Files
test2/intern/cycles/scene/scene.cpp

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

1095 lines
32 KiB
C++
Raw Permalink Normal View History

/* SPDX-FileCopyrightText: 2011-2022 Blender Foundation
*
* SPDX-License-Identifier: Apache-2.0 */
#include <cstdlib>
#include "bvh/bvh.h"
#include "device/device.h"
#include "scene/background.h"
#include "scene/bake.h"
#include "scene/camera.h"
#include "scene/curves.h"
2023-04-25 20:18:44 +02:00
#include "scene/devicescene.h"
#include "scene/film.h"
#include "scene/hair.h"
#include "scene/integrator.h"
#include "scene/light.h"
#include "scene/mesh.h"
#include "scene/object.h"
#include "scene/osl.h"
#include "scene/particles.h"
#include "scene/pointcloud.h"
#include "scene/procedural.h"
#include "scene/scene.h"
#include "scene/shader.h"
#include "scene/svm.h"
#include "scene/tables.h"
#include "scene/volume.h"
#include "session/session.h"
#include "util/guarded_allocator.h"
#include "util/log.h"
#include "util/progress.h"
CCL_NAMESPACE_BEGIN
Scene ::Scene(const SceneParams &params_, Device *device)
: name("Scene"),
default_surface(nullptr),
default_volume(nullptr),
default_light(nullptr),
default_background(nullptr),
default_empty(nullptr),
device(device),
dscene(device),
params(params_),
update_stats(nullptr),
kernels_loaded(false),
/* TODO(sergey): Check if it's indeed optimal value for the split kernel.
*/
max_closure_global(1)
{
memset((void *)&dscene.data, 0, sizeof(dscene.data));
osl_manager = make_unique<OSLManager>(device);
shader_manager = ShaderManager::create(device->info.has_osl ? params.shadingsystem :
SHADINGSYSTEM_SVM);
light_manager = make_unique<LightManager>();
geometry_manager = make_unique<GeometryManager>();
object_manager = make_unique<ObjectManager>();
image_manager = make_unique<ImageManager>(device->info);
particle_system_manager = make_unique<ParticleSystemManager>();
bake_manager = make_unique<BakeManager>();
procedural_manager = make_unique<ProceduralManager>();
volume_manager = make_unique<VolumeManager>();
/* Create nodes after managers, since create_node() can tag the managers. */
camera = create_node<Camera>();
dicing_camera = create_node<Camera>();
lookup_tables = make_unique<LookupTables>();
film = create_node<Film>();
background = create_node<Background>();
integrator = create_node<Integrator>();
ccl::Film::add_default(this);
ccl::ShaderManager::add_default(this);
}
Scene::~Scene()
{
free_memory(true);
}
void Scene::free_memory(bool final)
{
bvh.reset();
/* The order of deletion is important to make sure data is freed based on
* possible dependencies as the Nodes' reference counts are decremented in the
* destructors:
*
* - Procedurals can create and hold pointers to any other types.
* - Objects can hold pointers to Geometries and ParticleSystems
* - Lights and Geometries can hold pointers to Shaders.
*
* Similarly, we first delete all nodes and their associated device data, and
* then the managers and their associated device data.
*/
procedurals.clear();
objects.clear();
geometry.clear();
particle_systems.clear();
passes.clear();
if (device) {
camera->device_free(device, &dscene, this);
film->device_free(device, &dscene, this);
background->device_free(device, &dscene);
integrator->device_free(device, &dscene, true);
}
if (final) {
cameras.clear();
integrators.clear();
films.clear();
backgrounds.clear();
camera = nullptr;
dicing_camera = nullptr;
integrator = nullptr;
film = nullptr;
background = nullptr;
}
/* Delete Shaders after every other nodes to ensure that we do not try to
* decrement the reference count on some dangling pointer. */
shaders.clear();
/* Now that all nodes have been deleted, we can safely delete managers and
* device data. */
if (device) {
object_manager->device_free(device, &dscene, true);
geometry_manager->device_free(device, &dscene, true);
shader_manager->device_free(device, &dscene, this);
osl_manager->device_free(device, &dscene, this);
light_manager->device_free(device, &dscene);
particle_system_manager->device_free(device, &dscene);
bake_manager->device_free(device, &dscene);
volume_manager->device_free(&dscene);
if (final) {
image_manager->device_free(device);
}
else {
image_manager->device_free_builtin(device);
}
lookup_tables->device_free(device, &dscene);
}
if (final) {
lookup_tables.reset();
object_manager.reset();
geometry_manager.reset();
shader_manager.reset();
osl_manager.reset();
light_manager.reset();
particle_system_manager.reset();
image_manager.reset();
bake_manager.reset();
update_stats.reset();
procedural_manager.reset();
volume_manager.reset();
}
}
void Scene::device_update(Device *device_, Progress &progress)
{
if (!device) {
device = device_;
}
const bool print_stats = need_data_update();
bool kernels_reloaded = false;
while (true) {
if (update_stats) {
update_stats->clear();
}
const scoped_callback_timer timer([this, print_stats](double time) {
if (update_stats) {
update_stats->scene.times.add_entry({"device_update", time});
if (print_stats) {
printf("Update statistics:\n%s\n", update_stats->full_report().c_str());
}
}
});
/* The order of updates is important, because there's dependencies between
* the different managers, using data computed by previous managers. */
if (film->update_lightgroups(this)) {
light_manager->tag_update(this, ccl::LightManager::LIGHT_MODIFIED);
object_manager->tag_update(this, ccl::ObjectManager::OBJECT_MODIFIED);
background->tag_modified();
}
if (film->exposure_is_modified()) {
integrator->tag_modified();
}
/* Compile shaders and get information about features they used. */
progress.set_status("Updating Shaders");
osl_manager->device_update_pre(device, this);
shader_manager->device_update_pre(device, &dscene, this, progress);
if (progress.get_cancel() || device->have_error()) {
return;
}
/* Passes. After shader manager as this depends on the shaders. */
film->update_passes(this);
/* Update kernel features. After shaders and passes since those affect features. */
update_kernel_features();
/* Load render kernels, before uploading most data to the GPU, and before displacement and
* background light need to run kernels.
*
* Do it outside of the scene mutex since the heavy part of the loading (i.e. kernel
* compilation) does not depend on the scene and some other functionality (like display
* driver) might be waiting on the scene mutex to synchronize display pass.
*
* This does mean the scene might have gotten updated in the meantime, in which case
* we have to redo the first part of the scene update. */
const uint kernel_features = dscene.data.kernel_features;
scene_updated_while_loading_kernels = false;
if (!kernels_loaded || loaded_kernel_features != kernel_features) {
mutex.unlock();
kernels_reloaded |= load_kernels(progress);
mutex.lock();
}
if (progress.get_cancel() || device->have_error()) {
return;
}
if (!scene_updated_while_loading_kernels) {
break;
}
}
/* Upload shaders to GPU and compile OSL kernels, after kernels have been loaded. */
shader_manager->device_update_post(device, &dscene, this, progress);
osl_manager->device_update_post(device, this, progress, kernels_reloaded);
if (progress.get_cancel() || device->have_error()) {
return;
}
procedural_manager->update(this, progress);
if (progress.get_cancel()) {
return;
}
progress.set_status("Updating Background");
background->device_update(device, &dscene, this);
if (progress.get_cancel() || device->have_error()) {
return;
}
/* Camera will be used by adaptive subdivision, so do early. */
progress.set_status("Updating Camera");
camera->device_update(device, &dscene, this);
if (progress.get_cancel() || device->have_error()) {
return;
}
geometry_manager->device_update_preprocess(device, this, progress);
if (progress.get_cancel() || device->have_error()) {
return;
}
/* Update objects after geometry preprocessing. */
progress.set_status("Updating Objects");
object_manager->device_update(device, &dscene, this, progress);
if (progress.get_cancel() || device->have_error()) {
return;
}
progress.set_status("Updating Particle Systems");
particle_system_manager->device_update(device, &dscene, this, progress);
if (progress.get_cancel() || device->have_error()) {
return;
}
/* Camera and shaders must be ready here for adaptive subdivision and displacement. */
progress.set_status("Updating Meshes");
geometry_manager->device_update(device, &dscene, this, progress);
if (progress.get_cancel() || device->have_error()) {
return;
}
/* Update object flags with final geometry. */
progress.set_status("Updating Objects Flags");
object_manager->device_update_flags(device, &dscene, this, progress);
if (progress.get_cancel() || device->have_error()) {
return;
}
/* Update BVH primitive objects with final geometry. */
progress.set_status("Updating Primitive Offsets");
object_manager->device_update_prim_offsets(device, &dscene, this);
if (progress.get_cancel() || device->have_error()) {
return;
}
/* Images last, as they should be more likely to use host memory fallback than geometry.
* Some images may have been uploaded early for displacement already at this point. */
progress.set_status("Updating Images");
image_manager->device_update(device, this, progress);
if (progress.get_cancel() || device->have_error()) {
return;
}
/* Evaluate volume shader to build volume octrees. */
progress.set_status("Updating Volume");
volume_manager->device_update(device, &dscene, this, progress);
if (progress.get_cancel() || device->have_error()) {
return;
}
progress.set_status("Updating Camera Volume");
camera->device_update_volume(device, &dscene, this);
if (progress.get_cancel() || device->have_error()) {
return;
}
2018-11-25 08:01:14 +11:00
progress.set_status("Updating Lookup Tables");
lookup_tables->device_update(device, &dscene, this);
if (progress.get_cancel() || device->have_error()) {
return;
}
/* Light manager needs shaders and final meshes for triangles in light tree. */
progress.set_status("Updating Lights");
light_manager->device_update(device, &dscene, this, progress);
if (progress.get_cancel() || device->have_error()) {
return;
}
progress.set_status("Updating Integrator");
integrator->device_update(device, &dscene, this);
if (progress.get_cancel() || device->have_error()) {
return;
}
progress.set_status("Updating Film");
film->device_update(device, &dscene, this);
if (progress.get_cancel() || device->have_error()) {
return;
}
/* Update lookup tables a second time for film tables. */
progress.set_status("Updating Lookup Tables");
lookup_tables->device_update(device, &dscene, this);
if (progress.get_cancel() || device->have_error()) {
return;
}
progress.set_status("Updating Baking");
bake_manager->device_update(device, &dscene, this, progress);
if (progress.get_cancel() || device->have_error()) {
return;
}
if (device->have_error() == false) {
dscene.data.volume_stack_size = get_volume_stack_size();
progress.set_status("Updating Device", "Writing constant memory");
device->const_copy_to("data", &dscene.data, sizeof(dscene.data));
}
device->optimize_for_scene(this);
if (print_stats) {
const size_t mem_used = util_guarded_get_mem_used();
const size_t mem_peak = util_guarded_get_mem_peak();
LOG_INFO << "System memory statistics after full device sync:\n"
<< " Usage: " << string_human_readable_number(mem_used) << " ("
<< string_human_readable_size(mem_used) << ")\n"
<< " Peak: " << string_human_readable_number(mem_peak) << " ("
<< string_human_readable_size(mem_peak) << ")";
}
}
Scene::MotionType Scene::need_motion() const
{
if (integrator->get_motion_blur()) {
return MOTION_BLUR;
}
if (Pass::contains(passes, PASS_MOTION)) {
return MOTION_PASS;
}
return MOTION_NONE;
}
float Scene::motion_shutter_time()
{
if (need_motion() == Scene::MOTION_PASS) {
return 2.0f;
}
return camera->get_shuttertime();
}
bool Scene::need_global_attribute(AttributeStandard std)
{
if (std == ATTR_STD_UV) {
return Pass::contains(passes, PASS_UV);
}
if (std == ATTR_STD_MOTION_VERTEX_POSITION) {
return need_motion() != MOTION_NONE;
}
if (std == ATTR_STD_MOTION_VERTEX_NORMAL) {
return need_motion() == MOTION_BLUR;
}
if (std == ATTR_STD_VOLUME_VELOCITY || std == ATTR_STD_VOLUME_VELOCITY_X ||
std == ATTR_STD_VOLUME_VELOCITY_Y || std == ATTR_STD_VOLUME_VELOCITY_Z)
{
return need_motion() != MOTION_NONE;
}
return false;
}
void Scene::need_global_attributes(AttributeRequestSet &attributes)
{
for (int std = ATTR_STD_NONE; std < ATTR_STD_NUM; std++) {
if (need_global_attribute((AttributeStandard)std)) {
attributes.add((AttributeStandard)std);
}
}
}
bool Scene::need_update()
{
return (need_reset() || film->is_modified());
}
bool Scene::need_data_update()
{
return (background->is_modified() || image_manager->need_update() ||
object_manager->need_update() || geometry_manager->need_update() ||
light_manager->need_update() || lookup_tables->need_update() ||
integrator->is_modified() || shader_manager->need_update() ||
particle_system_manager->need_update() || bake_manager->need_update() ||
film->is_modified() || procedural_manager->need_update());
}
bool Scene::need_reset(const bool check_camera)
{
return need_data_update() || (check_camera && camera->is_modified());
}
void Scene::reset()
{
osl_manager->reset(this);
ShaderManager::add_default(this);
/* ensure all objects are updated */
camera->tag_modified();
dicing_camera->tag_modified();
film->tag_modified();
background->tag_modified();
background->tag_update(this);
integrator->tag_update(this, Integrator::UPDATE_ALL);
object_manager->tag_update(this, ObjectManager::UPDATE_ALL);
geometry_manager->tag_update(this, GeometryManager::UPDATE_ALL);
light_manager->tag_update(this, LightManager::UPDATE_ALL);
particle_system_manager->tag_update(this);
procedural_manager->tag_update();
}
void Scene::device_free()
{
free_memory(false);
}
void Scene::collect_statistics(RenderStats *stats)
{
geometry_manager->collect_statistics(this, stats);
image_manager->collect_statistics(stats);
}
void Scene::enable_update_stats()
{
if (!update_stats) {
update_stats = make_unique<SceneUpdateStats>();
}
}
void Scene::update_kernel_features()
{
if (!need_update()) {
return;
}
/* These features are not being tweaked as often as shaders,
* so could be done selective magic for the viewport as well. */
uint kernel_features = shader_manager->get_kernel_features(this);
const bool use_motion = need_motion() == Scene::MotionType::MOTION_BLUR;
kernel_features |= KERNEL_FEATURE_PATH_TRACING;
Cycles: approximate shadow caustics using manifold next event estimation This adds support for selective rendering of caustics in shadows of refractive objects. Example uses are rendering of underwater caustics and eye caustics. This is based on "Manifold Next Event Estimation", a method developed for production rendering. The idea is to selectively enable shadow caustics on a few objects in the scene where they have a big visual impact, without impacting render performance for the rest of the scene. The Shadow Caustic option must be manually enabled on light, caustic receiver and caster objects. For such light paths, the Filter Glossy option will be ignored and replaced by sharp caustics. Currently this method has a various limitations: * Only caustics in shadows of refractive objects work, which means no caustics from reflection or caustics that outside shadows. Only up to 4 refractive caustic bounces are supported. * Caustic caster objects should have smooth normals. * Not currently support for Metal GPU rendering. In the future this method may be extended for more general caustics. TECHNICAL DETAILS This code adds manifold next event estimation through refractive surface(s) as a new sampling technique for direct lighting, i.e. finding the point on the refractive surface(s) along the path to a light sample, which satisfies Fermat's principle for a given microfacet normal and the path's end points. This technique involves walking on the "specular manifold" using a pseudo newton solver. Such a manifold is defined by the specular constraint matrix from the manifold exploration framework [2]. For each refractive interface, this constraint is defined by enforcing that the generalized half-vector projection onto the interface local tangent plane is null. The newton solver guides the walk by linearizing the manifold locally before reprojecting the linear solution onto the refractive surface. See paper [1] for more details about the technique itself and [3] for the half-vector light transport formulation, from which it is derived. [1] Manifold Next Event Estimation Johannes Hanika, Marc Droske, and Luca Fascione. 2015. Comput. Graph. Forum 34, 4 (July 2015), 87–97. https://jo.dreggn.org/home/2015_mnee.pdf [2] Manifold exploration: a Markov Chain Monte Carlo technique for rendering scenes with difficult specular transport Wenzel Jakob and Steve Marschner. 2012. ACM Trans. Graph. 31, 4, Article 58 (July 2012), 13 pages. https://www.cs.cornell.edu/projects/manifolds-sg12/ [3] The Natural-Constraint Representation of the Path Space for Efficient Light Transport Simulation. Anton S. Kaplanyan, Johannes Hanika, and Carsten Dachsbacher. 2014. ACM Trans. Graph. 33, 4, Article 102 (July 2014), 13 pages. https://cg.ivd.kit.edu/english/HSLT.php The code for this samping technique was inserted at the light sampling stage (direct lighting). If the walk is successful, it turns off path regularization using a specialized flag in the path state (PATH_MNEE_SUCCESS). This flag tells the integrator not to blur the brdf roughness further down the path (in a child ray created from BSDF sampling). In addition, using a cascading mechanism of flag values, we cull connections to caustic lights for this and children rays, which should be resolved through MNEE. This mechanism also cancels the MIS bsdf counter part at the casutic receiver depth, in essence leaving MNEE as the only sampling technique from receivers through refractive casters to caustic lights. This choice might not be optimal when the light gets large wrt to the receiver, though this is usually not when you want to use MNEE. This connection culling strategy removes a fair amount of fireflies, at the cost of introducing a slight bias. Because of the selective nature of the culling mechanism, reflective caustics still benefit from the native path regularization, which further removes fireflies on other surfaces (bouncing light off casters). Differential Revision: https://developer.blender.org/D13533
2022-04-01 15:44:24 +02:00
/* Track the max prim count in case the backend needs to rebuild BVHs or
* kernels to support different limits. */
size_t kernel_max_prim_count = 0;
2022-06-30 12:14:22 +10:00
/* Figure out whether the scene will use shader ray-trace we need at least
Cycles: approximate shadow caustics using manifold next event estimation This adds support for selective rendering of caustics in shadows of refractive objects. Example uses are rendering of underwater caustics and eye caustics. This is based on "Manifold Next Event Estimation", a method developed for production rendering. The idea is to selectively enable shadow caustics on a few objects in the scene where they have a big visual impact, without impacting render performance for the rest of the scene. The Shadow Caustic option must be manually enabled on light, caustic receiver and caster objects. For such light paths, the Filter Glossy option will be ignored and replaced by sharp caustics. Currently this method has a various limitations: * Only caustics in shadows of refractive objects work, which means no caustics from reflection or caustics that outside shadows. Only up to 4 refractive caustic bounces are supported. * Caustic caster objects should have smooth normals. * Not currently support for Metal GPU rendering. In the future this method may be extended for more general caustics. TECHNICAL DETAILS This code adds manifold next event estimation through refractive surface(s) as a new sampling technique for direct lighting, i.e. finding the point on the refractive surface(s) along the path to a light sample, which satisfies Fermat's principle for a given microfacet normal and the path's end points. This technique involves walking on the "specular manifold" using a pseudo newton solver. Such a manifold is defined by the specular constraint matrix from the manifold exploration framework [2]. For each refractive interface, this constraint is defined by enforcing that the generalized half-vector projection onto the interface local tangent plane is null. The newton solver guides the walk by linearizing the manifold locally before reprojecting the linear solution onto the refractive surface. See paper [1] for more details about the technique itself and [3] for the half-vector light transport formulation, from which it is derived. [1] Manifold Next Event Estimation Johannes Hanika, Marc Droske, and Luca Fascione. 2015. Comput. Graph. Forum 34, 4 (July 2015), 87–97. https://jo.dreggn.org/home/2015_mnee.pdf [2] Manifold exploration: a Markov Chain Monte Carlo technique for rendering scenes with difficult specular transport Wenzel Jakob and Steve Marschner. 2012. ACM Trans. Graph. 31, 4, Article 58 (July 2012), 13 pages. https://www.cs.cornell.edu/projects/manifolds-sg12/ [3] The Natural-Constraint Representation of the Path Space for Efficient Light Transport Simulation. Anton S. Kaplanyan, Johannes Hanika, and Carsten Dachsbacher. 2014. ACM Trans. Graph. 33, 4, Article 102 (July 2014), 13 pages. https://cg.ivd.kit.edu/english/HSLT.php The code for this samping technique was inserted at the light sampling stage (direct lighting). If the walk is successful, it turns off path regularization using a specialized flag in the path state (PATH_MNEE_SUCCESS). This flag tells the integrator not to blur the brdf roughness further down the path (in a child ray created from BSDF sampling). In addition, using a cascading mechanism of flag values, we cull connections to caustic lights for this and children rays, which should be resolved through MNEE. This mechanism also cancels the MIS bsdf counter part at the casutic receiver depth, in essence leaving MNEE as the only sampling technique from receivers through refractive casters to caustic lights. This choice might not be optimal when the light gets large wrt to the receiver, though this is usually not when you want to use MNEE. This connection culling strategy removes a fair amount of fireflies, at the cost of introducing a slight bias. Because of the selective nature of the culling mechanism, reflective caustics still benefit from the native path regularization, which further removes fireflies on other surfaces (bouncing light off casters). Differential Revision: https://developer.blender.org/D13533
2022-04-01 15:44:24 +02:00
* one caustic light, one caustic caster and one caustic receiver to use
2022-06-30 12:14:22 +10:00
* and enable the MNEE code path. */
Cycles: approximate shadow caustics using manifold next event estimation This adds support for selective rendering of caustics in shadows of refractive objects. Example uses are rendering of underwater caustics and eye caustics. This is based on "Manifold Next Event Estimation", a method developed for production rendering. The idea is to selectively enable shadow caustics on a few objects in the scene where they have a big visual impact, without impacting render performance for the rest of the scene. The Shadow Caustic option must be manually enabled on light, caustic receiver and caster objects. For such light paths, the Filter Glossy option will be ignored and replaced by sharp caustics. Currently this method has a various limitations: * Only caustics in shadows of refractive objects work, which means no caustics from reflection or caustics that outside shadows. Only up to 4 refractive caustic bounces are supported. * Caustic caster objects should have smooth normals. * Not currently support for Metal GPU rendering. In the future this method may be extended for more general caustics. TECHNICAL DETAILS This code adds manifold next event estimation through refractive surface(s) as a new sampling technique for direct lighting, i.e. finding the point on the refractive surface(s) along the path to a light sample, which satisfies Fermat's principle for a given microfacet normal and the path's end points. This technique involves walking on the "specular manifold" using a pseudo newton solver. Such a manifold is defined by the specular constraint matrix from the manifold exploration framework [2]. For each refractive interface, this constraint is defined by enforcing that the generalized half-vector projection onto the interface local tangent plane is null. The newton solver guides the walk by linearizing the manifold locally before reprojecting the linear solution onto the refractive surface. See paper [1] for more details about the technique itself and [3] for the half-vector light transport formulation, from which it is derived. [1] Manifold Next Event Estimation Johannes Hanika, Marc Droske, and Luca Fascione. 2015. Comput. Graph. Forum 34, 4 (July 2015), 87–97. https://jo.dreggn.org/home/2015_mnee.pdf [2] Manifold exploration: a Markov Chain Monte Carlo technique for rendering scenes with difficult specular transport Wenzel Jakob and Steve Marschner. 2012. ACM Trans. Graph. 31, 4, Article 58 (July 2012), 13 pages. https://www.cs.cornell.edu/projects/manifolds-sg12/ [3] The Natural-Constraint Representation of the Path Space for Efficient Light Transport Simulation. Anton S. Kaplanyan, Johannes Hanika, and Carsten Dachsbacher. 2014. ACM Trans. Graph. 33, 4, Article 102 (July 2014), 13 pages. https://cg.ivd.kit.edu/english/HSLT.php The code for this samping technique was inserted at the light sampling stage (direct lighting). If the walk is successful, it turns off path regularization using a specialized flag in the path state (PATH_MNEE_SUCCESS). This flag tells the integrator not to blur the brdf roughness further down the path (in a child ray created from BSDF sampling). In addition, using a cascading mechanism of flag values, we cull connections to caustic lights for this and children rays, which should be resolved through MNEE. This mechanism also cancels the MIS bsdf counter part at the casutic receiver depth, in essence leaving MNEE as the only sampling technique from receivers through refractive casters to caustic lights. This choice might not be optimal when the light gets large wrt to the receiver, though this is usually not when you want to use MNEE. This connection culling strategy removes a fair amount of fireflies, at the cost of introducing a slight bias. Because of the selective nature of the culling mechanism, reflective caustics still benefit from the native path regularization, which further removes fireflies on other surfaces (bouncing light off casters). Differential Revision: https://developer.blender.org/D13533
2022-04-01 15:44:24 +02:00
bool has_caustics_receiver = false;
bool has_caustics_caster = false;
bool has_caustics_light = false;
for (Object *object : objects) {
Cycles: approximate shadow caustics using manifold next event estimation This adds support for selective rendering of caustics in shadows of refractive objects. Example uses are rendering of underwater caustics and eye caustics. This is based on "Manifold Next Event Estimation", a method developed for production rendering. The idea is to selectively enable shadow caustics on a few objects in the scene where they have a big visual impact, without impacting render performance for the rest of the scene. The Shadow Caustic option must be manually enabled on light, caustic receiver and caster objects. For such light paths, the Filter Glossy option will be ignored and replaced by sharp caustics. Currently this method has a various limitations: * Only caustics in shadows of refractive objects work, which means no caustics from reflection or caustics that outside shadows. Only up to 4 refractive caustic bounces are supported. * Caustic caster objects should have smooth normals. * Not currently support for Metal GPU rendering. In the future this method may be extended for more general caustics. TECHNICAL DETAILS This code adds manifold next event estimation through refractive surface(s) as a new sampling technique for direct lighting, i.e. finding the point on the refractive surface(s) along the path to a light sample, which satisfies Fermat's principle for a given microfacet normal and the path's end points. This technique involves walking on the "specular manifold" using a pseudo newton solver. Such a manifold is defined by the specular constraint matrix from the manifold exploration framework [2]. For each refractive interface, this constraint is defined by enforcing that the generalized half-vector projection onto the interface local tangent plane is null. The newton solver guides the walk by linearizing the manifold locally before reprojecting the linear solution onto the refractive surface. See paper [1] for more details about the technique itself and [3] for the half-vector light transport formulation, from which it is derived. [1] Manifold Next Event Estimation Johannes Hanika, Marc Droske, and Luca Fascione. 2015. Comput. Graph. Forum 34, 4 (July 2015), 87–97. https://jo.dreggn.org/home/2015_mnee.pdf [2] Manifold exploration: a Markov Chain Monte Carlo technique for rendering scenes with difficult specular transport Wenzel Jakob and Steve Marschner. 2012. ACM Trans. Graph. 31, 4, Article 58 (July 2012), 13 pages. https://www.cs.cornell.edu/projects/manifolds-sg12/ [3] The Natural-Constraint Representation of the Path Space for Efficient Light Transport Simulation. Anton S. Kaplanyan, Johannes Hanika, and Carsten Dachsbacher. 2014. ACM Trans. Graph. 33, 4, Article 102 (July 2014), 13 pages. https://cg.ivd.kit.edu/english/HSLT.php The code for this samping technique was inserted at the light sampling stage (direct lighting). If the walk is successful, it turns off path regularization using a specialized flag in the path state (PATH_MNEE_SUCCESS). This flag tells the integrator not to blur the brdf roughness further down the path (in a child ray created from BSDF sampling). In addition, using a cascading mechanism of flag values, we cull connections to caustic lights for this and children rays, which should be resolved through MNEE. This mechanism also cancels the MIS bsdf counter part at the casutic receiver depth, in essence leaving MNEE as the only sampling technique from receivers through refractive casters to caustic lights. This choice might not be optimal when the light gets large wrt to the receiver, though this is usually not when you want to use MNEE. This connection culling strategy removes a fair amount of fireflies, at the cost of introducing a slight bias. Because of the selective nature of the culling mechanism, reflective caustics still benefit from the native path regularization, which further removes fireflies on other surfaces (bouncing light off casters). Differential Revision: https://developer.blender.org/D13533
2022-04-01 15:44:24 +02:00
if (object->get_is_caustics_caster()) {
has_caustics_caster = true;
}
else if (object->get_is_caustics_receiver()) {
has_caustics_receiver = true;
}
Geometry *geom = object->get_geometry();
if (use_motion) {
if (object->use_motion() || geom->get_use_motion_blur()) {
kernel_features |= KERNEL_FEATURE_OBJECT_MOTION;
}
}
if (object->get_is_shadow_catcher() && !geom->is_light()) {
kernel_features |= KERNEL_FEATURE_SHADOW_CATCHER;
}
if (geom->is_hair()) {
const Hair *hair = static_cast<const Hair *>(geom);
kernel_features |= (hair->curve_shape == CURVE_RIBBON) ? KERNEL_FEATURE_HAIR_RIBBON :
KERNEL_FEATURE_HAIR_THICK;
kernel_max_prim_count = max(kernel_max_prim_count, hair->num_segments());
}
else if (geom->is_pointcloud()) {
kernel_features |= KERNEL_FEATURE_POINTCLOUD;
kernel_max_prim_count = max(kernel_max_prim_count,
static_cast<PointCloud *>(geom)->num_points());
}
else if (geom->is_mesh()) {
kernel_max_prim_count = max(kernel_max_prim_count,
static_cast<Mesh *>(geom)->num_triangles());
}
else if (geom->is_light()) {
const Light *light = static_cast<const Light *>(object->get_geometry());
if (light->get_use_caustics()) {
has_caustics_light = true;
}
}
if (object->has_light_linking()) {
kernel_features |= KERNEL_FEATURE_LIGHT_LINKING;
}
if (object->has_shadow_linking()) {
kernel_features |= KERNEL_FEATURE_SHADOW_LINKING;
}
}
Cycles: approximate shadow caustics using manifold next event estimation This adds support for selective rendering of caustics in shadows of refractive objects. Example uses are rendering of underwater caustics and eye caustics. This is based on "Manifold Next Event Estimation", a method developed for production rendering. The idea is to selectively enable shadow caustics on a few objects in the scene where they have a big visual impact, without impacting render performance for the rest of the scene. The Shadow Caustic option must be manually enabled on light, caustic receiver and caster objects. For such light paths, the Filter Glossy option will be ignored and replaced by sharp caustics. Currently this method has a various limitations: * Only caustics in shadows of refractive objects work, which means no caustics from reflection or caustics that outside shadows. Only up to 4 refractive caustic bounces are supported. * Caustic caster objects should have smooth normals. * Not currently support for Metal GPU rendering. In the future this method may be extended for more general caustics. TECHNICAL DETAILS This code adds manifold next event estimation through refractive surface(s) as a new sampling technique for direct lighting, i.e. finding the point on the refractive surface(s) along the path to a light sample, which satisfies Fermat's principle for a given microfacet normal and the path's end points. This technique involves walking on the "specular manifold" using a pseudo newton solver. Such a manifold is defined by the specular constraint matrix from the manifold exploration framework [2]. For each refractive interface, this constraint is defined by enforcing that the generalized half-vector projection onto the interface local tangent plane is null. The newton solver guides the walk by linearizing the manifold locally before reprojecting the linear solution onto the refractive surface. See paper [1] for more details about the technique itself and [3] for the half-vector light transport formulation, from which it is derived. [1] Manifold Next Event Estimation Johannes Hanika, Marc Droske, and Luca Fascione. 2015. Comput. Graph. Forum 34, 4 (July 2015), 87–97. https://jo.dreggn.org/home/2015_mnee.pdf [2] Manifold exploration: a Markov Chain Monte Carlo technique for rendering scenes with difficult specular transport Wenzel Jakob and Steve Marschner. 2012. ACM Trans. Graph. 31, 4, Article 58 (July 2012), 13 pages. https://www.cs.cornell.edu/projects/manifolds-sg12/ [3] The Natural-Constraint Representation of the Path Space for Efficient Light Transport Simulation. Anton S. Kaplanyan, Johannes Hanika, and Carsten Dachsbacher. 2014. ACM Trans. Graph. 33, 4, Article 102 (July 2014), 13 pages. https://cg.ivd.kit.edu/english/HSLT.php The code for this samping technique was inserted at the light sampling stage (direct lighting). If the walk is successful, it turns off path regularization using a specialized flag in the path state (PATH_MNEE_SUCCESS). This flag tells the integrator not to blur the brdf roughness further down the path (in a child ray created from BSDF sampling). In addition, using a cascading mechanism of flag values, we cull connections to caustic lights for this and children rays, which should be resolved through MNEE. This mechanism also cancels the MIS bsdf counter part at the casutic receiver depth, in essence leaving MNEE as the only sampling technique from receivers through refractive casters to caustic lights. This choice might not be optimal when the light gets large wrt to the receiver, though this is usually not when you want to use MNEE. This connection culling strategy removes a fair amount of fireflies, at the cost of introducing a slight bias. Because of the selective nature of the culling mechanism, reflective caustics still benefit from the native path regularization, which further removes fireflies on other surfaces (bouncing light off casters). Differential Revision: https://developer.blender.org/D13533
2022-04-01 15:44:24 +02:00
dscene.data.integrator.use_caustics = false;
if (device->info.has_mnee && has_caustics_caster && has_caustics_receiver && has_caustics_light)
{
Cycles: approximate shadow caustics using manifold next event estimation This adds support for selective rendering of caustics in shadows of refractive objects. Example uses are rendering of underwater caustics and eye caustics. This is based on "Manifold Next Event Estimation", a method developed for production rendering. The idea is to selectively enable shadow caustics on a few objects in the scene where they have a big visual impact, without impacting render performance for the rest of the scene. The Shadow Caustic option must be manually enabled on light, caustic receiver and caster objects. For such light paths, the Filter Glossy option will be ignored and replaced by sharp caustics. Currently this method has a various limitations: * Only caustics in shadows of refractive objects work, which means no caustics from reflection or caustics that outside shadows. Only up to 4 refractive caustic bounces are supported. * Caustic caster objects should have smooth normals. * Not currently support for Metal GPU rendering. In the future this method may be extended for more general caustics. TECHNICAL DETAILS This code adds manifold next event estimation through refractive surface(s) as a new sampling technique for direct lighting, i.e. finding the point on the refractive surface(s) along the path to a light sample, which satisfies Fermat's principle for a given microfacet normal and the path's end points. This technique involves walking on the "specular manifold" using a pseudo newton solver. Such a manifold is defined by the specular constraint matrix from the manifold exploration framework [2]. For each refractive interface, this constraint is defined by enforcing that the generalized half-vector projection onto the interface local tangent plane is null. The newton solver guides the walk by linearizing the manifold locally before reprojecting the linear solution onto the refractive surface. See paper [1] for more details about the technique itself and [3] for the half-vector light transport formulation, from which it is derived. [1] Manifold Next Event Estimation Johannes Hanika, Marc Droske, and Luca Fascione. 2015. Comput. Graph. Forum 34, 4 (July 2015), 87–97. https://jo.dreggn.org/home/2015_mnee.pdf [2] Manifold exploration: a Markov Chain Monte Carlo technique for rendering scenes with difficult specular transport Wenzel Jakob and Steve Marschner. 2012. ACM Trans. Graph. 31, 4, Article 58 (July 2012), 13 pages. https://www.cs.cornell.edu/projects/manifolds-sg12/ [3] The Natural-Constraint Representation of the Path Space for Efficient Light Transport Simulation. Anton S. Kaplanyan, Johannes Hanika, and Carsten Dachsbacher. 2014. ACM Trans. Graph. 33, 4, Article 102 (July 2014), 13 pages. https://cg.ivd.kit.edu/english/HSLT.php The code for this samping technique was inserted at the light sampling stage (direct lighting). If the walk is successful, it turns off path regularization using a specialized flag in the path state (PATH_MNEE_SUCCESS). This flag tells the integrator not to blur the brdf roughness further down the path (in a child ray created from BSDF sampling). In addition, using a cascading mechanism of flag values, we cull connections to caustic lights for this and children rays, which should be resolved through MNEE. This mechanism also cancels the MIS bsdf counter part at the casutic receiver depth, in essence leaving MNEE as the only sampling technique from receivers through refractive casters to caustic lights. This choice might not be optimal when the light gets large wrt to the receiver, though this is usually not when you want to use MNEE. This connection culling strategy removes a fair amount of fireflies, at the cost of introducing a slight bias. Because of the selective nature of the culling mechanism, reflective caustics still benefit from the native path regularization, which further removes fireflies on other surfaces (bouncing light off casters). Differential Revision: https://developer.blender.org/D13533
2022-04-01 15:44:24 +02:00
dscene.data.integrator.use_caustics = true;
kernel_features |= KERNEL_FEATURE_MNEE;
Cycles: approximate shadow caustics using manifold next event estimation This adds support for selective rendering of caustics in shadows of refractive objects. Example uses are rendering of underwater caustics and eye caustics. This is based on "Manifold Next Event Estimation", a method developed for production rendering. The idea is to selectively enable shadow caustics on a few objects in the scene where they have a big visual impact, without impacting render performance for the rest of the scene. The Shadow Caustic option must be manually enabled on light, caustic receiver and caster objects. For such light paths, the Filter Glossy option will be ignored and replaced by sharp caustics. Currently this method has a various limitations: * Only caustics in shadows of refractive objects work, which means no caustics from reflection or caustics that outside shadows. Only up to 4 refractive caustic bounces are supported. * Caustic caster objects should have smooth normals. * Not currently support for Metal GPU rendering. In the future this method may be extended for more general caustics. TECHNICAL DETAILS This code adds manifold next event estimation through refractive surface(s) as a new sampling technique for direct lighting, i.e. finding the point on the refractive surface(s) along the path to a light sample, which satisfies Fermat's principle for a given microfacet normal and the path's end points. This technique involves walking on the "specular manifold" using a pseudo newton solver. Such a manifold is defined by the specular constraint matrix from the manifold exploration framework [2]. For each refractive interface, this constraint is defined by enforcing that the generalized half-vector projection onto the interface local tangent plane is null. The newton solver guides the walk by linearizing the manifold locally before reprojecting the linear solution onto the refractive surface. See paper [1] for more details about the technique itself and [3] for the half-vector light transport formulation, from which it is derived. [1] Manifold Next Event Estimation Johannes Hanika, Marc Droske, and Luca Fascione. 2015. Comput. Graph. Forum 34, 4 (July 2015), 87–97. https://jo.dreggn.org/home/2015_mnee.pdf [2] Manifold exploration: a Markov Chain Monte Carlo technique for rendering scenes with difficult specular transport Wenzel Jakob and Steve Marschner. 2012. ACM Trans. Graph. 31, 4, Article 58 (July 2012), 13 pages. https://www.cs.cornell.edu/projects/manifolds-sg12/ [3] The Natural-Constraint Representation of the Path Space for Efficient Light Transport Simulation. Anton S. Kaplanyan, Johannes Hanika, and Carsten Dachsbacher. 2014. ACM Trans. Graph. 33, 4, Article 102 (July 2014), 13 pages. https://cg.ivd.kit.edu/english/HSLT.php The code for this samping technique was inserted at the light sampling stage (direct lighting). If the walk is successful, it turns off path regularization using a specialized flag in the path state (PATH_MNEE_SUCCESS). This flag tells the integrator not to blur the brdf roughness further down the path (in a child ray created from BSDF sampling). In addition, using a cascading mechanism of flag values, we cull connections to caustic lights for this and children rays, which should be resolved through MNEE. This mechanism also cancels the MIS bsdf counter part at the casutic receiver depth, in essence leaving MNEE as the only sampling technique from receivers through refractive casters to caustic lights. This choice might not be optimal when the light gets large wrt to the receiver, though this is usually not when you want to use MNEE. This connection culling strategy removes a fair amount of fireflies, at the cost of introducing a slight bias. Because of the selective nature of the culling mechanism, reflective caustics still benefit from the native path regularization, which further removes fireflies on other surfaces (bouncing light off casters). Differential Revision: https://developer.blender.org/D13533
2022-04-01 15:44:24 +02:00
}
if (integrator->get_guiding_params(device).use) {
kernel_features |= KERNEL_FEATURE_PATH_GUIDING;
}
if (bake_manager->get_baking()) {
kernel_features |= KERNEL_FEATURE_BAKING;
}
kernel_features |= film->get_kernel_features(this);
2021-10-26 15:30:12 +02:00
kernel_features |= integrator->get_kernel_features();
kernel_features |= camera->get_kernel_features();
dscene.data.kernel_features = kernel_features;
/* Currently viewport render is faster with higher max_closures, needs
* investigating. */
const uint max_closures = (params.background) ? get_max_closure_count() : MAX_CLOSURE;
dscene.data.max_closures = max_closures;
dscene.data.max_shaders = shaders.size();
/* Inform the device of the BVH limits. If this returns true, all BVHs
* and kernels need to be rebuilt. */
if (device->set_bvh_limits(objects.size(), kernel_max_prim_count)) {
kernels_loaded = false;
for (Geometry *geom : geometry) {
geom->need_update_rebuild = true;
geom->tag_modified();
}
}
}
bool Scene::update(Progress &progress)
{
if (!need_update()) {
return false;
}
/* Upload scene data to the GPU. */
progress.set_status("Updating Scene");
MEM_GUARDED_CALL(&progress, device_update, device, progress);
return true;
}
bool Scene::update_camera_resolution(Progress &progress, int width, int height)
{
if (!camera->set_screen_size(width, height)) {
return false;
}
camera->device_update(device, &dscene, this);
progress.set_status("Updating Device", "Writing constant memory");
device->const_copy_to("data", &dscene.data, sizeof(dscene.data));
return true;
}
static void log_kernel_features(const uint features)
{
LOG_INFO << "Requested features:";
LOG_INFO << "Use BSDF " << string_from_bool(features & KERNEL_FEATURE_NODE_BSDF);
LOG_INFO << "Use Emission " << string_from_bool(features & KERNEL_FEATURE_NODE_EMISSION);
LOG_INFO << "Use Volume " << string_from_bool(features & KERNEL_FEATURE_NODE_VOLUME);
LOG_INFO << "Use Bump " << string_from_bool(features & KERNEL_FEATURE_NODE_BUMP);
LOG_INFO << "Use Voronoi " << string_from_bool(features & KERNEL_FEATURE_NODE_VORONOI_EXTRA);
LOG_INFO << "Use Shader Raytrace " << string_from_bool(features & KERNEL_FEATURE_NODE_RAYTRACE);
LOG_INFO << "Use MNEE " << string_from_bool(features & KERNEL_FEATURE_MNEE);
LOG_INFO << "Use Transparent " << string_from_bool(features & KERNEL_FEATURE_TRANSPARENT);
LOG_INFO << "Use Denoising " << string_from_bool(features & KERNEL_FEATURE_DENOISING);
LOG_INFO << "Use Path Tracing " << string_from_bool(features & KERNEL_FEATURE_PATH_TRACING);
LOG_INFO << "Use Hair " << string_from_bool(features & KERNEL_FEATURE_HAIR);
LOG_INFO << "Use Pointclouds " << string_from_bool(features & KERNEL_FEATURE_POINTCLOUD);
LOG_INFO << "Use Object Motion " << string_from_bool(features & KERNEL_FEATURE_OBJECT_MOTION);
LOG_INFO << "Use Baking " << string_from_bool(features & KERNEL_FEATURE_BAKING);
LOG_INFO << "Use Subsurface " << string_from_bool(features & KERNEL_FEATURE_SUBSURFACE);
LOG_INFO << "Use Volume " << string_from_bool(features & KERNEL_FEATURE_VOLUME);
LOG_INFO << "Use Shadow Catcher " << string_from_bool(features & KERNEL_FEATURE_SHADOW_CATCHER);
Cycles: Add Portal Depth light pass information It allows to implement tricks based on a knowledge whether the path ever cam through a portal or not, and even something more advanced based on the number of portals. The main current objective is for strokes shading: stroke shader uses Ray Portal BSDF to place ray to the center of the stroke and point it in the direction of the surface it is generated for. This gives stroke a single color which matches shading of the original object. For this usecase to work the ray bounced from the original surface should ignore the strokes, which is now possible by using Portal Depth input and mixing with the Transparent BSDF. It also helps to make shading look better when there are multiple stroke layers. A solution of using portal depth is chosen over a single flag due to various factors: - Last time we've looked into it it was a bit tricky to implement as a flag due to us running out of bits. - It feels to be more flexible solution, even though it is a bit hard to come up with 100% compelling setup for it. - It needs to be slightly different from the current "Is Foo" flags, and be more "Is Portal Descendant" or something. An extra uint16 is added to the state to count the portal depth, but it is only allocated for scenes that use Ray Portal BSDF. Portal BSDF still increments Transparent bounce, as it is required to have some "limiting" factor so that ray does not get infinitely move to different place of the scene. Ref #125213 Pull Request: https://projects.blender.org/blender/blender/pulls/143107
2025-07-25 18:09:38 +02:00
LOG_INFO << "Use Portal Node " << string_from_bool(features & KERNEL_FEATURE_NODE_PORTAL);
}
bool Scene::load_kernels(Progress &progress)
{
progress.set_status("Loading render kernels (may take a few minutes the first time)");
const scoped_timer timer;
const uint kernel_features = dscene.data.kernel_features;
log_kernel_features(kernel_features);
if (!device->load_kernels(kernel_features)) {
string message = device->error_message();
if (message.empty()) {
message = "Failed loading render kernel, see console for errors";
}
progress.set_error(message);
progress.set_status(message);
progress.set_update();
return false;
}
kernels_loaded = true;
loaded_kernel_features = kernel_features;
return true;
}
int Scene::get_max_closure_count()
{
if (shader_manager->use_osl()) {
/* OSL always needs the maximum as we can't predict the
* number of closures a shader might generate. */
return MAX_CLOSURE;
}
int max_closures = 0;
for (int i = 0; i < shaders.size(); i++) {
Shader *shader = shaders[i];
if (shader->reference_count()) {
const int num_closures = shader->graph->get_num_closures();
max_closures = max(max_closures, num_closures);
}
}
max_closure_global = max(max_closure_global, max_closures);
if (max_closure_global > MAX_CLOSURE) {
/* This is usually harmless as more complex shader tend to get many
* closures discarded due to mixing or low weights. We need to limit
* to MAX_CLOSURE as this is hardcoded in CPU/mega kernels, and it
* avoids excessive memory usage for split kernels. */
LOG_WARNING << "Maximum number of closures exceeded: " << max_closure_global << " > "
<< MAX_CLOSURE;
max_closure_global = MAX_CLOSURE;
}
return max_closure_global;
}
int Scene::get_volume_stack_size() const
{
int volume_stack_size = 0;
/* Space for background volume and terminator.
* Don't do optional here because camera ray initialization expects that there
* is space for at least those elements (avoiding extra condition to check if
* there is actual volume or not).
*/
volume_stack_size += 2;
/* Quick non-expensive check. Can over-estimate maximum possible nested level,
* but does not require expensive calculation during pre-processing. */
bool has_volume_object = false;
for (const Object *object : objects) {
if (!object->get_geometry()->has_volume) {
continue;
}
if (object->intersects_volume) {
/* Object intersects another volume, assume it's possible to go deeper in
* the stack. */
/* TODO(sergey): This might count nesting twice (A intersects B and B
* intersects A), but can't think of a computationally cheap algorithm.
* Dividing my 2 doesn't work because of Venn diagram example with 3
* circles. */
++volume_stack_size;
}
else if (!has_volume_object) {
/* Allocate space for at least one volume object. */
++volume_stack_size;
}
has_volume_object = true;
if (volume_stack_size == MAX_VOLUME_STACK_SIZE) {
break;
}
}
volume_stack_size = min(volume_stack_size, MAX_VOLUME_STACK_SIZE);
LOG_DEBUG << "Detected required volume stack size " << volume_stack_size;
return volume_stack_size;
}
bool Scene::has_shadow_catcher()
{
if (shadow_catcher_modified_) {
has_shadow_catcher_ = false;
for (Object *object : objects) {
/* Shadow catcher flags on lights only controls effect on other objects, it's
* not catching shadows itself. This is on by default, so ignore to avoid
* performance impact when there is no actual shadow catcher. */
if (object->get_is_shadow_catcher() && !object->get_geometry()->is_light()) {
has_shadow_catcher_ = true;
break;
}
}
shadow_catcher_modified_ = false;
}
return has_shadow_catcher_;
}
void Scene::tag_shadow_catcher_modified()
{
shadow_catcher_modified_ = true;
}
bool Scene::has_volume()
{
has_volume_modified_ = false;
return dscene.data.integrator.use_volumes;
}
bool Scene::has_volume_modified() const
{
return has_volume_modified_;
}
void Scene::tag_has_volume_modified()
{
has_volume_modified_ = true;
}
template<> Light *Scene::create_node<Light>()
{
unique_ptr<Light> node = make_unique<Light>();
Light *node_ptr = node.get();
node->set_owner(this);
geometry.push_back(std::move(node));
light_manager->tag_update(this, LightManager::LIGHT_ADDED);
return node_ptr;
}
template<> Mesh *Scene::create_node<Mesh>()
{
unique_ptr<Mesh> node = make_unique<Mesh>();
Mesh *node_ptr = node.get();
node->set_owner(this);
geometry.push_back(std::move(node));
geometry_manager->tag_update(this, GeometryManager::MESH_ADDED);
return node_ptr;
}
template<> Hair *Scene::create_node<Hair>()
{
unique_ptr<Hair> node = make_unique<Hair>();
Hair *node_ptr = node.get();
node->set_owner(this);
geometry.push_back(std::move(node));
geometry_manager->tag_update(this, GeometryManager::HAIR_ADDED);
return node_ptr;
}
template<> Volume *Scene::create_node<Volume>()
{
unique_ptr<Volume> node = make_unique<Volume>();
Volume *node_ptr = node.get();
node->set_owner(this);
geometry.push_back(std::move(node));
geometry_manager->tag_update(this, GeometryManager::MESH_ADDED);
return node_ptr;
}
template<> PointCloud *Scene::create_node<PointCloud>()
{
unique_ptr<PointCloud> node = make_unique<PointCloud>();
PointCloud *node_ptr = node.get();
node->set_owner(this);
geometry.push_back(std::move(node));
geometry_manager->tag_update(this, GeometryManager::POINT_ADDED);
return node_ptr;
}
template<> Object *Scene::create_node<Object>()
{
unique_ptr<Object> node = make_unique<Object>();
Object *node_ptr = node.get();
node->set_owner(this);
objects.push_back(std::move(node));
object_manager->tag_update(this, ObjectManager::OBJECT_ADDED);
return node_ptr;
}
template<> ParticleSystem *Scene::create_node<ParticleSystem>()
{
unique_ptr<ParticleSystem> node = make_unique<ParticleSystem>();
ParticleSystem *node_ptr = node.get();
node->set_owner(this);
particle_systems.push_back(std::move(node));
particle_system_manager->tag_update(this);
return node_ptr;
}
template<> Shader *Scene::create_node<Shader>()
{
unique_ptr<Shader> node = make_unique<Shader>();
Shader *node_ptr = node.get();
node->set_owner(this);
shaders.push_back(std::move(node));
shader_manager->tag_update(this, ShaderManager::SHADER_ADDED);
return node_ptr;
}
template<> Pass *Scene::create_node<Pass>()
{
unique_ptr<Pass> node = make_unique<Pass>();
Pass *node_ptr = node.get();
node->set_owner(this);
passes.push_back(std::move(node));
film->tag_modified();
return node_ptr;
}
template<> Camera *Scene::create_node<Camera>()
{
unique_ptr<Camera> node = make_unique<Camera>();
Camera *node_ptr = node.get();
node->set_owner(this);
cameras.push_back(std::move(node));
return node_ptr;
}
template<> Integrator *Scene::create_node<Integrator>()
{
unique_ptr<Integrator> node = make_unique<Integrator>();
Integrator *node_ptr = node.get();
node->set_owner(this);
integrators.push_back(std::move(node));
return node_ptr;
}
template<> Background *Scene::create_node<Background>()
{
unique_ptr<Background> node = make_unique<Background>();
Background *node_ptr = node.get();
node->set_owner(this);
backgrounds.push_back(std::move(node));
return node_ptr;
}
template<> Film *Scene::create_node<Film>()
{
unique_ptr<Film> node = make_unique<Film>();
Film *node_ptr = node.get();
node->set_owner(this);
films.push_back(std::move(node));
return node_ptr;
}
template<> void Scene::delete_node(Light *node)
{
assert(node->get_owner() == this);
geometry.erase_by_swap(node);
light_manager->tag_update(this, LightManager::LIGHT_REMOVED);
}
template<> void Scene::delete_node(Mesh *node)
{
assert(node->get_owner() == this);
geometry.erase_by_swap(node);
geometry_manager->tag_update(this, GeometryManager::MESH_REMOVED);
}
template<> void Scene::delete_node(Hair *node)
{
assert(node->get_owner() == this);
geometry.erase_by_swap(node);
geometry_manager->tag_update(this, GeometryManager::HAIR_REMOVED);
}
template<> void Scene::delete_node(Volume *node)
{
assert(node->get_owner() == this);
geometry.erase_by_swap(node);
geometry_manager->tag_update(this, GeometryManager::MESH_REMOVED);
}
template<> void Scene::delete_node(PointCloud *node)
{
assert(node->get_owner() == this);
geometry.erase_by_swap(node);
geometry_manager->tag_update(this, GeometryManager::POINT_REMOVED);
}
template<> void Scene::delete_node(Geometry *node)
{
assert(node->get_owner() == this);
uint flag;
if (node->is_hair()) {
flag = GeometryManager::HAIR_REMOVED;
}
else {
flag = GeometryManager::MESH_REMOVED;
if (node->has_volume) {
volume_manager->tag_update(node);
}
}
geometry.erase_by_swap(node);
geometry_manager->tag_update(this, flag);
}
template<> void Scene::delete_node(Object *node)
{
assert(node->get_owner() == this);
uint flag = ObjectManager::OBJECT_REMOVED;
if (node->get_geometry()->has_volume) {
volume_manager->tag_update(node, flag);
}
objects.erase_by_swap(node);
object_manager->tag_update(this, flag);
}
template<> void Scene::delete_node(ParticleSystem *node)
{
assert(node->get_owner() == this);
particle_systems.erase_by_swap(node);
particle_system_manager->tag_update(this);
}
template<> void Scene::delete_node(Shader *node)
{
assert(node->get_owner() == this);
/* don't delete unused shaders, not supported */
node->clear_reference_count();
}
template<> void Scene::delete_node(Procedural *node)
{
assert(node->get_owner() == this);
procedurals.erase_by_swap(node);
procedural_manager->tag_update();
}
template<> void Scene::delete_node(Pass *node)
{
assert(node->get_owner() == this);
passes.erase_by_swap(node);
film->tag_modified();
}
template<typename T> static void assert_same_owner(const set<T *> &nodes, const NodeOwner *owner)
{
#ifdef NDEBUG
(void)nodes;
(void)owner;
#else
for (const T *node : nodes) {
assert(node->get_owner() == owner);
}
#endif
}
template<> void Scene::delete_nodes(const set<Geometry *> &nodes, const NodeOwner *owner)
{
assert_same_owner(nodes, owner);
geometry.erase_in_set(nodes);
geometry_manager->tag_update(this, GeometryManager::GEOMETRY_REMOVED);
light_manager->tag_update(this, LightManager::LIGHT_REMOVED);
}
template<> void Scene::delete_nodes(const set<Object *> &nodes, const NodeOwner *owner)
{
assert_same_owner(nodes, owner);
objects.erase_in_set(nodes);
object_manager->tag_update(this, ObjectManager::OBJECT_REMOVED);
}
template<> void Scene::delete_nodes(const set<ParticleSystem *> &nodes, const NodeOwner *owner)
{
assert_same_owner(nodes, owner);
particle_systems.erase_in_set(nodes);
particle_system_manager->tag_update(this);
}
template<> void Scene::delete_nodes(const set<Shader *> &nodes, const NodeOwner * /*owner*/)
{
/* don't delete unused shaders, not supported */
for (Shader *shader : nodes) {
shader->clear_reference_count();
}
}
template<> void Scene::delete_nodes(const set<Procedural *> &nodes, const NodeOwner *owner)
{
assert_same_owner(nodes, owner);
procedurals.erase_in_set(nodes);
procedural_manager->tag_update();
}
template<> void Scene::delete_nodes(const set<Pass *> &nodes, const NodeOwner *owner)
{
assert_same_owner(nodes, owner);
passes.erase_in_set(nodes);
film->tag_modified();
}
CCL_NAMESPACE_END