Silk.NET•3y ago

doesn't Span have a really efficient equality implementation already?

Aqua•10/20/23, 10:39 PM

intrinsics can often be slower if you're just using them for one singular operation

MMaxine casting a Vector3<int> to Vector3 (float) for equality should be fine, right

PerkseyOP•10/20/23, 10:39 PM

// USE AN EPSILON DAMNIT!// USE AN EPSILON DAMNIT!

PPerksey doesn't Span have a really efficient equality implementation already?

Maxine•10/20/23, 10:39 PM

so i should just create a Span<T> for this and other and equals that?

Maxine•10/20/23, 10:39 PM

that would both be generalizable and fast

Maxine•10/20/23, 10:39 PM

possibly

Maxine•10/20/23, 10:39 PM

if the JIT created the right instructions

PerkseyOP•10/20/23, 10:40 PM

sure but you're trading correctness for performance

Maxine•10/20/23, 10:40 PM

what do you mean correctness, Vector3 Equals was never meant to use an epsilon

PerkseyOP•10/20/23, 10:40 PM

it is conceivably possible that the same effective float value has differing bits

Maxine•10/20/23, 10:41 PM

yep i just figured out why this wouldn't work

Maxine•10/20/23, 10:41 PM

things like -0 and +0

Maxine•10/20/23, 10:41 PM

NaN equality

PPerksey sure but you're trading correctness for performance

Aqua•10/20/23, 10:41 PM

i'm not even sure it would be more performant

Aqua•10/20/23, 10:41 PM

or if it is, its negligable

PerkseyOP•10/20/23, 10:41 PM

it'd be more performant for gigantic structs

Aqua•10/20/23, 10:41 PM

you're replacing instructions with other instructions

Aqua•10/20/23, 10:41 PM

true, but not for a small one like vectors

Aqua•10/20/23, 10:41 PM

matrix probably

PerkseyOP•10/20/23, 10:41 PM

oh yeah

Aqua•10/20/23, 10:41 PM

but not vectors

PerkseyOP•10/20/23, 10:41 PM

like seriously please just use specialised impls

AAqua i'm not even sure it would be more performant

Maxine•10/20/23, 10:42 PM

i jsut adapted this

Aqua•10/20/23, 10:42 PM

maybe there is something i am missing ¯\_(ツ)_/¯

Maxine•10/20/23, 10:42 PM

should i throw this all out and jus compare XYZ? that's fine with me

Aqua•10/20/23, 10:42 PM

i've done the experiments with SIMD and vectors

Aqua•10/20/23, 10:43 PM

and in my tests its far slower than to just do the operations directly

Aqua•10/20/23, 10:43 PM

because of the amount of extra bulk that goes along with it

Aqua•10/20/23, 10:43 PM

simd comes into its own when you do an operation multiple times, just once its often worse for perf

Aqua•10/20/23, 10:43 PM

which is why matrix benefits from simd

Aqua•10/20/23, 10:43 PM

if i'm wrong please correct me, this is just from the results of my own experiments

TechPizza•10/20/23, 10:44 PM

i would say that using SIMD for Vector4 is perfectly fine

TechPizza•10/20/23, 10:44 PM

Vector2 and Vector3 is way too painful due to not being the size of a Vector128

PerkseyOP•10/20/23, 10:46 PM

if (Vector128.IsHardwareAccelerated && typeof(T) == typeof(float))
{
    return Unsafe.BitCast<Vector3<T>, Vector3>(this).AsVector128().Equals(Unsafe.BitCast<Vector3<T>, Vector3>(other).AsVector128());
}

if (Vector128.IsHardwareAccelerated && typeof(T) == typeof(float))
{
    return Unsafe.BitCast<Vector3<T>, Vector3>(this).AsVector128().Equals(Unsafe.BitCast<Vector3<T>, Vector3>(other).AsVector128());
}

Maxine•10/20/23, 10:46 PM

is that a .net 8 method

Maxine•10/20/23, 10:47 PM

i like it

PPerksey ```cs if (Vector128.IsHardwareAccelerated && typeof(T) == typeof(float)) { r...

Aqua•10/20/23, 10:48 PM

again, just replacing instructions with other instructions
sure, BitCastBitCast may be fast, but you're introducing a comparison, a function call, etc.

Maxine•10/20/23, 10:48 PM

those should be inlined

Maxine•10/20/23, 10:48 PM

typeof(T) == typeof(Y) gets inlined

Aqua•10/20/23, 10:48 PM

hmmm yes fair point

Maxine•10/20/23, 10:48 PM

IsHardwareAcceleration gets eliminated

Aqua•10/20/23, 10:48 PM

i know that

Maxine•10/20/23, 10:49 PM

the most expensive thing here is the new Vector4

PerkseyOP•10/20/23, 10:49 PM

yeah all of that will be inlined and understood by the JIT

Maxine•10/20/23, 10:49 PM

and the SIMD itself

Maxine•10/20/23, 10:49 PM

for short and byte, the default equality should get optimized into a simple comparison, right?

Aqua•10/20/23, 10:49 PM

how does nativeaot handle this situation?

PerkseyOP•10/20/23, 10:49 PM

the same way the JIT does

PerkseyOP•10/20/23, 10:49 PM

because it is the JIT

PerkseyOP•10/20/23, 10:49 PM

just ahead of time

if (Vector128.IsHardwareAccelerated && typeof(T) == typeof(float)) { return Unsafe.BitCast<Vector3<T>, Vector3>(this).AsVector128().Equals(Unsafe.BitCast<Vector3<T>, Vector3>(other).AsVector128()); }

doesn't Span have a really efficient equality implementation already?

Similar Threads