Gravity from Jupiter!!?

MGB · 2005-06-09T06:53:14

OK here's a strange one for non-SI-unit people out there ;) My world has 1 unit = 1 foot. I've been using a value for gravity that 'felt right', which was 0.00014. All ok - aircraft seems to handle fine. Then I realise I should be using a value of 0.032152 ft/ms to simulate real-world gravity. Result: aircraft falls out of the sky at seemingly ridiculous speeds (falls 200 ft in just over 100ms). But measuring the falling speed after 100ms gives 0.98m/s, which I believe is correct. There really isn't that much going into the calculations for gravity, basically: TVector3 accel(mLinearForceSum / TotalMass()); accel.y -= CONSTANTS().KGravitySpeed; // Apply gravity. mVelocity += accel * aFrameTimeMs; mPosn.Add(mVelocity * aFrameTimeMs); mLinearForceSum.Zero(); Am I missing something out here?! I've modelled air resistance as per real-world values (rho=0.002378 slugs/cubic ft) And same goes for my projectiles. Do I need to go to the lengths of simulating the Reynolds effect to keep em up in the air?!

Math and Physics Programming

Started by MGB June 06, 2005 07:48 AM

16 comments, last by gamechampionx 18 years, 10 months ago

Eelco

301

June 07, 2005 12:58 AM

Quote:Original post by CombatWombat
There isnt anything wrong with the english system per-se.

Consider yourself a well-rounded person if you know all its whacky details = o

is it just me or does your second sentence completely take the wind out of the sails of your first?

gamechampionx

140

June 07, 2005 05:47 AM

Just remember that acceleration depends on distance between the two objects and their masses. If you use a = 9.8 m/(s^2), your results will degrade as the object moves away from the center of gravity.

Check out Drunken Brawl at http://www.angelfire.com/games6/drunken_brawl!

CombatWombat

681

June 07, 2005 06:11 PM

Quote:Original post by Eelco
Quote:Original post by CombatWombat
There isnt anything wrong with the english system per-se.

Consider yourself a well-rounded person if you know all its whacky details = o

is it just me or does your second sentence completely take the wind out of the sails of your first?

Well by "isnt anything wrong with [the system]" i mean if you get all the details sorted it does work. The results will be correct and all that good stuff. Once you learn a few things about it, it works just like metric but with different conversion numbers (which if you have a calculator or PC doing the math isnt *that* major of a problem)

The second part I was refering to how a lot of people completely forsake it, which is an ignorant decision considering we WILL be having to deal with the system for a long time to come. The US would have to retool ALOT of stuff to switch to metric.

MGB

288

Author

June 08, 2005 06:20 AM

Cheers Wombat - a few things to look at there...
I did start off with things totally unscaled, but recently saw the error of my ways ;)

gamechampionx: by 'degrade' do you mean 'differ from reality'? I don't think I'll be any worse off by not modelling earth's gravity attenuation.

Squirm

481

June 08, 2005 06:47 AM

The old system has some advantages... 12 is a much more sensible number than 10, for example, though most of our population on gamedev would probably like to see 8 or 16. Metric... it's such a shame... if we had just got a decent power of 2 in there life would be so much simpler :}

alvaro

21,604

June 08, 2005 07:48 AM

You've gotta love Google for these things: Clicky

I would always do all computations in SI units (kg, m, s) and when you need to display a value to the user you can convert it to whatever units he wants. It's the same philosophy I follow for people counting from 1 (count from 0 internally, add 1 when you display the number) or for other weird ways in which humans think. Americans are a particularly defective type of humans when it comes to these things...

c = 1.8026175 × 10^12 furlongs / fortnight

:o)

Eelco

301

June 08, 2005 10:13 AM

Quote:Original post by CombatWombat
Quote:Original post by Eelco
Quote:Original post by CombatWombat
There isnt anything wrong with the english system per-se.

Consider yourself a well-rounded person if you know all its whacky details = o

is it just me or does your second sentence completely take the wind out of the sails of your first?

Well by "isnt anything wrong with [the system]" i mean if you get all the details sorted it does work. The results will be correct and all that good stuff. Once you learn a few things about it, it works just like metric but with different conversion numbers (which if you have a calculator or PC doing the math isnt *that* major of a problem)

The second part I was refering to how a lot of people completely forsake it, which is an ignorant decision considering we WILL be having to deal with the system for a long time to come. The US would have to retool ALOT of stuff to switch to metric.

well depends on what you define as wrong. personally, i greatly value simplicity and consistency

gamechampionx

140

June 09, 2005 06:53 AM

Yes, by degrade, I simply mean differ from reality.

9.8 m/s^2 is maybe good enough for acceleration based on objects within the earth's atmosphere, for example, but using that number for calculating forces between celestial bodies would be suicide, in which case the "full" formula of:

F = G*m1*m2 / (d^2) would be the only reasonable way to model such motion.

Check out Drunken Brawl at http://www.angelfire.com/games6/drunken_brawl!

Gravity from Jupiter!!?

This topic is closed to new replies.

Popular Topics

Recommended Tutorials

Gravity from Jupiter!!?

This topic is closed to new replies.

Popular Topics

Recommended Tutorials

Reticulating splines