zeux

State of Roblox graphics API across all platforms, with percentage deltas since EOY 2018. Updated December 29 2019.

Windows

API	Share
Direct3D 11+	85% (+5%)
Direct3D 10.1	8.5% (-1.5%)
Direct3D 10.0	5.5% (-2.5%)
Direct3D 9	1% (-1%)

zeux

a1.medium $ ./build/release/meshoptimizer buddha.obj
# buddha.obj: 549409 vertices, 1087474 triangles; read in 483.46 msec; indexed in 354.61 msec
Original : ACMR 1.556966 ATVR 3.081784 (NV 3.124747 AMD 3.277660 Intel 2.289651) Overfetch 2.105950 Overdraw 1.200370 in 0.00 msec
Random   : ACMR 2.999919 ATVR 5.937897 (NV 5.937882 AMD 5.937935 Intel 5.936783) Overfetch 10.839888 Overdraw 1.218682 in 33.38 msec
Cache    : ACMR 0.661465 ATVR 1.309272 (NV 1.590738 AMD 1.434356 Intel 1.138871) Overfetch 1.509062 Overdraw 1.206893 in 477.86 msec
CacheFifo: ACMR 0.689948 ATVR 1.365651 (NV 1.706663 AMD 1.516610 Intel 1.229416) Overfetch 1.552013 Overdraw 1.197034 in 146.00 msec
Overdraw : ACMR 2.776432 ATVR 5.495538 (NV 5.508446 AMD 5.527603 Intel 5.314811) Overfetch 8.624212 Overdraw 1.086317 in 209.29 msec
Fetch    : ACMR 1.556966 ATVR 3.081784 (NV 3.124747 AMD 3.277660 Intel 2.289651) Overfetch 2.105950 Overdraw 1.200370 in 25.13 msec
FetchMap : ACMR 1.556966 ATVR 3.081784 (NV 3.124747 AMD 3.277660 Intel 2.289651) O

zeux

~/pugixml$ cat ucd.cpp
#include "pugixml.hpp"
#include <malloc.h>

int main()
{
        pugi::xml_document doc;
        doc.load_file("ucd.all.grouped.xml");
        malloc_stats();
}

zeux

class RemapInterfaceIdsPass : public spvtools::opt::Pass
{
public:
    const char* name() const override { return "remap-interface-ids"; }

    RemapInterfaceIdsPass(uint32_t start_id): start_id(start_id)
    {
    }

    Status Process() override

zeux

#pragma once

#include <string>

namespace fmt {

#define KIND(X) \
	X(bool,bool) \
	X(signed char,schar) \
	X(unsigned char,uchar) \

zeux

namespace tstd
{
	template <typename T>
	class vector
	{
	public:
		typedef T* iterator;
		typedef const T* const_iterator;

		iterator begin() { return begin_; }

zeux

#include <stdio.h>
#include <stdint.h>

__declspec(noinline)
uint32_t giveMeRand()
{
	static uint32_t result = 0xdeadbeef;
	result *= 17;
	return ++result;
}

zeux

diff --git a/src/indexcodec.cpp b/src/indexcodec.cpp
index 4ab92cb..18ccd43 100644
--- a/src/indexcodec.cpp
+++ b/src/indexcodec.cpp
@@ -7,9 +7,6 @@
 // This work is based on:
 // Fabian Giesen. Simple lossless index buffer compression & follow-up. 2013
 // Conor Stokes. Vertex Cache Optimised Index Buffer Compression. 2014
-namespace meshopt
-{

zeux

Using fixed-point encoding for position (14 bits per component) and texture coordinates (12 bits per component), with 32-bit index buffer
and this vertex format:

// 12 bytes
struct PackedVertexOct
{
	unsigned short px, py, pz;
	unsigned char nu, nv; // octahedron encoded normal, aliases .pw
	unsigned short tx, ty;
};

zeux

Algorithms used for Cone* preprocess the mesh in some way, then split sequentially into 64-triangle clusters:
ConeBase: optimize mesh for transform cache
ConeSort: split mesh into large planar connected clusters, bin clusters into 6 buckets by cardinal axes, optimize each bucket for transform cache
ConeAcmr: optimize mesh for transform cache, split sequentially into variable length clusters that are relatively planar, sort clusters by avg normal
ConeCash: optimize mesh for transform cache, picking triangles that reduce ACMR but prioritizing those that keep current cluster planar
MaskBase: split sequentially into 64-triangle clusters, store a 64-bit conservative triangle mask for 6 frustums (cube faces)
ManyConeN: split sequentially into 64-triangle clusters, store N (up to 4) cones for each cluster and a cone id per triangle (2 bit)

Note that all Cone* solutions get significantly worse results with 128 or 256 triangle clusters; it doesn't matter much for Mask.
The biggest challenge with Cone* algorithms is t