hide? i've been participating in the conversation
hide? i've been participating in the conversation
Awhy not?
Amy math structs are all using generic math and they work excellently
the only flaw that i can see is that some functions won't work on integer types as they require
the only flaw that i can see is that some functions won't work on integer types as they require
IRootFunctions or whatever which the integer ones don't implement
Keh just a joke 
Kjust fund it funny I saw you there and here, and tried to come up with something witty to say
KDo you math structs involved SIMD?
Kalso what is CKnaufe
Kbest I could come up with was this
K
Ksomehow I think that's not it
Anot currently, i tried adding simd and the only thing it did was reduce performance
so likely the only place simd will be added is matrix multiplication where it does make sense to have it
so likely the only place simd will be added is matrix multiplication where it does make sense to have it
Lyup thats it you got it right im referencing old german writings
Kright, so in cases where you require special code paths / exact type, it gets finnicky
Kprops to you for recognizing that as old german writing
Kalso I have never heard of this but it's damn interesting
Kit's a super old documentation of the german langugage
K"Deutsche Sprache" or as it was then "Teutsche Sprache" (apparently)
Anah the reducing performance thing wasn't even to do with the generic math, from what i've read multiplying 2 vectors together using simd is just a waste, where simd only has benefits if you're doing like matrix multiplication, or large amounts of vector multiplication at once
Loh i was being sarcastic it's knife spelled with au and starting with the letter C, long story, but yeah i could tell it's old german, i studied german for a couple months awhile back
Lim quite good at recognizing languages
KThat's an Iceberg right there
Koh sure it wasn't from that, I'm just suggesting that that's where it gets hard
KSIMD also really only has benefits in loops so
Aah
yeah
simd will be more challenging with generic math for sure
yeah
simd will be more challenging with generic math for sure
Ksince just multiplying 4 numbers will just result in
Kso the muls are independent and processor pipeline will handle it
Ayep
Ai found just multiplying the 4 numbers together was much faster
Abut hey, worth experimenting
Lthe solution is to fit all the memory you need for your app into 1 register, then you never have to deal with any of that
Kah yes
KKthat's what my apps also look like xD
Atop 10 games of all time
L> void
> return 5
> return 5
Kyes all my apps have compiler errors
Kproblem?
K
Lyes, method of type void can not have a return value
Kthat's what the compiler keeps telling me
Kit's really mean :(
KBEYLEY HAVE YOU SEEN THIS
Lyes
LUNUX
KUNUX lmao
Pthis is not true, the JIT does not recognise Vector4D and friends as TYP_SIMD thus it will not place our custom struct in VEX registers. This means that there is an enregistration penalty every time we want to do an operation using System.Numerics. That being said, it should still be possible and less expensive than regular float maths.
PThere was some work being done to recognise user structs as TYP_SIMD which, once implemented, will allow us to have the same performance as System.Numerics
Mahhhh
Mi didn't even think of that, fair cop
KGitHub
Summary of the PR
Fixes BreakneckSleep in the existing CPU usage PR to store the last tick timestamp and compare to that instead of to when Sleep is called.
The branch still doesnt actually use Bre...
Fixes BreakneckSleep in the existing CPU usage PR to store the last tick timestamp and compare to that instead of to when Sleep is called.
The branch still doesnt actually use Bre...