fixed point was a mistake
-
fixed point was in fact *not* a mistake
@eniko what a rollercoaster of emotions!
-
@eniko pretty important for, like, finance and shit
-
@ratsnakegames @eniko I wanted to write COBOL not FORTRAN but my brain missed a beat
-
@ratsnakegames @eniko I wanted to write COBOL not FORTRAN but my brain missed a beat
@gabrielesvelto @eniko we still do that sh*t in Java or .NET these days because financial regulators really frown upon money vanishing because of floating point errors
-
@gabrielesvelto @eniko we still do that sh*t in Java or .NET these days because financial regulators really frown upon money vanishing because of floating point errors
@ratsnakegames and I commend you for that! In a past life I worked for a high-frquency trading firm and their software did financial calculations using floating point numbers 🤷
-
fixed point was in fact *not* a mistake
@eniko jumpscare
-
when 64kb ram was all anyone would ever need
-
@ratsnakegames and I commend you for that! In a past life I worked for a high-frquency trading firm and their software did financial calculations using floating point numbers 🤷
@gabrielesvelto @ratsnakegames
Eh, as long as there's enough bits in the mantissa... -
fixed point was in fact *not* a mistake
@eniko following with interest.
-
fixed point was in fact *not* a mistake
I have completed the triangle rasterizer fixed point conversion
The benchmarks have all improved between 0 and 35%
Except for threaded random triangles with flat color. That metric has increased 106%. As in its twice as fast as before
I have no idea why but I'm fairly sure I've ruled out bugs in my benchmarking
I am very confused
-
fixed point was in fact *not* a mistake
@eniko fixed point is very fun
-
I have completed the triangle rasterizer fixed point conversion
The benchmarks have all improved between 0 and 35%
Except for threaded random triangles with flat color. That metric has increased 106%. As in its twice as fast as before
I have no idea why but I'm fairly sure I've ruled out bugs in my benchmarking
I am very confused
@eniko Hyperthreading? That has very different behavior depending on the kind of code being run. A single core might have twice the int resources to handle two threads of throughput but not twice the float resources (and then it depends on how bottlenecked by things like memory latency the benchmark is).
-
@eniko Hyperthreading? That has very different behavior depending on the kind of code being run. A single core might have twice the int resources to handle two threads of throughput but not twice the float resources (and then it depends on how bottlenecked by things like memory latency the benchmark is).
@lina I don't know? >_> I just split the screen between 4 worker threads that all draw each triangle to their quadrant
-
@lina I don't know? >_> I just split the screen between 4 worker threads that all draw each triangle to their quadrant
@eniko How many cores/threads does your machine have? Can you pin app threads to specific cores/threads?
-
@lina I don't know? >_> I just split the screen between 4 worker threads that all draw each triangle to their quadrant
@eniko@mastodon.gamedev.place @lina@vt.social do you happen to own a Bulldozer CPU?
-
@eniko How many cores/threads does your machine have? Can you pin app threads to specific cores/threads?
@lina not sure offhand and I'm in bed now but my CPU is a Ryzen 7 5600g
-
@eniko@mastodon.gamedev.place @lina@vt.social do you happen to own a Bulldozer CPU?
-
I have completed the triangle rasterizer fixed point conversion
The benchmarks have all improved between 0 and 35%
Except for threaded random triangles with flat color. That metric has increased 106%. As in its twice as fast as before
I have no idea why but I'm fairly sure I've ruled out bugs in my benchmarking
I am very confused
@eniko mutex/locking/semaphores?
-
I have completed the triangle rasterizer fixed point conversion
The benchmarks have all improved between 0 and 35%
Except for threaded random triangles with flat color. That metric has increased 106%. As in its twice as fast as before
I have no idea why but I'm fairly sure I've ruled out bugs in my benchmarking
I am very confused
@eniko (Highly speculative, since FPUs are pretty good these days:) If the CPU has hyperthreading: Maybe two fixed-point threads share ALUs better than two floating-point threads share FPUs?
You might even find that combined fixed and float makes better overall use of a modern CPU, provided you can do both on a single thread or jump through the hoops to get both threads scheduled on the same core.
-
@lina not sure offhand and I'm in bed now but my CPU is a Ryzen 7 5600g
@eniko That doesn't exist... Ryzen 5 or different model number?
If it's the 5 5600G then that's 6 cores, so with 4 threads you shouldn't have HT effects as long as the OS scheduler isn't dumb about it...
