
Advertisement
Fulcrum.013
Member
Content count
44 
Joined

Last visited

Days Won
1
Fulcrum.013 last won the day on June 18
Fulcrum.013 had the most liked content!
Community Reputation
39 NeutralAbout Fulcrum.013

Rank
Member
Personal Information

Role
Programmer

Interests
Programming
Recent Profile Visitors
The recent visitors block is disabled and is not being shown to other users.

Would Unique paritcles slow down a physics simulation?
Fulcrum.013 replied to NoAbsoloops's topic in Math and Physics
Really single body dynamic involving a Newton laws and law of summing forces and torques. So its is simple for understand and implement. To extend it to multibody dynamics is require to compute response impulses of contacts only. This calculation involves again Newtons laws that is simple, and compuing of contact points and normals that can be very very complexive. But collision detection and prediction is a analitic geometry, not a phisics. And it really can be very complexive. For example convex aproximation of bodies is simple, but have a quaric computation complexity and gives inaccurate normals. Curved approximation is more complexive to understand, but in most cases have a linear complexity and normals is allways accurate. But with extending complexity of bodies, growth complexity of math involved to collisions detection . Some time it bring problems that very hard to solve, becouse tools of solutions is not works. For example the separation axis theorem will not work for curved torus, becouse it can not be cuted to convex bodies. Just looks like you studing to become a Software Engeneer, and all your dreams about making nextgen gaming engines. Fill difference with just a gamer that call something a phisics. To make engines you have to know how exactly it works inside and have skills to research something that nobody implemented yet. But to do it, at least required to know what same branch of phisics and mathematics involved, just to know what same to research and what same books can help with it. Just for example tehcnique with aprroximation of bodies by grids of paricles comes from CADs, where it used at least for 15 years for advanced soft bodies simulations (really nothing is ridgid in real word). Only problem that this technique have  is huge memory requirments to model a complexe world. So to be used for games it require just to find a fast algo to convert curved bodies to grid approximation when it waked up or have other moving objects that anble to affect it. In case guys that made this demo has find this algo, thay really genious. 
Would Unique paritcles slow down a physics simulation?
Fulcrum.013 replied to NoAbsoloops's topic in Math and Physics
Really, it depends from game concepts. Some developers have as target to develop a arcade games. Its game mechanics usualy use nonrealistic phisics. For example space arcade usualy use a aproaches that make a piloting of spasechip similar to piloting of airplane, but its can not by similar by phisics laws. So they just raplece real phisics by dummy game mechanics like maximal speed and turn radius. Simulate a realistic space piloting is not so hard, but it would be about impossible to play. Just becouse real spaceship controlled by two sticks and orbital piloting require a very advancet pilots skills. Really, top air force pilots spending a years of training to become a spaceships pilots. So space simulation game just can not be phisical realistic, until it ordered by NASA for real spacepilots trainig purposes. But other games really have to involve a realistic phisics into their game mechanics to be playable. For example a air combat simulators. 
Image sprites and shadow maps  beginner questions
Fulcrum.013 replied to Catalin Danciu's topic in 2D and 3D Art
Latest Microsoft api for drawing sprites is DirectDraw that depricated at DX 7.0 and removed at DX 8.0. OpenGL never have a sprite drawing API. Sprites can be easely simulated by 3D GAPI by maping texture to flat rectangular mesh maked from two triangles. It technicue is more powerful, becouse also give ability to stretch, rotate, and do with sprite anything else, that can be done with planar 3D object. 
Would Unique paritcles slow down a physics simulation?
Fulcrum.013 replied to NoAbsoloops's topic in Math and Physics
Really ridgid bodys dynamic is simple like as loop counter. Complexive is analitical geometry, that involved to computation of loop increment. Animated doors just have geometry conditions precomputed. Approximating body by stack of spherical particlles is just a way to simplify involved analitical geometry. Spheresphere and sphere  plane is simpliest of collision detection algos, also it crazzy efficient. It involves dot products only, that even CPU can compute each cycle per core. Also prediction of impacts for spheres is a simpliest of predictions algos, it is require just to find a distance betwin 2 straight lines at 4D space only. Really, anything genious is simply. Complexive is to find how to done complex task by simple tools. 
better approach angular_velocity.x = dot(body.inertionlaxe.x,angular_momentum) / Ix; angular_velocity.y = dot(body.inertionlaxe.y,angular_momentum) / Iy; angular_velocity.z = dot(body.inertionlaxe.z,angular_momentum) / Iz; then construct rotation matrix and rotate body locally, respectively to main inertions axes coordinate system. total angular momentum is sum of angular momentums for general inertion axes of body. total angular momentum is conservative respectively to global basis. but general inertion axes are local axes of body, so when body rotates, peojection of total angular momentum vector to body's general inertion axes are fluctuating, that cause angular velocities drift. In case total angular momentum vector have direction that close to middle inertion axe direction, fluctuations magnitude grows rapidly.

angular_velocity = angular_momentum.magnitude/(angular_momentum.dir*InertionTesnsor*angular_momentum.dir)

Would Unique paritcles slow down a physics simulation?
Fulcrum.013 replied to NoAbsoloops's topic in Math and Physics
Anycase yo have to detect and habdle collisions. How to handle depend from how much accurate and relistic simulation required by game concepts. But collisions is key feature in both cases. Othervice it will be racing of relativic horces in spherical vacuum. Really it could not be compared how much phisics uses rendering, dynamics simulation or game mechanics. Its all are uses phisics but different branchs. So it can be compared how mach calculation Its require. Just game mechanic commonly uses approach that can be integrated analiticaly, so uses less processor instructions, synamic simulation uses approaches that have be integrated numericaly, so uses more processor instructions. 
Would Unique paritcles slow down a physics simulation?
Fulcrum.013 replied to NoAbsoloops's topic in Math and Physics
You just mixing a terms "phisics simulation" and "phisic simulation engine component". It is not a same. Any games, excluding a card games,etc are phisic based. Difference is it involve approaches that physic simulation engine have "from box" or bring some other approaches only. Realy, game design is aprocess of adjustin a gaming engine universal abilities to concrete game concepts. Some approaches simulated by engine is independed from game concepts, otherwice it just not works. This part of approches adjusted by 100% datadriven way. But other approaches depends from game concepts, just becouse gaming phisics simulation is alweys compromise betwin accuracy, speed and gameplay concepts. So engine developer just can not predict wich compromise have be selected for concrete game. So this part of features have be coded/designed at game development time. For exammple let see to DooM'slike doors. Of course, it generally simulated by keyframe animation. But it still a phisic simulation, becouse animation have to match position limits, movement direction and speed constraints. So for this usecase keyfame animation is not a replacemet of simulation, it is a precalculated optimization of simulation. 
Would Unique paritcles slow down a physics simulation?
Fulcrum.013 replied to NoAbsoloops's topic in Math and Physics
Anithing into computing is a MATH, excluding layer betwin chair and keyboard only. Phisical simulation is MATH too. Really task of physic as science  is to describe world by math. Really any science began when it involves math. Software is whole math and nothing else. Really Rassion translation of "software" word can be translated back to English as a "computer mathematical fundations" 
Would Unique paritcles slow down a physics simulation?
Fulcrum.013 replied to NoAbsoloops's topic in Math and Physics
It depends from game concepts. For example has seen a mmorg game where assualt carbines range has been cuted to 150 meters and snipper ridles range has been cuted to 250 meters by game concepts. After this range boolet just dissapeared. Also it have no fast mooving objects like car o plane. Only humans, bears, dogs, goblins etc. But for example AK74 have direct shot range 440 meters and SVD rifle have direct short range 640 meters. Direct shot range mean that closer than it range it is no significant difference betwin straight line and real ballistical tragectory. So on concepted ranges shooting can be approached by line segments casting with enought to arcade gameplay accuracy. But it not is no simulation. It is bullet tragectories simulated by straight line segment 
Would Unique paritcles slow down a physics simulation?
Fulcrum.013 replied to NoAbsoloops's topic in Math and Physics
For first rendering a frame in video game is equalent of cout<<Result; into console application. And for main  rendering is also involved physical optics. Really mechanical dynamic is only a little bit of physics. Fire famge it fire famage. And shells damage is shells damage. When HE shell hit a ship it affect a explosive damage and have chance to burn a fire. But HE shell explode instanttly at hit. AP shell explode with litle delay. On flat tragectory it makes entry and outer hole in nonarmoured ship and explode outside. 
Would Unique paritcles slow down a physics simulation?
Fulcrum.013 replied to NoAbsoloops's topic in Math and Physics
just a simplified approach of fluid drag law with neogatin of turbulence and difference in air and water resistance. this approach may still accurate for not so big speeds and small waterline drift. all other is elementary computational geometry. It is many ways to obtain cross section from model both raster and vector exists. Raster way have example into DX SDK . wiki paper about fluid drag https://en.wikipedia.org/wiki/Drag_(physics) 
Would Unique paritcles slow down a physics simulation?
Fulcrum.013 replied to NoAbsoloops's topic in Math and Physics
Did you seen a spectrum game River Raider? It have same gameplay than this games. It can be interested on 10 year age. But it is to boring even at 15 age. Any game that have whechies/wessels/planes e.t.c have to model it. Quality of approach is second questions. But at can not be out of model. For example at Quake source code trajectories of all NPC is hardcoded together with firing. It is garbage approach of behavior, but it still approach. And than approach close to reality, than easely to adjust it work. For example again world of warships. It is good for gameplay to have accelerations of ships looks realistic. It have two ways. First ask a level disigner to do some dance with tambourine , and make other black magic with curves until acceleration will be seems realistic by his look thru the crystal sphere, and do same for each ship after each it tweak. And other way. We know a power of ship engine and his maximal speed from table data. Maximal speed mean that resistance force is equal to engine force. Engine power can be easely recalculated to force. Square area of front projection of underwater part can be easely calculated from geometric model too. So we easely calculate a Cx from just a two known digits. Only than we need to do else, is to add some response time to engine power changes. And its give really accurate aproach of ship acceleration for cost of nothing, that also automaticaly recalculable after tweeks of other parameters. 
Would Unique paritcles slow down a physics simulation?
Fulcrum.013 replied to NoAbsoloops's topic in Math and Physics
Did you seen a flight simulators like a DCS, LOCKON or IL2 Shturmovik, or even WarThumder where atleast half of IL2 physics has been cuted. Is gameplay of those games depends from anything else than phisics? Ok. Let consider situation when you have to burn something by fire under it. Simpliest approach to calculate energy of flame particles that hit it. Also what is difference betwin bullet and particle? Realy it is no difference. 
Would Unique paritcles slow down a physics simulation?
Fulcrum.013 replied to NoAbsoloops's topic in Math and Physics
It approach good to 9th class schoollboy. Really i has used same on this age. It was clone a of microsoft's game with monkey that throw bumerangs with angle to horizont. I has made it at 16 on pure basic od Soviet PDP11 clone. And even into this game physic simulation was a key feature of gameplay. But it really not good for AAA class games. For example World Of Warships modeling even fuse delay time. It why destroyers abd light cruiser recives a damdge from armore perching shells only on high incident angle. and low damage on direct shoot tragectories.

Advertisement