Cluster indexes, not vertexes

backface cone rejection
Merge branch 'skmp/fix-AllocateBoatWakeArray' into 'main'
2025-01-13 01:33:34 +02:00 · 2025-01-12 18:57:25 +02:00 · 2025-01-06 11:46:50 +00:00 · 2025-01-06 11:41:11 +00:00 · 2025-01-06 13:40:31 +02:00 · 2025-01-05 15:40:18 +02:00
10 changed files with 430 additions and 202 deletions
--- a/.vscode/launch.json
+++ b/.vscode/launch.json
@@ -24,6 +24,30 @@
          }
      ]
    },
+    {
+      "name": "dca3-sim (linux)",
+      "type": "cppdbg",
+      "request": "launch",
+      "program": "${workspaceFolder}/dreamcast/dca3-sim.elf",
+      "args": [],
+      "stopAtEntry": false,
+      "cwd": "${workspaceFolder}/dreamcast",
+      "environment": [],
+      "externalConsole": false,
+      "MIMode": "gdb",
+      "setupCommands": [
+          {
+              "description": "Enable pretty-printing for gdb",
+              "text": "-enable-pretty-printing",
+              "ignoreFailures": true
+          },
+          {
+              "description": "Set Disassembly Flavor to Intel",
+              "text": "-gdb-set disassembly-flavor intel",
+              "ignoreFailures": true
+          }
+      ]
+    },
    {
      "name": "dca3-sim (mac)",
      "type": "cppdbg",
--- a/README.md
+++ b/README.md
@@ -28,12 +28,12 @@ You will also need the following tools installed

 ### Cloning the dca3-game repo and downloading the prebuilt elf
 - Open dreamsdk shell
- type `git clone https://gitlab.com/skmp/dca3-game.git` (and press enter)
+- type `git clone --branch alpha https://gitlab.com/skmp/dca3-game.git` (and press enter)
 - It should take a moment and successfully clone the repo
 - type `cd dca3-game/dreamcast` (and press enter)
 - type `explorer .` (and press enter).
 - A folder named dreamcast with some files should be open. Keep it on the side.
- Download the artifacts from https://gitlab.com/skmp/dca3-game/-/jobs/8725216645
+- Download the Alpha Prebuilt Elf from https://gitlab.com/skmp/dca3-game/-/releases
 - Open artifacts.zip and extract dca3.elf to the folder that was kept open before.
 - Close the folder and dreamsdk shell

--- a/public/index.html
+++ b/public/index.html
@@ -3,100 +3,10 @@
 <head>
    <meta charset="UTF-8">
    <meta name="viewport" content="width=device-width, initial-scale=1.0">
-    <title>DCA3 - A port of RE3 for Dreamcast</title>
-    <!-- Bootstrap CSS -->
-    <link href="https://cdn.jsdelivr.net/npm/bootstrap@5.3.2/dist/css/bootstrap.min.css" rel="stylesheet">
-    <style>
-        body {
-            background-color: #121212;
-            color: #ffffff;
-        }
-        .navbar {
-            background-color: #1f1f1f;
-        }
-        .hero {
-            text-align: center;
-            padding: 100px 20px;
-            background-color: #333333;
-            border-radius: 10px;
-        }
-        .features {
-            margin-top: 50px;
-        }
-        .footer {
-            text-align: center;
-            padding: 20px;
-            background-color: #1f1f1f;
-            margin-top: 20px;
-            border-radius: 10px;
-        }
-    </style>
+    <title>DCA3 Redirect</title>
+    <meta http-equiv="refresh" content="5;url=https://dca3.net">
 </head>
 <body>
-    <!-- Navbar -->
-    <nav class="navbar navbar-expand-lg navbar-dark">
-        <div class="container">
-            <a class="navbar-brand" href="#">DCA3</a>
-            <button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navbarNav">
-                <span class="navbar-toggler-icon"></span>
-            </button>
-            <div class="collapse navbar-collapse" id="navbarNav">
-                <ul class="navbar-nav ms-auto">
-                    <li class="nav-item">
-                        <a class="nav-link" href="#">Home</a>
-                    </li>
-                    <li class="nav-item">
-                        <a class="nav-link" href="#features">Features</a>
-                    </li>
-                    <li class="nav-item">
-                        <a class="nav-link" href="#download">Download</a>
-                    </li>
-                </ul>
-            </div>
-        </div>
-    </nav>
-
-    <!-- Hero Section -->
-    <div class="container">
-        <div class="hero mt-4">
-            <h1>DCA3</h1>
-            <p>The first-ever port of Grand Theft Auto III to the SEGA Dreamcast, built from the REGTA reverse engineering project.</p>
-            <a href="#download" class="btn btn-primary">Download Now</a>
-        </div>
-    </div>
-
-    <!-- Features Section -->
-    <div class="container features text-center">
-        <h2 id="features" class="mb-4">Key Features</h2>
-        <div class="row">
-            <div class="col-md-4">
-                <h3>Faithful Port</h3>
-                <p>Experience GTA3 as it was meant to be, now on the Dreamcast with okay visuals and sluggish gameplay.</p>
-            </div>
-            <div class="col-md-4">
-                <h3>Open Source</h3>
-                <p>Built on the REGTA reverse engineering project, ensuring accuracy and community-driven development.</p>
-            </div>
-            <div class="col-md-4">
-                <h3>First Release</h3>
-                <p>Celebrating our first release with not very much optimized performance and compatibility for Dreamcast hardware.</p>
-            </div>
-        </div>
-    </div>
-
-    <!-- Download Section -->
-    <div class="container text-center mt-5">
-        <h2 id="download" class="mb-4">Download DCA3</h2>
-        <p>Get started with the first release of  and bring GTA3 to your Dreamcast. You'll need a PC copy of GTA3 to create your .cdi</p>
-        <a href="#" class="btn btn-success btn-lg">Download Version 1.0</a>
-    </div>
-
-    <!-- Footer -->
-    <footer class="footer mt-4">
-        <p>No rights reserved. Made with love for the Dreamcast community.</p>
-    </footer>
-
-    <!-- Bootstrap JS Bundle -->
-    <script src="https://cdn.jsdelivr.net/npm/bootstrap@5.3.2/dist/js/bootstrap.bundle.min.js"></script>
+    <p>You will be redirected to <a href="https://dca3.net">dca3.net</a> in 5 seconds.</p>
 </body>
-</html>
+</html>
--- a/src/audio/AudioLogic.cpp
+++ b/src/audio/AudioLogic.cpp
@@ -6302,7 +6302,7 @@ cPedComments::Process()
 		{
 		case LOADING_STATUS_NOT_LOADED:
 			SampleManager.LoadPedComment(sampleIndex);
-#ifdef GTA_PS2 // on PC ped comment is loaded at once
+#if defined(GTA_PS2) || defined(RW_DC) // on PC ped comment is loaded at once
 			break;
 #endif
 		case LOADING_STATUS_LOADED:
--- a/src/audio/sampman_dc.cpp
+++ b/src/audio/sampman_dc.cpp
@@ -166,8 +166,12 @@ struct sfx_bank {
 std::map<int, sfx_bank> sfx_banks;

 int nPedSlotSfx[MAX_PEDSFX];
+uint32 nPedSlotSfxReqId[MAX_PEDSFX];
 uintptr_t nPedSlotSfxAddr[MAX_PEDSFX];
 uint8_t nCurrentPedSlot;
+file_t fdPedSfx;
+volatile uint32 nPedSfxReqReadId = 1;
+volatile uint32 nPedSfxReqNextId = 1;

 struct WavHeader {
    // RIFF Header
@@ -442,15 +446,22 @@ cSampleManager::Initialise(void)
 		}
 	});
 	
+	nPedSfxReqNextId = 1;
+	nPedSfxReqReadId = 1;
 	for ( int32 i = 0; i < MAX_PEDSFX; i++ )
 	{
 		nPedSlotSfx[i]     = -1;
+		nPedSlotSfxReqId[i] = 0;
 		nPedSlotSfxAddr[i] = snd_mem_malloc(PED_BLOCKSIZE_ADPCM);
 		debugf("PedSlot %d buffer: %p\n", i, (void*)nPedSlotSfxAddr[i]);
 	}
 	
 	nCurrentPedSlot = 0;

+	fdPedSfx = fs_open(SampleBankDataFilename, O_RDONLY);
+
+	assert(fdPedSfx >= 0);
+
 	_dcAudioInitialized = true;
 	return TRUE;
 }
@@ -458,7 +469,7 @@ cSampleManager::Initialise(void)
 void
 cSampleManager::Terminate(void)
 {
-
+	fs_close(fdPedSfx);
 }

 bool8 cSampleManager::CheckForAnAudioFileOnCD(void)
@@ -637,7 +648,7 @@ cSampleManager::IsPedCommentLoaded(uint32 nComment)
 			slot += ARRAY_SIZE(nPedSlotSfx);
 #endif
 		if ( nComment == nPedSlotSfx[slot] )
-			return LOADING_STATUS_LOADED;
+			return nPedSlotSfxReqId[slot] <= nPedSfxReqReadId ? LOADING_STATUS_LOADED : LOADING_STATUS_LOADING;
 	}
 	
 	return LOADING_STATUS_NOT_LOADED;
@@ -698,27 +709,28 @@ cSampleManager::LoadPedComment(uint32 nComment)

 	assert(m_aSamples[nComment].nByteSize < PED_BLOCKSIZE_ADPCM);

-	file_t fd = fs_open(SampleBankDataFilename, O_RDONLY);
-
-	assert(fd >= 0);
-	debugf("Loading ped comment %d, offset: %d, size: %d\n", nComment, m_aSamples[nComment].nFileOffset, m_aSamples[nComment].nByteSize);
-	fs_seek(fd, m_aSamples[nComment].nFileOffset, SEEK_SET);
+	CdStreamQueueAudioRead(nComment, (void*)nPedSlotSfxAddr[nCurrentPedSlot], m_aSamples[nComment].nByteSize, m_aSamples[nComment].nFileOffset, [](AudioReadCmd* cmd) {
+		debugf("Loading ped comment %d, offset: %d, size: %d\n", nComment, m_aSamples[nComment].nFileOffset, m_aSamples[nComment].nByteSize);
+		fs_seek(fdPedSfx, cmd->seek, SEEK_SET);


-	// TODO: When we can dma directly to AICA, we can use this instead
-	// fs_read(fd, SPU_BASE_U8 + nPedSlotSfxAddr[nCurrentPedSlot], sizeof(nPedSlotSfxAddr));
+		// TODO: When we can dma directly to AICA, we can use this instead
+		// fs_read(fdPedSfx, SPU_BASE_U8 + (uintptr_t)cmd->dest, cmd->size);

-	void* stagingBuffer = memalign(32, m_aSamples[nComment].nByteSize);
-	assert(stagingBuffer != 0);
-	debugf("Allocated %d bytes at %p\n", m_aSamples[nComment].nByteSize, stagingBuffer);
-	int rs = fs_read(fd, stagingBuffer, m_aSamples[nComment].nByteSize);
-	debugf("Read %d bytes, expected %d\n", rs, m_aSamples[nComment].nByteSize);
-	assert(rs == m_aSamples[nComment].nByteSize);
+		void* stagingBuffer = memalign(32, cmd->size);
+		assert(stagingBuffer != 0);
+		debugf("Allocated %d bytes at %p\n", cmd->size, stagingBuffer);
+		int rs = fs_read(fdPedSfx, stagingBuffer, cmd->size);
+		debugf("Read %d bytes, expected %d\n", rs, cmd->size);
+		assert(rs == cmd->size);

-	fs_close(fd);
+		spu_memload((uintptr_t)cmd->dest, stagingBuffer, cmd->size);
+		free(stagingBuffer);
+		nPedSfxReqReadId = nPedSfxReqReadId + 1;
+	});

-	spu_memload(nPedSlotSfxAddr[nCurrentPedSlot], stagingBuffer, m_aSamples[nComment].nByteSize);
-	free(stagingBuffer);
+	nPedSlotSfxReqId[nCurrentPedSlot] = ++nPedSfxReqNextId;
+	
 	nPedSlotSfx[nCurrentPedSlot] = nComment;

 	if ( ++nCurrentPedSlot >= MAX_PEDSFX )
--- a/src/core/CdStream.h
+++ b/src/core/CdStream.h
@@ -1,4 +1,5 @@
 #pragma once
+#include <functional>

 #define CDSTREAM_SECTOR_SIZE 2048

@@ -43,7 +44,14 @@ char *CdStreamGetImageName(int32 cd);
 void CdStreamRemoveImages(void);
 int32 CdStreamGetNumImages(void);

-void CdStreamQueueAudioRead(int fd, void* pBuffer, size_t bytes, size_t seek);
+struct AudioReadCmd {
+	void* dest;
+	int fd;
+	size_t size;
+	size_t seek;
+	std::function<void(AudioReadCmd*)> callback;
+};
+void CdStreamQueueAudioRead(int fd, void* pBuffer, size_t bytes, size_t seek, std::function<void(AudioReadCmd*)> callback = nullptr);
 void CdStreamDiscardAudioRead(int fd);

 #ifdef FLUSHABLE_STREAMING
--- a/src/core/CdStreamDC.cpp
+++ b/src/core/CdStreamDC.cpp
@@ -26,12 +26,6 @@
 #include "CdStream.h"
 #include "rwcore.h"
 #include "MemoryMgr.h"
-struct AudioReadCmd {
-	void* dest;
-	int fd;
-	size_t size;
-	size_t seek;
-};

 #define CDDEBUG(f, ...)   debug ("%s: " f "\n", "cdvd_stream", ## __VA_ARGS__)
 #define CDTRACE(f, ...)   printf("%s: " f "\n", "cdvd_stream", ## __VA_ARGS__)
@@ -489,8 +483,16 @@ std::vector<AudioReadCmd> pendingAudioReads;
 std::mutex pendingAudioReadsMutex;
 #endif
 // Will replace a previous read request for the same file descriptor
-void CdStreamQueueAudioRead(int fd, void* pBuffer, size_t bytes, size_t seek) {
+void CdStreamQueueAudioRead(int fd, void* pBuffer, size_t bytes, size_t seek, std::function<void(AudioReadCmd*)> callback) {
 	AudioReadCmd cmd = { pBuffer, fd, bytes, seek};
+	if (!callback) {
+		cmd.callback = [](AudioReadCmd* cmd){
+			lseek(cmd->fd, cmd->seek, SEEK_SET);
+			read(cmd->fd, cmd->dest, cmd->size);
+		};
+	} else {
+		cmd.callback = callback;
+	}
 	{
 		#if !defined(DC_SH4)
 		std::lock_guard<std::mutex> lock(pendingAudioReadsMutex);
@@ -514,6 +516,7 @@ void CdStreamQueueAudioRead(int fd, void* pBuffer, size_t bytes, size_t seek) {
 	sem_post(gCdStreamSema);
 }

+
 void CdStreamDiscardAudioRead(int fd) {
 	#if !defined(DC_SH4)
 	std::lock_guard<std::mutex> lock(pendingAudioReadsMutex);
@@ -564,8 +567,7 @@ int read_loop(int fd, void* pBuffer, size_t bytes) {
 		total_read += read_bytes;
 		auto cmd = CdStreamNextAudioRead();
 		while (cmd.fd != -1) {
-			lseek(cmd.fd, cmd.seek, SEEK_SET);
-			read(cmd.fd, cmd.dest, cmd.size);
+			cmd.callback(&cmd);
 			cmd = CdStreamNextAudioRead();
 		}
 	}
@@ -581,8 +583,7 @@ void *CdStreamThread(void *param)

 		auto cmd = CdStreamNextAudioRead();
 		while (cmd.fd != -1) {
-			lseek(cmd.fd, cmd.seek, SEEK_SET);
-			read(cmd.fd, cmd.dest, cmd.size);
+			cmd.callback(&cmd);
 			cmd = CdStreamNextAudioRead();
 		}

--- a/src/math/Matrix.cpp
+++ b/src/math/Matrix.cpp
@@ -282,7 +282,7 @@ CMatrix::SetRotate(float xAngle, float yAngle, float zAngle)
 void
 CMatrix::RotateX(float x)
 {
-#ifdef DC_SH4
+#if 0 && defined(DC_SH4) // this is bugged and does not yield correct results
 	mat_load(reinterpret_cast<matrix_t *>(this));
 	mat_rotate_x(x);
 	mat_store(reinterpret_cast<matrix_t *>(this));
@@ -312,7 +312,7 @@ CMatrix::RotateX(float x)
 void
 CMatrix::RotateY(float y)
 {
-#ifdef DC_SH4
+#if 0 && defined(DC_SH4) // this is bugged and does not yield correct results
 	mat_load(reinterpret_cast<matrix_t *>(this));
 	mat_rotate_y(y);
 	mat_store(reinterpret_cast<matrix_t *>(this));
@@ -342,7 +342,7 @@ CMatrix::RotateY(float y)
 void
 CMatrix::RotateZ(float z)
 {
-#ifdef DC_SH4
+#if 0 && defined(DC_SH4) // this is bugged and does not yield correct results
 	mat_load(reinterpret_cast<matrix_t *>(this));
 	mat_rotate_z(z);
 	mat_store(reinterpret_cast<matrix_t *>(this));
@@ -372,7 +372,7 @@ CMatrix::RotateZ(float z)
 void
 CMatrix::Rotate(float x, float y, float z)
 {
-#ifdef DC_SH4
+#if 0 && defined(DC_SH4) // this is bugged and does not yield correct results
 	mat_load(reinterpret_cast<matrix_t *>(this));
 	mat_rotate(x, y, z);
 	mat_store(reinterpret_cast<matrix_t *>(this));
@@ -404,6 +404,20 @@ CMatrix::Rotate(float x, float y, float z)
 	float z2 = sZ * sY - (cZ * sX) * cY;
 	float z3 = cX * cY;

+	#if !defined(DC_TEXCONV) && !defined(DC_SIM)
+	this->rx = fipr(x1, y1, z1, 0,  rx, ry, rz, 0);
+	this->ry = fipr(x2, y2, z2, 0,  rx, ry, rz, 0);
+	this->rz = fipr(x3, y3, z3, 0,  rx, ry, rz, 0);
+	this->fx = fipr(x1, y1, z1, 0,  ux, uy, uz, 0);
+	this->fy = fipr(x2, y2, z2, 0,  ux, uy, uz, 0);
+	this->fz = fipr(x3, y3, z3, 0,  ux, uy, uz, 0);
+	this->ux = fipr(x1, y1, z1, 0,  ax, ay, az, 0);
+	this->uy = fipr(x2, y2, z2, 0,  ax, ay, az, 0);
+	this->uz = fipr(x3, y3, z3, 0,  ax, ay, az, 0);
+	this->px = fipr(x1, y1, z1, 0,  px, py, pz, 0);
+	this->py = fipr(x2, y2, z2, 0,  px, py, pz, 0);
+	this->pz = fipr(x3, y3, z3, 0,  px, py, pz, 0);
+	#else
 	this->rx = x1 * rx + y1 * ry + z1 * rz;
 	this->ry = x2 * rx + y2 * ry + z2 * rz;
 	this->rz = x3 * rx + y3 * ry + z3 * rz;
@@ -416,6 +430,7 @@ CMatrix::Rotate(float x, float y, float z)
 	this->px = x1 * px + y1 * py + z1 * pz;
 	this->py = x2 * px + y2 * py + z2 * pz;
 	this->pz = x3 * px + y3 * py + z3 * pz;
+	#endif
 #endif
 }

--- a/src/renderer/WaterLevel.cpp
+++ b/src/renderer/WaterLevel.cpp
@@ -1479,7 +1479,8 @@ CWaterLevel::AllocateBoatWakeArray()
 			apGeomArray[geom] = RpGeometryCreate(9*9, 8*8*2, rpGEOMETRYTRISTRIP
 															| rpGEOMETRYPRELIT
 															| rpGEOMETRYMODULATEMATERIALCOLOR
-															| rpGEOMETRYTEXTURED);
+															| rpGEOMETRYTEXTURED
+															| rw::Geometry::HAS_TRIANGLES/* RW_DC specific */);
 			ASSERT(apGeomArray[geom] != nil);

 			RpTriangle *geomTriangles = RpGeometryGetTriangles(apGeomArray[geom]);
--- a/vendor/librw/src/dc/rwdc.cpp
+++ b/vendor/librw/src/dc/rwdc.cpp
@@ -4,6 +4,11 @@
 #include <stdlib.h>
 #include <string.h>
 #include <assert.h>
+#include <cstdint>
+#include <vector>
+#include <iostream>
+#include <cmath>
+#include <set>

 #if !defined(DC_TEXCONV) && !defined(MACOS64)
 #include <malloc.h>
@@ -595,6 +600,8 @@ struct atomic_context_t {

 	matrix_t worldView, mtx;
 	UniformObject uniform;
+	V3d cameraDir;
+	float cosPhi;
 };
 /* END Ligting Structs and Defines */

@@ -1754,24 +1761,23 @@ void addInterpolatedVertex(const pvr_vertex16_t& v1, const pvr_vertex16_t& v2, u
 }

 struct MeshInfo {
-	int16_t meshletCount;
-	int16_t meshletOffset;
+	uint32_t meshletCount;
+	uint32_t meshletOffset;
 };
-static_assert(sizeof(MeshInfo) == 4);
+static_assert(sizeof(MeshInfo) == 8);

 struct MeshletInfo {
 	RwSphere boundingSphere;
-	uint16_t flags;
-	int8_t pad;
+	uint8_t flags;
+	uint8_t clusterCount;
 	int8_t vertexSize;
-	uint16_t vertexCount;
-	uint16_t indexCount;
+	uint8_t vertexCount;
 	uint32_t vertexOffset;
 	uint32_t indexOffset;
 	uint32_t skinIndexOffset;
 	uint32_t skinWeightOffset;
 };
-static_assert(sizeof(MeshletInfo) == 40); // or 32 if !skin
+static_assert(sizeof(MeshletInfo) == 36); // or 28 if !skin


 inline __attribute__((always_inline))  void setLights(Atomic *atomic, WorldLights *lightData, UniformObject &uniformObject)
@@ -3229,6 +3235,41 @@ size_t vertexBufferFree() {
 }


+float calculateMaxAngularOffset(
+    const V3d& cameraPos,      // Camera position
+    const V3d& cameraLookAt,   // Camera look-at point (not direction)
+    const V3d& sphereCenter,   // Sphere center
+    float sphereRadius)        // Sphere radius
+{
+    // Compute the look-at direction vector (normalized)
+    V3d frustumAxis = cameraLookAt;
+
+    // Vector from camera to sphere center
+    V3d V = sub(sphereCenter, cameraPos);
+
+       if (length(V) < sphereRadius) {
+               return 1;
+       }
+
+    // Distance along the camera look-at direction
+    float d_parallel = dot(V, frustumAxis);
+
+    // Perpendicular vector (V_perp = V - d_parallel * frustumAxis)
+    V3d V_perp = {
+        V.x - d_parallel * frustumAxis.x,
+        V.y - d_parallel * frustumAxis.y,
+        V.z - d_parallel * frustumAxis.z
+    };
+    float d_perpendicular = length(V_perp);
+
+    // Total perpendicular distance including sphere radius
+    float d_total = d_perpendicular + sphereRadius;
+
+    // Maximum angular offset
+    float theta_max = std::atan2(d_total, d_parallel);
+    return theta_max; // Result in radians
+}
+
 void defaultRenderCB(ObjPipeline *pipe, Atomic *atomic) {
    rw::Camera *cam = engine->currentCamera;
    // Frustum Culling
@@ -3286,6 +3327,28 @@ void defaultRenderCB(ObjPipeline *pipe, Atomic *atomic) {
 	mat_apply((matrix_t*)&atomicContexts.back().worldView);
 	mat_store((matrix_t*)&atomicContexts.back().mtx);

+	// TODO: Don't invert twice for the same atomic
+	Matrix magic;
+	Matrix::invert(&magic, atomic->getFrame()->getLTM());
+
+	V3d camera_dir;
+	V3d::transformVectors(&camera_dir, &cam->getFrame()->getLTM()->at, 1, &magic);
+	camera_dir = normalize(camera_dir);
+	// required because of camera up being negative
+	camera_dir.x*=-1;
+	camera_dir.y*=-1;
+	camera_dir.z*=-1;
+
+	atomicContexts.back().cameraDir = camera_dir;
+
+	float angle = calculateMaxAngularOffset(cam->getFrame()->getLTM()->pos, cam->getFrame()->getLTM()->at, atomic->getWorldBoundingSphere()->center, atomic->getWorldBoundingSphere()->radius);
+	
+	angle = fabs(angle);
+	if (angle > 70/2 * M_PI / 180) {
+		angle = 70/2 * M_PI / 180;
+	}
+	atomicContexts.back().cosPhi = cosf(90 * M_PI / 180 + 5.1 * M_PI / 180 + angle);
+
 	int16_t contextId = atomicContexts.size() - 1;

 	assert(numMeshes <= 32767);
@@ -3294,6 +3357,8 @@ void defaultRenderCB(ObjPipeline *pipe, Atomic *atomic) {

 	for (int16_t n = 0; n < numMeshes; n++) {
 		bool doBlend = meshes[n].material->color.alpha != 255; // TODO: check all vertexes for alpha?
+		bool doBlendMaterial = doBlend;
+
 		bool textured = geo->numTexCoordSets && meshes[n].material->texture;
 		if (textured) {
 			doBlend |= Raster::formatHasAlpha(meshes[n].material->texture->raster->format);
@@ -3343,7 +3408,7 @@ void defaultRenderCB(ObjPipeline *pipe, Atomic *atomic) {
 		pvr_poly_cxt_t cxt;
 		int pvrList;
 		if (doBlend || isMatFX) {
-			if (doAlphaTest) {
+			if (doAlphaTest && !doBlendMaterial) {
 				pvrList = PVR_LIST_PT_POLY;
 			} else {
 				pvrList = PVR_LIST_TR_POLY;
@@ -3613,9 +3678,37 @@ void defaultRenderCB(ObjPipeline *pipe, Atomic *atomic) {
 					auto indexData = (int8_t*)&dcModel->data[meshlet->indexOffset];

 					if (!clippingRequired) {
-						submitMeshletSelector[textured](OCR_SPACE, indexData, meshlet->indexCount);
+						unsigned numClusters = meshlet->clusterCount;
+						auto currentIndexData = indexData;
+						do {
+							V3d coneNormal = { currentIndexData[0] / 127.f, currentIndexData[1] / 127.f, currentIndexData[2] / 127.f };
+							float costheta = dot(acp->cameraDir, coneNormal);
+							unsigned indexCount = (uint8_t&)currentIndexData[3];
+
+							currentIndexData += 4;
+							if (costheta >= acp->cosPhi) {
+								submitMeshletSelector[textured](OCR_SPACE, currentIndexData, indexCount);
+							} else {
+								// printf("CONE CULL, %f %f, %f %f %f\n", costheta, acp->cosPhi, meshlet->coneNormal.x, meshlet->coneNormal.y, meshlet->coneNormal.z);
+							}
+							currentIndexData += indexCount;
+						} while(--numClusters != 0);
 					} else {
-						clipAndsubmitMeshletSelector[textured](OCR_SPACE, indexData, meshlet->indexCount);
+						unsigned numClusters = meshlet->clusterCount;
+						auto currentIndexData = indexData;
+						do {
+							V3d coneNormal = { currentIndexData[0] / 127.f, currentIndexData[1] / 127.f, currentIndexData[2] / 127.f };
+							float costheta = dot(acp->cameraDir, coneNormal);
+							unsigned indexCount = (uint8_t&)currentIndexData[3];
+
+							currentIndexData += 4;
+							if (costheta >= acp->cosPhi) {
+								clipAndsubmitMeshletSelector[textured](OCR_SPACE, currentIndexData, indexCount);
+							} else {
+								// printf("CONE CULL, %f %f, %f %f %f\n", costheta, acp->cosPhi, meshlet->coneNormal.x, meshlet->coneNormal.y, meshlet->coneNormal.z);
+							}
+							currentIndexData += indexCount;
+						} while(--numClusters != 0);
 					}

 					if (meshContext->matfxContextOffset != SIZE_MAX) {
@@ -3639,9 +3732,37 @@ void defaultRenderCB(ObjPipeline *pipe, Atomic *atomic) {
 						tnlMeshletEnvMap(OCR_SPACE, &dcModel->data[meshlet->vertexOffset] + normalOffset, meshlet->vertexCount, meshlet->vertexSize, &matfxContext->mtx, matfxContext->coefficient);

 						if (!clippingRequired) {
-							submitMeshletSelector[true](OCR_SPACE, indexData, meshlet->indexCount);
+							unsigned numClusters = meshlet->clusterCount;
+							auto currentIndexData = indexData;
+							do {
+								V3d coneNormal = { currentIndexData[0] / 127.f, currentIndexData[1] / 127.f, currentIndexData[2] / 127.f };
+								float costheta = dot(acp->cameraDir, coneNormal);
+								unsigned indexCount = (uint8_t&)currentIndexData[3];
+
+								currentIndexData += 4;
+								if (costheta >= acp->cosPhi) {
+									submitMeshletSelector[true](OCR_SPACE, currentIndexData, indexCount);
+								} else {
+									// printf("CONE CULL, %f %f, %f %f %f\n", costheta, acp->cosPhi, meshlet->coneNormal.x, meshlet->coneNormal.y, meshlet->coneNormal.z);
+								}
+								currentIndexData += indexCount;
+							} while(--numClusters != 0);
 						} else {
-							clipAndsubmitMeshletSelector[true](OCR_SPACE, indexData, meshlet->indexCount);
+							unsigned numClusters = meshlet->clusterCount;
+							auto currentIndexData = indexData;
+							do {
+								V3d coneNormal = { currentIndexData[0] / 127.f, currentIndexData[1] / 127.f, currentIndexData[2] / 127.f };
+								float costheta = dot(acp->cameraDir, coneNormal);
+								unsigned indexCount = (uint8_t&)currentIndexData[3];
+
+								currentIndexData += 4;
+								if (costheta >= acp->cosPhi) {
+									clipAndsubmitMeshletSelector[true](OCR_SPACE, currentIndexData, indexCount);
+								} else {
+									// printf("CONE CULL, %f %f, %f %f %f\n", costheta, acp->cosPhi, meshlet->coneNormal.x, meshlet->coneNormal.y, meshlet->coneNormal.z);
+								}
+								currentIndexData += indexCount;
+							} while(--numClusters != 0);
 						}
 					}
 				}
@@ -3709,7 +3830,7 @@ void defaultRenderCB(ObjPipeline *pipe, Atomic *atomic) {
 		};

 		if (doBlend || isMatFX) {
-			if (doAlphaTest) {
+			if (doAlphaTest && !doBlendMaterial) {
 				ptCallbacks.emplace_back(std::move(renderCB));
 			} else {
 				blendCallbacks.emplace_back(std::move(renderCB));
@@ -4780,17 +4901,84 @@ centerTexCoords(Geometry *g)
 	rwFree(groupIDs);
 }

-bool isDegenerate(const V3d& v1, const V3d& v2, const V3d& v3) {
+// Cluster structure
+struct ConeCluster {
+    V3d normal;               // Average normal of the cluster
+    std::vector<uint32_t> faces; // Indices of triangles in this cluster
+};
+
+V3d calculateNormal(const V3d& v1, const V3d& v2, const V3d& v3) {
    V3d u = {v2.x - v1.x, v2.y - v1.y, v2.z - v1.z};
    V3d v = {v3.x - v1.x, v3.y - v1.y, v3.z - v1.z};
 	V3d crs = cross(u, v);
-	if (length(crs) < 0.0000001f) {
-		return true;
+	if (length(crs) < 0.00001f) {
+			return {0, 0, 0};
 	} else {
-		return false;
+			return normalize(crs);
 	}
 }

+// Process mesh with cone clustering logic
+std::vector<ConeCluster> processMeshWithClustering(uint16_t* indices, uint32_t numIndices, const V3d* vertices, float cosThreshold) {
+
+    std::vector<V3d> triangleNormals(numIndices / 3); // Store triangle normals
+    std::vector<bool> processed(numIndices / 3, false);  // Track processed triangles
+
+    // Compute normals for each triangle
+    for (uint32_t i = 0; i < numIndices; i += 3) {
+        uint16_t idx1 = indices[i];
+        uint16_t idx2 = indices[i + 1];
+        uint16_t idx3 = indices[i + 2];
+
+        // Calculate and store the triangle normal
+        triangleNormals[i / 3] = calculateNormal(vertices[idx1], vertices[idx2], vertices[idx3]);
+    }
+
+    // Perform clustering
+    std::vector<ConeCluster> clusters;
+    for (size_t i = 0; i < triangleNormals.size(); ++i) {
+        if (processed[i]) continue;
+
+               if (length(triangleNormals[i]) < 0.000001f)
+                       continue;
+        // Create a new cluster
+        ConeCluster cluster;
+        cluster.normal = triangleNormals[i];
+        cluster.faces.push_back(i);
+        processed[i] = true;
+
+        // Find and add similar triangles to this cluster
+        for (size_t j = 0; j < triangleNormals.size(); ++j) {
+            if (processed[j]) continue;
+
+            float similarity = dot(cluster.normal, triangleNormals[j]);
+            if (similarity >= cosThreshold) {
+                cluster.faces.push_back(j);
+                processed[j] = true;
+            }
+        }
+
+        // Normalize the cluster normal (optional)
+        cluster.normal = normalize(cluster.normal);
+
+        // Store the cluster
+        clusters.push_back(cluster);
+    }
+
+    // // Process each cluster
+    // for (size_t c = 0; c < clusters.size(); ++c) {
+    //     std::cout << "Cluster " << c << ":\n";
+    //     std::cout << "  Average Normal: (" << clusters[c].normal.x << ", " << clusters[c].normal.y << ", " << clusters[c].normal.z << ")\n";
+    //     std::cout << "  Triangles: ";
+    //     for (auto face : clusters[c].faces) {
+    //         std::cout << face << " ";
+    //     }
+    //     std::cout << "\n";
+    // }
+
+    return clusters;
+}
+
 bool isDegenerateByIndex(uint16_t idx1, uint16_t idx2, uint16_t idx3) {
 	return idx1 == idx2 || idx1 == idx3 || idx2 == idx3;
 }
@@ -4961,10 +5149,16 @@ RwSphere calculateBoundingSphere(V3d* vertexData, size_t count) {

 	return sphere;
 }
+
+struct ConeClusterStrip {
+	triangle_stripper::primitive_group* strip;
+	ConeCluster* cluster;
+};
+
 struct meshlet {
 	std::set<uint16_t> vertices;
 	std::map<uint16_t, uint8_t> vertexToLocalIndex;
-	std::vector<triangle_stripper::primitive_group*> strips;
+	std::vector<ConeClusterStrip*> strips;
 	size_t vertexDataOffset;
 	size_t indexDataOffset;
 	size_t skinIndexDataOffset;
@@ -5022,7 +5216,8 @@ void processGeom(Geometry *geo) {
 	
 	int32 n = geo->meshHeader->numMeshes;
 	auto meshes = geo->meshHeader->getMeshes();
-	std::vector<primitive_vector> pvecs(n);
+	std::vector<std::vector<primitive_vector>> pvecs(n);
+	std::vector<std::vector<ConeCluster>> pclus(n);
 	std::vector<std::vector<meshlet>> meshMeshlets(n);

 	size_t totalIndices = 0, strips = 0,  totalTrilist = 0;
@@ -5048,6 +5243,10 @@ void processGeom(Geometry *geo) {
 		skinIndices = (uint32_t*)skin->indices;
 	}

+    float angularThreshold = 5.0f * M_PI / 180.0f;
+    float cosThreshold = std::cos(angularThreshold);
+
+

 	std::vector<size_t> canonicalIdx(geo->numVertices, SIZE_MAX);
 	for (size_t i = 0; i < geo->numVertices; i++) {
@@ -5095,6 +5294,7 @@ void processGeom(Geometry *geo) {
 		}
 	}
 	texconvf("Found %zu vertex duplicates, %.2f%%\n", dups, (float)dups/geo->numVertices*100);
+	
 	for (int meshNum = 0; meshNum < n; meshNum++) {
 		auto mesh = &meshes[meshNum];
 		
@@ -5130,28 +5330,44 @@ void processGeom(Geometry *geo) {
 			mesh->indices[i] = canonicalIdx[mesh->indices[i]];
 		}

-		{
-			indices Indices(mesh->indices, mesh->indices + mesh->numIndices);
-
+		auto clusters = processMeshWithClustering(mesh->indices, mesh->numIndices, geo->morphTargets[0].vertices, cosThreshold);
+		for (auto&& cluster: clusters) {
+			std::vector<uint16_t> idx;
+			idx.reserve(cluster.faces.size()*3);
+			for (auto && face: cluster.faces) {
+					if (isDegenerateByIndex(mesh->indices[face*3 + 0], mesh->indices[face*3 + 1], mesh->indices[face*3 + 2])) {
+							continue;
+					}
+					idx.push_back(mesh->indices[face*3 + 0]);
+					idx.push_back(mesh->indices[face*3 + 1]);
+					idx.push_back(mesh->indices[face*3 + 2]);
+			}
+			indices Indices(idx.begin(), idx.end());
+			primitive_vector PrimitivesVector;
 			tri_stripper TriStripper(Indices);

 			TriStripper.SetMinStripSize(0);
 			TriStripper.SetCacheSize(0);
 			TriStripper.SetBackwardSearch(true);

-			TriStripper.Strip(&pvecs[meshNum]);
+			TriStripper.Strip(&PrimitivesVector);
+
+			pvecs[meshNum].push_back(PrimitivesVector);
 		}
+		pclus[meshNum] = clusters;

 		mesh->indices = oldIndices;
 		mesh->numIndices = oldNumIndices;

-		for (auto &&strip: pvecs[meshNum]) {
-			totalIndices += strip.Indices.size();
-			if (strip.Type == TRIANGLES) {
-				assert(strip.Indices.size()%3==0);
-				strips += strip.Indices.size()/3;
-			} else {
-				strips ++;
+		for (auto &&strip2: pvecs[meshNum]) {
+			for (auto &&strip: strip2) {
+				totalIndices += strip.Indices.size();
+				if (strip.Type == TRIANGLES) {
+					assert(strip.Indices.size()%3==0);
+					strips += strip.Indices.size()/3;
+				} else {
+					strips ++;
+				}
 			}
 		}
 	}
@@ -5164,23 +5380,34 @@ void processGeom(Geometry *geo) {
 	size_t meshVerticesCount = 0;
 	size_t meshletIndexesCount = 0;
 	size_t meshletVerticesCount = 0;
+
+	std::vector<std::vector<ConeClusterStrip>> meshConeClusterStrips(pvecs.size());
+
 	for (int pvn = 0; pvn < pvecs.size(); pvn++) {
-		auto &&prims = pvecs[pvn];
+		auto &&prims2 = pvecs[pvn];

 		std::set<uint16_t> meshletVertices;
-		std::vector<primitive_group*> meshletStrips;
+		std::list<ConeClusterStrip*> strips;
+		std::vector<ConeClusterStrip*> meshletStrips;

-		std::list<primitive_group*> strips;
-		for (auto &&strip: prims) {
-			strips.push_back(&strip);
+		for (int cvn = 0; cvn < prims2.size(); cvn++) {
+			auto&& prims = prims2[cvn];
+
+			std::list<ConeClusterStrip> strips;
+			for (auto &&strip: prims) {
+				meshConeClusterStrips[pvn].push_back({&strip, &pclus[pvn][cvn]});
+			}
+		}
+
+		for (auto&& ccs: meshConeClusterStrips[pvn]) {
+			strips.push_back(&ccs);
 		}
-		#undef printf

 		while(strips.size()) {
 			for(;;) {
 				// pluck strip with fewest new indices

-				primitive_group* bestStrip = nullptr;
+				ConeClusterStrip* bestStrip = nullptr;

 				size_t remainingVertices = 128 - meshletVertices.size();
 				size_t bestSharedVertices = 0;
@@ -5189,7 +5416,7 @@ void processGeom(Geometry *geo) {
 					auto &&strip = *strip_ptr;
 					std::set<uint16_t> newVertices;
 					size_t sharedVertices = 0;
-					for (auto &&idx: strip.Indices) {
+					for (auto &&idx: strip.strip->Indices) {
 						if (meshletVertices.find(idx) == meshletVertices.end()) {
 							newVertices.insert(idx);
 						} else {
@@ -5212,7 +5439,7 @@ void processGeom(Geometry *geo) {

 				// add strip to meshlet
 				meshletStrips.push_back(bestStrip);
-				for (auto &&idx: bestStrip->Indices) {
+				for (auto &&idx: bestStrip->strip->Indices) {
 					meshletVertices.insert(idx);
 				}
 				strips.remove(bestStrip);
@@ -5222,7 +5449,7 @@ void processGeom(Geometry *geo) {

 			// printf("Meshlet constructed, %ld strips, %zu vertices\n", meshletStrips.size(), meshletVertices.size());
 			for (auto &&strip: meshletStrips) {
-				meshletIndexesCount += strip->Indices.size();
+				meshletIndexesCount += strip->strip->Indices.size();
 			}
 			meshletVerticesCount += meshletVertices.size();

@@ -5240,10 +5467,13 @@ void processGeom(Geometry *geo) {
 		}

 		std::set<uint16_t> meshVertices;
-		for (auto &&strip: prims) {
-			meshIndexesCount += strip.Indices.size();
-			for (auto &&idx: strip.Indices) {
-				meshVertices.insert(idx);
+		for (int cvn = 0; cvn < prims2.size(); cvn++) {
+			auto&& prims = prims2[cvn];
+			for (auto &&strip: prims) {
+				meshIndexesCount += strip.Indices.size();
+				for (auto &&idx: strip.Indices) {
+					meshVertices.insert(idx);
+				}
 			}
 		}
 		meshVerticesCount += meshVertices.size();
@@ -5260,19 +5490,15 @@ void processGeom(Geometry *geo) {
 	for (size_t i = 0; i < meshMeshlets.size(); i++) {
 		auto &&mesh = meshMeshlets[i];
 		
-		assert(mesh.size() <= 32767);
-		meshData.write<int16_t>(mesh.size());
+		assert(mesh.size() <= UINT32_MAX);
+		meshData.write<uint32_t>(mesh.size());

-		assert((meshletData.size() + meshMeshlets.size() * 4) <= 32767);
-		meshData.write<int16_t>(meshletData.size() + meshMeshlets.size() * 4);
+		assert((meshletData.size() + meshMeshlets.size() * sizeof(MeshInfo)) <= UINT32_MAX);
+		meshData.write<uint32_t>(meshletData.size() + meshMeshlets.size() * sizeof(MeshInfo));

 		for (auto && meshlet: mesh) {
 			auto boundingSphere = meshlet.calculateBoundingSphere(vertices);

-			uint32_t totalIndexes = 0;
-			for(auto&& strip: meshlet.strips) {
-				totalIndexes += strip->Indices.size();
-			}

 			// write out vertex data

@@ -5295,18 +5521,50 @@ void processGeom(Geometry *geo) {
 			// write out index data
 			meshlet.indexDataOffset = indexData.size();
 			
-			for(auto&& strip: meshlet.strips) {
-				if (strip->Type == TRIANGLES) {
-					for (size_t i = 0; i < strip->Indices.size(); i+=3) {
-						indexData.write<uint8_t>(meshlet.vertexToLocalIndex[strip->Indices[i]]);
-						indexData.write<uint8_t>(meshlet.vertexToLocalIndex[strip->Indices[i+1]]);
-						indexData.write<uint8_t>(meshlet.vertexToLocalIndex[strip->Indices[i+2]] | 128);
+			std::map<ConeCluster*, std::vector<triangle_stripper::primitive_group*>> groupedStrips;
+			for(auto&& clusterStrip: meshlet.strips) {
+				groupedStrips[clusterStrip->cluster].push_back(clusterStrip->strip);
+			}
+
+			size_t clusterCount = 0;
+			
+			for(auto&& stripGroup: groupedStrips) {
+				uint32_t totalIndexes = 256;	// force a new cluster
+				size_t totalIndexPatchPoint = 0;
+
+				for (auto&& strip: stripGroup.second) {
+					if (totalIndexes + strip->Indices.size() > 255) {
+						if (totalIndexPatchPoint) {
+							assert(totalIndexes != 0);
+							assert(totalIndexes <= 255);
+							indexData[totalIndexPatchPoint] = totalIndexes;
+						}
+						totalIndexes = 0;
+						clusterCount++;
+
+						indexData.write<int8_t>(stripGroup.first->normal.x * 127);
+						indexData.write<int8_t>(stripGroup.first->normal.y * 127);
+						indexData.write<int8_t>(stripGroup.first->normal.z * 127);
+						totalIndexPatchPoint = indexData.size();
+						indexData.write<uint8_t>(0);
 					}
-				} else {
-					for (size_t i = 0; i < strip->Indices.size(); i++) {
-						indexData.write<uint8_t>(meshlet.vertexToLocalIndex[strip->Indices[i]] | ((i + 1) == strip->Indices.size() ? 128 : 0));
+					totalIndexes += strip->Indices.size();
+					if (strip->Type == TRIANGLES) {
+						for (size_t i = 0; i < strip->Indices.size(); i+=3) {
+							indexData.write<uint8_t>(meshlet.vertexToLocalIndex[strip->Indices[i]]);
+							indexData.write<uint8_t>(meshlet.vertexToLocalIndex[strip->Indices[i+1]]);
+							indexData.write<uint8_t>(meshlet.vertexToLocalIndex[strip->Indices[i+2]] | 128);
+						}
+					} else {
+						for (size_t i = 0; i < strip->Indices.size(); i++) {
+							indexData.write<uint8_t>(meshlet.vertexToLocalIndex[strip->Indices[i]] | ((i + 1) == strip->Indices.size() ? 128 : 0));
+						}
 					}
 				}
+				assert(totalIndexPatchPoint != 0);
+				assert(totalIndexes != 0);
+				assert(totalIndexes <= 255);
+				indexData[totalIndexPatchPoint] = totalIndexes;
 			}

 			// write out skinning data
@@ -5457,15 +5715,14 @@ void processGeom(Geometry *geo) {
 			// write out meshlet data
 			meshletData.write(boundingSphere);
 			//isTextured, isNormaled, isColored, small_xyz, pad_xyz, small_uv
-			uint16_t flags = texcoorded | (normaled << 1) | (colored << 2) | (!big_vertex << 3) | (pad_xyz << 4) | (!big_uv << 5);
-			meshletData.write<uint16_t>(flags);
-			meshletData.write<uint8_t>(0);
-			//bool textured, bool normaled, bool colored, bool big_vertex, bool big_uv, bool pad_xyz
+			uint8_t flags = texcoorded | (normaled << 1) | (colored << 2) | (!big_vertex << 3) | (pad_xyz << 4) | (!big_uv << 5);
+			meshletData.write<uint8_t>(flags);
+			assert(clusterCount <= 255);
+			meshletData.write<uint8_t>(clusterCount);
 			meshletData.write<uint8_t>(vertexSize);
-			assert(meshlet.vertices.size() <= 65535);
-			meshletData.write<uint16_t>(meshlet.vertices.size());
-			assert(totalIndexes <= 65535);
-			meshletData.write<uint16_t>(totalIndexes);
+			//bool textured, bool normaled, bool colored, bool big_vertex, bool big_uv, bool pad_xyz
+			assert(meshlet.vertices.size() <= 255);
+			meshletData.write<uint8_t>(meshlet.vertices.size());
 			meshlet.rewriteOffsetVDO = meshletData.size();
 			meshletData.write<uint32_t>(meshlet.vertexDataOffset); // will be patched
 			meshlet.rewriteOffsetIDO = meshletData.size();
Author	SHA1	Message	Date
Stefanos Kornilios Mitsis Poiitidis	284f24c48c	Cluster indexes, not vertexes	2025-01-13 01:33:34 +02:00
Stefanos Kornilios Mitsis Poiitidis	e4b296c4cf	backface cone rejection	2025-01-12 18:57:25 +02:00
Stefanos Kornilios Mitsis Poiitidis	2065c9a93f	Merge branch 'skmp/fix-AllocateBoatWakeArray' into 'main' Some checks failed re3 conan+cmake / build-cmake (openal, glfw, macos-latest, gl3) (push) Has been cancelled Details re3 conan+cmake / build-cmake (openal, glfw, ubuntu-18.04, gl3) (push) Has been cancelled Details re3 conan+cmake / build-cmake (openal, glfw, windows-latest, gl3) (push) Has been cancelled Details re3 conan+cmake / build-cmake (openal, windows-latest, d3d9) (push) Has been cancelled Details re3 cmake devkitA64 (Nintendo Switch) / build-nintendo-switch (push) Has been cancelled Details re3 premake amd64 / build (Debug, win-amd64-librw_d3d9-oal) (push) Has been cancelled Details re3 premake amd64 / build (Debug, win-amd64-librw_gl3_glfw-oal) (push) Has been cancelled Details re3 premake amd64 / build (Release, win-amd64-librw_d3d9-oal) (push) Has been cancelled Details re3 premake amd64 / build (Release, win-amd64-librw_gl3_glfw-oal) (push) Has been cancelled Details re3 premake x86 / build (Debug, win-x86-librw_d3d9-mss) (push) Has been cancelled Details re3 premake x86 / build (Debug, win-x86-librw_d3d9-oal) (push) Has been cancelled Details re3 premake x86 / build (Debug, win-x86-librw_gl3_glfw-mss) (push) Has been cancelled Details re3 premake x86 / build (Debug, win-x86-librw_gl3_glfw-oal) (push) Has been cancelled Details re3 premake x86 / build (Release, win-x86-librw_d3d9-mss) (push) Has been cancelled Details re3 premake x86 / build (Release, win-x86-librw_d3d9-oal) (push) Has been cancelled Details re3 premake x86 / build (Release, win-x86-librw_gl3_glfw-mss) (push) Has been cancelled Details re3 premake x86 / build (Release, win-x86-librw_gl3_glfw-oal) (push) Has been cancelled Details re3 premake x86 / build (Vanilla, win-x86-librw_d3d9-mss) (push) Has been cancelled Details re3 premake x86 / build (Vanilla, win-x86-librw_d3d9-oal) (push) Has been cancelled Details re3 premake x86 / build (Vanilla, win-x86-librw_gl3_glfw-mss) (push) Has been cancelled Details re3 premake x86 / build (Vanilla, win-x86-librw_gl3_glfw-oal) (push) Has been cancelled Details Fix AllocateBoatWakeArray, it needs RpGeometryCreate with HAS_TRIANGLES See merge request skmp/dca3-game!12	2025-01-06 11:46:50 +00:00
Stefanos Kornilios Mitsis Poiitidis	a2caa56275	Merge branch 'skmp/material-alpha-override-pt' into 'main' Perform full blending for mat.a != 0xFF, should fade-in sprites (eg, Trees) See merge request skmp/dca3-game!11	2025-01-06 11:41:11 +00:00
Stefanos Kornilios Mitsis Poiitidis	fa65f637ad	Fix AllocateBoatWakeArray, it needs RpGeometryCreate with HAS_TRIANGLES	2025-01-06 13:40:31 +02:00
Stefanos Kornilios Mitsis Poiitidis	0a0a664428	Perform full blending for mat.a != 0xFF, should fade-in sprites (eg, Trees)	2025-01-05 15:40:18 +02:00
Stefanos Kornilios Mitsis Poiitidis	5abd0641a4	Merge branch 'skmp/fix-staring-into-the-void' into 'main' Some checks are pending re3 conan+cmake / build-cmake (openal, glfw, macos-latest, gl3) (push) Waiting to run Details re3 conan+cmake / build-cmake (openal, glfw, ubuntu-18.04, gl3) (push) Waiting to run Details re3 conan+cmake / build-cmake (openal, glfw, windows-latest, gl3) (push) Waiting to run Details re3 conan+cmake / build-cmake (openal, windows-latest, d3d9) (push) Waiting to run Details re3 cmake devkitA64 (Nintendo Switch) / build-nintendo-switch (push) Waiting to run Details re3 premake amd64 / build (Debug, win-amd64-librw_d3d9-oal) (push) Waiting to run Details re3 premake amd64 / build (Debug, win-amd64-librw_gl3_glfw-oal) (push) Waiting to run Details re3 premake amd64 / build (Release, win-amd64-librw_d3d9-oal) (push) Waiting to run Details re3 premake amd64 / build (Release, win-amd64-librw_gl3_glfw-oal) (push) Waiting to run Details re3 premake x86 / build (Debug, win-x86-librw_d3d9-mss) (push) Waiting to run Details re3 premake x86 / build (Debug, win-x86-librw_d3d9-oal) (push) Waiting to run Details re3 premake x86 / build (Debug, win-x86-librw_gl3_glfw-mss) (push) Waiting to run Details re3 premake x86 / build (Debug, win-x86-librw_gl3_glfw-oal) (push) Waiting to run Details re3 premake x86 / build (Release, win-x86-librw_d3d9-mss) (push) Waiting to run Details re3 premake x86 / build (Release, win-x86-librw_d3d9-oal) (push) Waiting to run Details re3 premake x86 / build (Release, win-x86-librw_gl3_glfw-mss) (push) Waiting to run Details re3 premake x86 / build (Release, win-x86-librw_gl3_glfw-oal) (push) Waiting to run Details re3 premake x86 / build (Vanilla, win-x86-librw_d3d9-mss) (push) Waiting to run Details re3 premake x86 / build (Vanilla, win-x86-librw_d3d9-oal) (push) Waiting to run Details re3 premake x86 / build (Vanilla, win-x86-librw_gl3_glfw-mss) (push) Waiting to run Details re3 premake x86 / build (Vanilla, win-x86-librw_gl3_glfw-oal) (push) Waiting to run Details Revert buggy optimizations in CMatrix::Rotate*. Didn't investigate much, they aren't often called Closes #8 and #6 See merge request skmp/dca3-game!9	2025-01-05 08:21:54 +00:00
Stefanos Kornilios Mitsis Poiitidis	ca866bdb2a	Fix sim build	2025-01-05 10:15:30 +02:00
Stefanos Kornilios Mitsis Poiitidis	bf0e36858e	Use fipr in CMatrix::Rotate, simplifies it so it doesn't ICE Some checks failed re3 conan+cmake / build-cmake (openal, glfw, macos-latest, gl3) (push) Has been cancelled Details re3 conan+cmake / build-cmake (openal, glfw, ubuntu-18.04, gl3) (push) Has been cancelled Details re3 conan+cmake / build-cmake (openal, glfw, windows-latest, gl3) (push) Has been cancelled Details re3 conan+cmake / build-cmake (openal, windows-latest, d3d9) (push) Has been cancelled Details re3 cmake devkitA64 (Nintendo Switch) / build-nintendo-switch (push) Has been cancelled Details re3 premake amd64 / build (Debug, win-amd64-librw_d3d9-oal) (push) Has been cancelled Details re3 premake amd64 / build (Debug, win-amd64-librw_gl3_glfw-oal) (push) Has been cancelled Details re3 premake amd64 / build (Release, win-amd64-librw_d3d9-oal) (push) Has been cancelled Details re3 premake amd64 / build (Release, win-amd64-librw_gl3_glfw-oal) (push) Has been cancelled Details re3 premake x86 / build (Debug, win-x86-librw_d3d9-mss) (push) Has been cancelled Details re3 premake x86 / build (Debug, win-x86-librw_d3d9-oal) (push) Has been cancelled Details re3 premake x86 / build (Debug, win-x86-librw_gl3_glfw-mss) (push) Has been cancelled Details re3 premake x86 / build (Debug, win-x86-librw_gl3_glfw-oal) (push) Has been cancelled Details re3 premake x86 / build (Release, win-x86-librw_d3d9-mss) (push) Has been cancelled Details re3 premake x86 / build (Release, win-x86-librw_d3d9-oal) (push) Has been cancelled Details re3 premake x86 / build (Release, win-x86-librw_gl3_glfw-mss) (push) Has been cancelled Details re3 premake x86 / build (Release, win-x86-librw_gl3_glfw-oal) (push) Has been cancelled Details re3 premake x86 / build (Vanilla, win-x86-librw_d3d9-mss) (push) Has been cancelled Details re3 premake x86 / build (Vanilla, win-x86-librw_d3d9-oal) (push) Has been cancelled Details re3 premake x86 / build (Vanilla, win-x86-librw_gl3_glfw-mss) (push) Has been cancelled Details re3 premake x86 / build (Vanilla, win-x86-librw_gl3_glfw-oal) (push) Has been cancelled Details	2025-01-05 09:19:02 +02:00
Stefanos Kornilios Mitsis Poiitidis	b90e8433da	Revert buggy optimizations in CMatrix::Rotate*. Didn't investigate much, they aren't often called	2025-01-05 09:12:56 +02:00
Stefanos Kornilios Mitsis Poiitidis	133110c15c	Merge branch 'skmp/async-pedsfx-load' into 'main' Some checks are pending re3 conan+cmake / build-cmake (openal, glfw, macos-latest, gl3) (push) Waiting to run Details re3 conan+cmake / build-cmake (openal, glfw, ubuntu-18.04, gl3) (push) Waiting to run Details re3 conan+cmake / build-cmake (openal, glfw, windows-latest, gl3) (push) Waiting to run Details re3 conan+cmake / build-cmake (openal, windows-latest, d3d9) (push) Waiting to run Details re3 cmake devkitA64 (Nintendo Switch) / build-nintendo-switch (push) Waiting to run Details re3 premake amd64 / build (Debug, win-amd64-librw_d3d9-oal) (push) Waiting to run Details re3 premake amd64 / build (Debug, win-amd64-librw_gl3_glfw-oal) (push) Waiting to run Details re3 premake amd64 / build (Release, win-amd64-librw_d3d9-oal) (push) Waiting to run Details re3 premake amd64 / build (Release, win-amd64-librw_gl3_glfw-oal) (push) Waiting to run Details re3 premake x86 / build (Debug, win-x86-librw_d3d9-mss) (push) Waiting to run Details re3 premake x86 / build (Debug, win-x86-librw_d3d9-oal) (push) Waiting to run Details re3 premake x86 / build (Debug, win-x86-librw_gl3_glfw-mss) (push) Waiting to run Details re3 premake x86 / build (Debug, win-x86-librw_gl3_glfw-oal) (push) Waiting to run Details re3 premake x86 / build (Release, win-x86-librw_d3d9-mss) (push) Waiting to run Details re3 premake x86 / build (Release, win-x86-librw_d3d9-oal) (push) Waiting to run Details re3 premake x86 / build (Release, win-x86-librw_gl3_glfw-mss) (push) Waiting to run Details re3 premake x86 / build (Release, win-x86-librw_gl3_glfw-oal) (push) Waiting to run Details re3 premake x86 / build (Vanilla, win-x86-librw_d3d9-mss) (push) Waiting to run Details re3 premake x86 / build (Vanilla, win-x86-librw_d3d9-oal) (push) Waiting to run Details re3 premake x86 / build (Vanilla, win-x86-librw_gl3_glfw-mss) (push) Waiting to run Details re3 premake x86 / build (Vanilla, win-x86-librw_gl3_glfw-oal) (push) Waiting to run Details PedSfx: Read from I/O thread, load them on the background See merge request skmp/dca3-game!10	2025-01-04 19:15:39 +00:00
Stefanos Kornilios Mitsis Poiitidis	6f71cfc458	PedSfx: Read from I/O thread, load them on the background	2025-01-04 17:46:48 +02:00
Stefanos Kornilios Mitsis Poiitidis	02d0103b20	Redirect to dca3.net	2024-12-31 09:05:08 +00:00
Stefanos Kornilios Mitsis Poiitidis	72da27ef0b	Edit README.md	2024-12-29 22:23:08 +00:00