diff --git a/tiny_bvh.h b/tiny_bvh.h
index 9e6ab35..0a210bb 100644
--- a/tiny_bvh.h
+++ b/tiny_bvh.h
@@ -1378,7 +1378,7 @@ void BVH::Convert( const BVHLayout from, const BVHLayout to, const bool deleteOr
 		// Convert a 4-wide BVH to a format suitable for CPU traversal.
 		// See Faster Incoherent Ray Traversal Using 8-Wide AVX InstructionsLayout,
 		// Atilla T. Áfra, 2013.
-		unsigned spaceNeeded = usedBVH4Nodes * 4; // here, 'block' is 16 bytes.
+		unsigned spaceNeeded = usedBVH4Nodes;
 		if (allocatedAlt4bNodes < spaceNeeded)
 		{
 			FATAL_ERROR_IF( bvh4Node == 0, "BVH::Convert( BASIC_BVH4, BVH4_AFRA ), bvh4Node == 0." );
@@ -1400,19 +1400,25 @@ void BVH::Convert( const BVHLayout from, const BVHLayout to, const bool deleteOr
 			for (int cidx = 0, i = 0; i < 4; i++) if (orig.child[i])
 			{
 				const BVHNode4& child = bvh4Node[orig.child[i]];
-				((float*)&newNode.xmin4)[cidx] = child.aabbMin.x, ((float*)&newNode.ymin4)[cidx] = child.aabbMin.y;
-				((float*)&newNode.zmin4)[cidx] = child.aabbMin.z, ((float*)&newNode.xmax4)[cidx] = child.aabbMax.x;
-				((float*)&newNode.ymax4)[cidx] = child.aabbMax.y, ((float*)&newNode.zmax4)[cidx] = child.aabbMax.z;
+				((float*)&newNode.xmin4)[cidx] = child.aabbMin.x;
+				((float*)&newNode.ymin4)[cidx] = child.aabbMin.y;
+				((float*)&newNode.zmin4)[cidx] = child.aabbMin.z;
+				((float*)&newNode.xmax4)[cidx] = child.aabbMax.x;
+				((float*)&newNode.ymax4)[cidx] = child.aabbMax.y;
+				((float*)&newNode.zmax4)[cidx] = child.aabbMax.z;
 				if (child.isLeaf())
-					newNode.childFirst[cidx] = orig.firstTri, newNode.triCount[cidx] = orig.triCount;
+					newNode.childFirst[cidx] = child.firstTri,
+					newNode.triCount[cidx] = child.triCount;
 				else
-					stack[stackPtr++] = (unsigned)((float*)&newNode.childFirst[cidx] - (float*)bvh4Alt2),
+					stack[stackPtr++] = (unsigned)((unsigned*)&newNode.childFirst[cidx] - (unsigned*)bvh4Alt2),
 					stack[stackPtr++] = orig.child[i];
 				cidx++;
 			}
 			// pop next task
 			if (!stackPtr) break;
-			nodeIdx = stack[--stackPtr], ((unsigned*)bvh4Alt2)[stack[--stackPtr]] = newAlt4Ptr;
+			nodeIdx = stack[--stackPtr];
+			unsigned offset = stack[--stackPtr];
+			((unsigned*)bvh4Alt2)[offset] = newAlt4Ptr;
 		#ifdef __GNUC__
 		#pragma GCC diagnostic pop
 		#endif
@@ -3172,30 +3178,31 @@ int BVH::Intersect_AltSoA( Ray& ray ) const
 
 int BVH::Intersect_Afra( Ray& ray ) const
 {
-	// SIMDVEC4 xmin4, ymin4, zmin4;
-	// SIMDVEC4 xmax4, ymax4, zmax4;
-	// unsigned childFirst[4];
-	// unsigned triCount[4];
 #if 1
 	// quick-and-dirty intersect to verify data structure
-	unsigned nodeIdx = 0, stack[64], stackPtr = 0, steps = 0;
+	unsigned nodeIdx = 0, stack[1024], stackPtr = 0, steps = 0;
 	while (1)
 	{
 		steps++;
 		BVHNode4Alt2& node = bvh4Alt2[nodeIdx];
-		for( unsigned i = 0; i < 4; i++ )
+	#ifdef __GNUC__
+	#pragma GCC diagnostic push
+	#pragma GCC diagnostic ignored "-Wstrict-aliasing"
+	#endif
+		for (unsigned i = 0; i < 4; i++) if (node.childFirst[i] + node.triCount[i] > 0)
 		{
 			bvhvec3 bmin, bmax;
 			bmin.x = ((float*)&node.xmin4)[i], bmax.x = ((float*)&node.xmax4)[i];
 			bmin.y = ((float*)&node.ymin4)[i], bmax.y = ((float*)&node.ymax4)[i];
 			bmin.z = ((float*)&node.zmin4)[i], bmax.z = ((float*)&node.zmax4)[i];
-			if (IntersectAABB( ray, bmin, bmax ))
+			float t = IntersectAABB( ray, bmin, bmax );
+			if (t < 1e30f)
 			{
 				if (node.triCount[i] > 0)
 				{
 					// process leaf
 					const unsigned first = node.childFirst[i], count = node.triCount[i];
-					for( unsigned j = 0; j < count; j++ ) IntersectTri( ray, triIdx[first + j] );
+					for (unsigned j = 0; j < count; j++) IntersectTri( ray, triIdx[first + j] );
 				}
 				else
 				{
@@ -3204,6 +3211,9 @@ int BVH::Intersect_Afra( Ray& ray ) const
 				}
 			}
 		}
+	#ifdef __GNUC__
+	#pragma GCC diagnostic pop
+	#endif
 		if (stackPtr == 0) break; else nodeIdx = stack[--stackPtr];
 	}
 #else
diff --git a/tiny_bvh_fenster.cpp b/tiny_bvh_fenster.cpp
index 7d11998..d40914d 100644
--- a/tiny_bvh_fenster.cpp
+++ b/tiny_bvh_fenster.cpp
@@ -5,6 +5,7 @@
 
 #define TINYBVH_IMPLEMENTATION
 #include "tiny_bvh.h"
+#include <fstream>
 
 using namespace tinybvh;
 
@@ -21,8 +22,7 @@ BVH bvh;
 
 #ifdef LOADSCENE
 bvhvec4* triangles = 0;
-const char scene[] = "happybuddha.bin";
-#include <fstream>
+const char scene[] = "cryteksponza.bin";
 #else
 ALIGNED( 16 ) bvhvec4 triangles[259 /* level 3 */ * 6 * 2 * 49 * 3]{};
 #endif
@@ -35,7 +35,8 @@ int verts = 0;
 static bvhvec3 eye( 0, 13, 30 ), p1, p2, p3;
 static bvhvec3 view = normalize( bvhvec3( 0, 0.01f, -1 ) );
 #else
-static bvhvec3 eye( -3.5f, -1.5f, -6.5f ), view = normalize( bvhvec3( 3, 1.5f, 5 ) );
+static bvhvec3 eye( -3.5f, -1.5f, -6.5f ), p1, p2, p3;
+static bvhvec3 view = normalize( bvhvec3( 3, 1.5f, 5 ) );
 #endif
 
 void sphere_flake( float x, float y, float z, float s, int d = 0 )
@@ -109,8 +110,8 @@ void Init()
 	// build a BVH over the scene
 #if defined(BVH_USEAVX)
 	bvh.BuildHQ( triangles, verts / 3 );
-	bvh.Convert( BVH::WALD_32BYTE, BVH::BASIC_BVH8 );
-	bvh.Convert( BVH::BASIC_BVH8, BVH::CWBVH );
+	bvh.Convert( BVH::WALD_32BYTE, BVH::BASIC_BVH4 );
+	bvh.Convert( BVH::BASIC_BVH4, BVH::BVH4_AFRA );
 #elif defined(BVH_USENEON)
 	bvh.BuildNEON( triangles, verts / 3 );
 #else
@@ -172,7 +173,7 @@ void Tick( uint32_t* buf )
 
 	// trace primary rays
 #if !defined USE_EMBREE
-	for (int i = 0; i < N; i++) bvh.Intersect( rays[i], BVH::CWBVH );
+	for (int i = 0; i < N; i++) bvh.Intersect( rays[i], BVH::BVH4_AFRA );
 #else
 	struct RTCRayHit rayhit;
 	for (int i = 0; i < N; i++)