSetViewports(context, 1, viewport&);
ID3D11DeviceContext_ClearRenderTargetView(context, backbuffer_rtv, clear_color[0]&);
ID3D11DeviceContext_ClearDepthStencilView(context, depth_dsv, D3D11_CLEAR_DEPTH | D3D11_CLEAR_STENCIL, 1.0, 0);
ID3D11DeviceContext_IASetPrimitiveTopology(context, D3D11_PRIMITIVE_TOPOLOGY.D3D11_PRIMITIVE_TOPOLOGY_TRIANGLELIST);
bind_constants(context, 0, transforms_buffer);
t: transforms = {
.view = camera_view(cam&),
.projection = m4_perspective(radians(cam.zoom), cast(scr_width, f32) / cast(scr_height, f32), 0.1, 100.0),
.view_position = cam.position,
};
// the lights wander a little, so the highlights move across the spheres
for (i: u32 = 0; i < 4; i += 1) {
lights.positions[i].x = (i % 2 == 0 ? -10.0 : 10.0) + 5.0 * sinf(current_frame);
}
update_buffer(context, lights_buffer, lights&, sizeof(lights_constants));
shader_use(context, s&);
bind_constants(context, 1, material_buffer);
bind_constants(context, 2, lights_buffer);
ID3D11DeviceContext_PSSetShaderResources(context, 0, 5, rusted_iron[0]&);
m: material_constants = { .albedo = v3(0.5, 0.0, 0.0), .ao = 1.0, .textured = textured };
// render rows*column number of spheres with varying metallic/roughness values:
// metallic from bottom to top, roughness from left to right
for (row: i32 = 0; row < 7; row += 1) {
for (col: i32 = 0; col < 7; col += 1) {
m.metallic = cast(row, f32) / 6.0;
// perfectly smooth reflects one point of light, which looks wrong
// for a point light: start just above zero
m.roughness = clampf(cast(col, f32) / 6.0, 0.05, 1.0);
update_buffer(context, material_buffer, m&, sizeof(material_constants));
draw_sphere(context, transforms_buffer, t&, ball&, m4_translate(v3(cast(col - 3, f32) * 2.5, cast(row - 3, f32) * 2.5, 0.0)));
}
}
// dxgi: show the image just drawn, waiting for the next vertical blank
// --------------------------------------------------------------------
IDXGISwapChain_Present(swap_chain, 1, 0);
}
return 0;
}