To what order of magnitude would be sufficient to step time based functions for physics applications ? Since direct solutions off multivariant co dependent problems usually cant be found until you update things like position velocity and or energy over a very short tike period first. Example being gravitation or electrostatics

Like nano seconds ?

It really depends on what hardware and language you're working on. There's no point writing code to run faster than the clock speed of the hardware that it's running on. I know someone who codes with VHDL, which is a language that can compiles into circuit board diagrams. It even takes into account the time that it takes for an electrical signal to travel through each trace. Every line of code runs in parallel so you can't access a variable until the next cycle after it's defined.

They have to test out different hardware for just the compilation based on what the code is doing because it could take days to compile on the wrong hardware. 10 hours to compile is fast for them.

@cmarshall10450 oh no no
I mean for plain simulation

Like take a multiple charge system where the charhed objects can move like air particles and none of them are fixed

Defining by them in terms of time creates position based dependency hell!

You have to take what you know and determine ney force on each particle at each time and physical position and accept some error to work the system

What timewise is usually acceptable ?

@cmarshall10450 imagine every second creates cumulative error in the system

Every frame rather

Too short and you run the risk of not being able to represent the smaller numbers. Too long and you get too much error per frame. I would think you would need to try different frame sizes and compare error output and determine what is acceptable. Have you talked to people who simulate things like nuclear bomb blasts? Or other similar things? Is there an industry standard?

@TheWrongGod I guess it really depends on how accurate your simulation needs to be. No matter how granular your time frame is, you will lose some calculations. Even in the real physical world things can be too granular to calculate - quantum tunneling for example.

I do sim racing, which has to be very physically accurate to the real life cars based on real data, but even then, 400hz is seen as an amazing clock speed.

What I was saying before was obviously an extreme case. That's a case when the circuits are having to work in the real world in an aerospace setting so need to be calculating everything in the GHz range rather than just hz.

@TheWrongGod the errors would only be cumulative if they were fed back into the circuit as an input, otherwise it should be a deterministic error on each clock cycle.

@cmarshall10450 do you think when the force is based on relative distance which results in error akin to rieman sums vs integrals ?

@TheWrongGod an approximation is always just an approximation. Quantum tunneling of electrons shows that even in the physical world that you're attempting to simulate, things can be missed.

What are you wanting to simulate and how often do the physics need to be calculated? The complexity of the calculation will also play a part in how fast you can do it as the longer it takes to calculate the next frame, the slower your physics can be refreshed.

@cmarshall10450 well when I was sitting around with access to a mm physics text I started thinking a lot about electricity

And I kind of just for shits and giggles wanted to tackle a queuing system for numerous interactions but make it with charged particles which are interdependent and just simulate them.frame by frame until there were no more changes. Seems like this and the logic for minimizing the number of factors in say a collision system involving many many objects would have been solved already.

On a lighter note I was wondering if modern scum women might be more controllable if I just came out and said "well thought to be extremely jealous controlling and suspicious of me given the knowledge of relative nature's and the likely imaginary happiness being experienced and only compensated for by sex with a woman in their 20s or 30s of superb physical attractiveness since that's the only worthwhile trait anymore."

Like

Sign me up to be consistently viewed as a piece of trash that can never behave enough given the overall worthless core of my fucked up nature and the fact that you feel forced by lack of alternative to rut with me to satisfy your considerable physical and psychological male adult sexual impulses which I continually exacerbate and don't want to shop for pussy constantly and would rather keep one at home. And eat my food and use my shower and bed and call me a dumb meddling piece of shit whore who'd be better off dead.

Lol about sums it all up really

Well that took quite a turn very quickly. Please get help.