SH4 FTRV Optimizations - Revision history

GyroVorbis: /* Bounding Sphere vs View Frustum Culling */

2025-08-06T08:32:21Z

Bounding Sphere vs View Frustum Culling

← Older revision		Revision as of 08:32, 6 August 2025
Line 88:		Line 88:
	for intersection. */		for intersection. */
	for(unsigned i = 0; i < 4; ++i) {		for(unsigned i = 0; i < 4; ++i) {
	if(s->radius < distances.elem[0])		if(s->radius < distances.elem[i])
	return SPHEREOUTSIDE;		return SPHEREOUTSIDE;
	else if(s->radius > -distances.elem[0])		else if(s->radius > -distances.elem[i])
	res = SPHEREBOUNDARY;		res = SPHEREBOUNDARY;
	}		}

2025-08-06T08:13:07Z

Bounding Sphere vs View Frustum Culling

← Older revision		Revision as of 08:13, 6 August 2025
Line 83:		Line 83:
	component is equal to the dot product between the sphere's centroid and the		component is equal to the dot product between the sphere's centroid and the
	corresponding plane column vector. */		corresponding plane column vector. */
	shz_vec4_t distances = shz_xmtrx_trans_vec4(s->center);		shz_vec4_t distances = shz_xmtrx_trans_vec4(shz_vec4_t(s->center, -1.0f));

	/* Now we simply iterate over each of our 4 result components to check		/* Now we simply iterate over each of our 4 result components to check

2025-08-06T08:11:35Z

Bounding Sphere vs View Frustum Culling

← Older revision		Revision as of 08:11, 6 August 2025
Line 65:		Line 65:
	// Use FIPR to accelerate first two dot products independently		// Use FIPR to accelerate first two dot products independently
	for(unsigned i = 0; i < 2; ++i) {		for(unsigned i = 0; i < 2; ++i) {
	float distance = ~~shz_vec4_dot~~(s->center, p->plane);		float distance = shz_vec3_dot(s->center, p->plane) - p->plane.distance;

	if(s->radius < distance)		if(s->radius < distance)

2025-08-06T08:04:52Z

When to use FTRV

← Older revision		Revision as of 08:04, 6 August 2025
Line 16:		Line 16:
	There are two things to look out for when developing an intuition for when you can leverage FTRV:		There are two things to look out for when developing an intuition for when you can leverage FTRV:
	# There are 3 or more dot products (ax + by + cz + dw) being calculated back-to-back.		# There are 3 or more dot products (ax + by + cz + dw) being calculated back-to-back.
	# One of the vectors is held constant while the other vector argument is variable.		# One of the vectors is held constant while the other vector argument is variable across each.

	When you see such a scenario, the first thing that should pop into your head is that you are dealing with a vector x matrix transform, and you can use '''FTRV''' to accelerate it.		When you see such a scenario, the first thing that should pop into your head is that you are dealing with a vector x matrix transform, and you can use '''FTRV''' to accelerate it.

2025-08-06T08:04:16Z

When to use FTRV

← Older revision		Revision as of 08:04, 6 August 2025
Line 14:		Line 14:

	==When to use FTRV==		==When to use FTRV==
			There are two things to look out for when developing an intuition for when you can leverage FTRV:
			# There are 3 or more dot products (ax + by + cz + dw) being calculated back-to-back.
			# One of the vectors is held constant while the other vector argument is variable.

			When you see such a scenario, the first thing that should pop into your head is that you are dealing with a vector x matrix transform, and you can use '''FTRV''' to accelerate it.

	==Real-World Examples==		==Real-World Examples==

2025-08-06T07:57:49Z

Bounding Sphere vs View Frustum Culling

2025-08-06T07:54:55Z

Bounding Sphere vs View Frustum Culling

← Older revision		Revision as of 07:54, 6 August 2025
Line 76:		Line 76:
	our 4 plane vectors. This gives us a result vector where the value of each		our 4 plane vectors. This gives us a result vector where the value of each
	component is equal to the dot product between the sphere's centroid and the		component is equal to the dot product between the sphere's centroid and the
	corresponding plane column vector. */		corresponding plane column vector. */
	shz_vec4_t distances = shz_xmtrx_trans_vec4(s->center);		shz_vec4_t distances = shz_xmtrx_trans_vec4(s->center);

2025-08-06T07:53:52Z

Bounding Sphere vs View Frustum Culling

← Older revision		Revision as of 07:53, 6 August 2025
Line 35:		Line 35:
	const FrustumPlane *p = this->frustumPlanes;		const FrustumPlane *p = this->frustumPlanes;
	for(int32 i = 0; i < 6; i++){		for(int32 i = 0; i < 6; i++){
	/* Compute dot product between each plane and the centroid		// Compute dot product between each plane and the centroid.
	Your FIPR senses should be tingling, since we've got dot products! */		// Your FIPR senses should be tingling, since we've got dot products!
	float32 distance = dot(p->plane, s->center) - p->plane.distance;		float32 distance = dot(p->plane, s->center) - p->plane.distance;

2025-08-06T07:53:24Z

Bounding Sphere vs View Frustum Culling

← Older revision		Revision as of 07:53, 6 August 2025
Line 25:		Line 25:
	===Collision and Physics===		===Collision and Physics===
	===Bounding Sphere vs View Frustum Culling===		===Bounding Sphere vs View Frustum Culling===
	The following code snippet is taken from the DCA3 codebase as part of its Renderware driver back-end for Dreamcast. To check for intersection between a bounding sphere and the view frustum, you must compute the dot product between the centroid of the bounding sphere against all 6 of ~~its~~ planes.		The following code snippet is taken from the DCA3 codebase as part of its Renderware driver back-end for Dreamcast. To check for intersection between a bounding sphere and the view frustum, you must compute the dot product between the centroid of the bounding sphere against all 6 of the view frustum planes. The result of each dot product represents the sphere's distance from that plane.

	The original algorithm was as follows:		The original algorithm was as follows:

2025-08-06T07:52:37Z

@@ Line 25: / Line 25: @@
 ===Collision and Physics===
 ===Bounding Sphere vs View Frustum Culling===
 ===ADPCM Decoding===