Benchmark: point in triangle

Script Preparation code:

/*  var polygon = [{x:Math.random(), y:Math.random()}, {x:Math.random(), y:Math.random()}, {x:Math.random(), y:Math.random()}];
  px = Math.random();
  py = Math.random();
var v0x = polygon[1].x-polygon[0].x;
var v0y = polygon[1].y-polygon[0].y;
var v1x = polygon[2].x-polygon[0].x;
var v1y = polygon[2].y-polygon[0].y;

// if v0 cross v1 is positive, triangle is counter clockwise when viewed from above.
var normal = v0x*v1y - v1x*v0y;

point {x: 47.87142596679951, y: -6.500680708968048}
triangle
"{"a":{"x":-40.49496078491211,"y":-8.738582611083984},
  "b":{"x":-41.94913723574891,"y":-6.313193949870829},
  "c":{"x":-41.94775545887751,"y":-6.315498584878696}
*/
debugger;

Tests:

winding
var polygon = [{x:Math.random(), y:Math.random()}, {x:Math.random(), y:Math.random()}, {x:Math.random(), y:Math.random()}]; px = Math.random(); py = Math.random(); var v0x = polygon[1].x-polygon[0].x; var v0y = polygon[1].y-polygon[0].y; var v1x = polygon[2].x-polygon[0].x; var v1y = polygon[2].y-polygon[0].y; // if v0 cross v1 is positive, triangle is counter clockwise when viewed from above. var normal = v0x*v1y - v1x*v0y; var length = polygon.length; var windingNumber = 0; var i = 0; var l = polygon.length - 1; var edgePoint1 = polygon[l]; var edgePoint2; for (; i <= l; i++) { edgePoint2 = polygon[i]; var edgePoint1x = edgePoint1.x; var edgePoint1y = edgePoint1.y; var edgePoint2x = edgePoint2.x; var edgePoint2y = edgePoint2.y; // if it passes the point going upward if ( edgePoint1y <= py && edgePoint2y > py) { // and the point is to the left // EQUATION COPIED FROM GeometryUtils.isLeft for performance reasons if ((edgePoint2x - edgePoint1x) * (py - edgePoint1y) - (px - edgePoint1x) * (edgePoint2y - edgePoint1y) > 0 ) { // increment windingNumber windingNumber++; } } // else if it passes the point going down else if ( edgePoint1y > py && edgePoint2y <= py) { // and the point is to the right if ((edgePoint2x - edgePoint1x) * (py - edgePoint1y) - (px - edgePoint1x) * (edgePoint2y - edgePoint1y) < 0) { // decrement windingNumber windingNumber--; } } edgePoint1 = edgePoint2; } // the point is inside the polygon as long as the winding number isn't zero. return windingNumber !== 0;
bary
var polygon = [{x:Math.random(), y:Math.random()}, {x:Math.random(), y:Math.random()}, {x:Math.random(), y:Math.random()}]; px = Math.random(); py = Math.random(); var v0x = polygon[1].x-polygon[0].x; var v0y = polygon[1].y-polygon[0].y; var v1x = polygon[2].x-polygon[0].x; var v1y = polygon[2].y-polygon[0].y; // if v0 cross v1 is positive, triangle is counter clockwise when viewed from above. var normal = v0x*v1y - v1x*v0y; var a = polygon[0]; var b = polygon[1]; var c = polygon[2]; // Compute vectors var v0x = c.x - a.x; var v0y = c.y - a.y; var v1x = b.x - a.x; var v1y = b.y - a.y; var v2x = px - a.x; var v2y = py - a.y; var dot00 = v0x * v0x + v0y * v0y; var dot01 = v0x * v1x + v0y * v1y; var dot02 = v0x * v2x + v0y * v2y; var dot11 = v1x * v1x + v1y * v1y; var dot12 = v1x * v2x + v1y * v2y; var denom = dot00 * dot11 - dot01 * dot01; var u = (dot11 * dot02 - dot01 * dot12) / denom; var v = (dot00 * dot12 - dot01 * dot02) /denom; // Check if point is in triangle return (u >= 0) && (v >= 0) && (u + v < 1);
bary2
var polygon = [{x:Math.random(), y:Math.random()}, {x:Math.random(), y:Math.random()}, {x:Math.random(), y:Math.random()}]; px = Math.random(); py = Math.random(); var v0x = polygon[1].x-polygon[0].x; var v0y = polygon[1].y-polygon[0].y; var v1x = polygon[2].x-polygon[0].x; var v1y = polygon[2].y-polygon[0].y; // if v0 cross v1 is positive, triangle is counter clockwise when viewed from above. var normal = v0x*v1y - v1x*v0y; var a = polygon[0]; var b = polygon[1]; var c = polygon[2]; var ax = a.x; var ay = a.y; // Compute vectors var v0x = c.x - ax; var v0y = c.y - ay; var v1x = b.x - ax; var v1y = b.y - ay; var v2x = px - ax; var v2y = py - ay; var dot00 = v0x * v0x + v0y * v0y; var dot01 = v0x * v1x + v0y * v1y; var dot02 = v0x * v2x + v0y * v2y; var dot11 = v1x * v1x + v1y * v1y; var dot12 = v1x * v2x + v1y * v2y; var denom = dot00 * dot11 - dot01 * dot01; var u = (dot11 * dot02 - dot01 * dot12) / denom; var v = (dot00 * dot12 - dot01 * dot02) /denom; // Check if point is in triangle return (u >= 0) && (v >= 0) && (u + v < 1) && (normal > 0);
half planes
var polygon = [{x:Math.random(), y:Math.random()}, {x:Math.random(), y:Math.random()}, {x:Math.random(), y:Math.random()}]; px = Math.random(); py = Math.random(); var v0x = polygon[1].x-polygon[0].x; var v0y = polygon[1].y-polygon[0].y; var v1x = polygon[2].x-polygon[0].x; var v1y = polygon[2].y-polygon[0].y; // if v0 cross v1 is positive, triangle is counter clockwise when viewed from above. var normal = v0x*v1y - v1x*v0y; var ax = polygon[0].x; var bx = polygon[1].x; var cx = polygon[2].x; var ay = polygon[0].y; var by = polygon[1].y; var cy = polygon[2].y; var b1 = (px - bx) * (ay - by) - (ax - bx) * (py - by); var b2 = (px - cx) * (by - cy) - (bx - cx) * (py - cy); var b3 = (px - ax) * (cy - ay) - (cx - ax) * (py - ay); return (b1==b2) && (b2==b3);
half planes with normal
var polygon = [{x:Math.random(), y:Math.random()}, {x:Math.random(), y:Math.random()}, {x:Math.random(), y:Math.random()}]; px = Math.random(); py = Math.random(); var v0x = polygon[1].x-polygon[0].x; var v0y = polygon[1].y-polygon[0].y; var v1x = polygon[2].x-polygon[0].x; var v1y = polygon[2].y-polygon[0].y; // if v0 cross v1 is positive, triangle is counter clockwise when viewed from above. var normal = v0x*v1y - v1x*v0y; var ax = polygon[0].x; var bx = polygon[1].x; var cx = polygon[2].x; var ay = polygon[0].y; var by = polygon[1].y; var cy = polygon[2].y; if (((px - ax) * (by - ay) - (bx - ax) * (py - ay)) > 0 == normal > 0) return true; if (((px - bx) * (cy - by) - (cx - bx) * (py - by)) > 0 == normal > 0) { return true; } return ((px - cx) * (ay - cy) - (ax - cx) * (py - cy)) > 0 == normal > 0;
half planes with normal and vars
var polygon = [{x:Math.random(), y:Math.random()}, {x:Math.random(), y:Math.random()}, {x:Math.random(), y:Math.random()}]; px = Math.random(); py = Math.random(); var v0x = polygon[1].x-polygon[0].x; var v0y = polygon[1].y-polygon[0].y; var v1x = polygon[2].x-polygon[0].x; var v1y = polygon[2].y-polygon[0].y; // if v0 cross v1 is positive, triangle is counter clockwise when viewed from above. var normal = v0x*v1y - v1x*v0y; var a = polygon[0]; var b = polygon[1]; var c = polygon[2]; var ax = a.x; var bx = b.x; var ay = a.y; var by = b.y; var pxax = px-ax; var byay = by-ay; var bxax = bx-ax; var pyay = py-ay; if ((pxax * byay - bxax * pyay) * normal > 0) return true; var cx = c.x; var cy = c.y; var pxbx = px-bx; var cyby = cy-by; var cxbx = cx-bx; var pyby = py-by; if ((pxbx * cyby - cxbx * pyby) * normal > 0) return true; var pxcx = px-cx; var aycy = ay-cy; var axcx = ax-cx; var pycy = py-cy; return ((pxcx * aycy - axcx * pycy) * normal) > 0;

Rendered benchmark preparation results:

Suite status: <idle, ready to run>

Previous results

Test case name	Result
winding
bary
bary2
half planes
half planes with normal
half planes with normal and vars

Fastest: N/A

Slowest: N/A

Autogenerated LLM Summary (model llama3.2:3b, generated 26 days ago):

LLMs can make mistakes. Check important info.

It seems like you're providing benchmark results for some JavaScript tests. I'll help you analyze the data.

There are six test runs, each representing a different scenario:

1. `bary`
2. `bary2`
3. `winding`
4. `half planes`
5. `half planes with normal and vars`
6. `half planes with normal`

Let's look at the values that might be of interest:

* `ExecutionsPerSecond`: This value represents the number of test executions per second.
* `TestName`: Each test has a unique name, which might indicate the specific scenario or requirement being tested.

Now, let's see if we can identify any trends or insights from these results:

**Most efficient browser:** Chrome 58 is consistently performing well across all tests, with an average execution speed of around 180,000 executions per second.

**Least efficient browsers:** The other two test runs (`winding` and `half planes`) have the lowest execution speeds, averaging around 120,000 to 140,000 executions per second.

**Test variations:** Tests like `bary2`, `half planes with normal and vars`, and `half planes` are more computationally intensive than others (e.g., `winding`). These tests might require more processing power or resources, which can impact execution speed.

Keep in mind that these results are just one aspect of the overall benchmarking process. To draw meaningful conclusions, you'd need to consider other factors, such as code performance, optimization opportunities, and hardware-specific issues (if applicable).

Latest run results:

Run details: (Test run date: 7 years ago)

User agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/58.0.3029.110 Safari/537.36

Browser/OS: Chrome 58 on Linux

View result in a separate tab

Test name	Executions per second
winding	212920.8 Ops/sec
bary	377547.8 Ops/sec
bary2	319438.4 Ops/sec
half planes	203060.2 Ops/sec
half planes with normal	179106.9 Ops/sec
half planes with normal and vars	186834.0 Ops/sec

Related benchmarks:

point in triangle (version: 0)

Comparing performance of: winding vs bary vs bary2 vs half planes vs half planes with normal vs half planes with normal and vars

Created: 7 years ago by: Registered User

Jump to the latest result

winding

bary

bary2

half planes

half planes with normal

half planes with normal and vars

Suite status: <idle, ready to run>

Experimental features:

Fastest: N/A

Slowest: N/A

Autogenerated LLM Summary (model llama3.2:3b, generated 26 days ago):