Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[PytonPago]

Hi there, i just started to thing about getting stuff at my model get a bit complex in the design:
 - having rigid axle suspensions (single and multiple wheels)
 - gear-complex simulation (simply put, all gear rods between engine-gearbox-wheel axles-functional equipment (like a winchdrive) whyte proper rotation speed proportions of each element even to the actually set gear) (thats 3 parts for rods multiplied by 7)
 - driving wheel complex simulation (at the same time the suspension works, hawing around 6 different movable parts)
 - other gear driven equipment (in my case the BM-21 launcher that will be put on the chase later (maybe 12 or so))
 - Motor fan rotation
 - a lot of little movable things :
 -- door handle
 -- door window control (mech. or electronic)
 -- gear setting
 -- sidelights and other internal stuff (possible fans, all function-able buttons, internal GPS or control LKSs etc.)

As there will be no different model for the cockpit and at the hi-detail LOD range could be maybe 10 such vehicles doing different stuff, how does that affect hardware usage and where should I draw the line in those systems ? My ural has just a 6 wheel base (single and double rigid axle suspension):








 ... doe, there are much more complex possibilities (a 8 or 10 wheel base whyte the same level of sim-modeling of the hi-weight crane at its back like the already described) and yes, the graphics involved to the rocket launcher at the actual one, where the controls for it are at theyr base, that i would like to be used directly there:



 (so it will have another set of complexly-movable meshes to the model (not so much as it horrifyingly looks like)). So generally, there is a lot of stuff that will dependently move at once and I have a bad feeling about putting that final thing to Outerra that way, but no idea how should it be simplified reasonably and still multifunction-ally at a great number of details.

There still be a little complexion added in the model itself at different levels (engine and its compartment is not finished, its a open structure vehicle, so at least a simplified engine-mock up should have been done as it is visible at a great number of angles, what was my  motivation to make the gearbox-complex working the way as descried), so yes the model can end up hi-poly not looking like that at first look ..

[ZeosPantera]

You are taking this way too EXACTLY CLOSE TO REALITY as I have come to expect from a fellow Outerran. Keep up the good work and don't forget to pressure Cameni and Pig to the breaking point!

I would love full physics drivetrains to match reality. Here is one I attempted in Garrys Mod.

[PytonPago]

You are taking this way too EXACTLY CLOSE TO REALITY as I have come to expect from a fellow Outerran. Keep up the good work and don't forget to pressure Cameni and Pig to the breaking point!

I would love full physics drivetrains to match reality. Here is one I attempted in Garrys Mod.

 ... : P .... not so close as it seems, i just make primary objects rotate at realistic proportions to the gear set ... your way in garrys mod is a whole different thing as it simulates the gears at a (maybe something simplified, but still) real-world contact pressure gear system (yes, i had seen some engines too, made even whyte additives to fuel, nice play) ... the problem is, its a real crack on CPU, as it has to be got in OT as just some vehicle in the whole world of other simulations taking place at once, its much better to just simulate the rotation proportions correctly, where special things will be taken at aether a very well done animation, or a potential situation dependent combined-animation system (making some kind of variations at each animation, where the specific interaction places would be combination-able whyte others at a function to theyr actual position and/or characteristic constant).

 .... so my point is to have some assumption of how much things can be simultaneously be done on a vehicle in this simplistic way of mine whyle not getting OT panic when a Grad battalion gets theyr salvo fired at the Hi-res model range, at the fact to have so much movable objects at the model (as i counted its just for the chasis around 60-70 pcs. plus there would be around 30 in cockpit and 30-40 around the BM-20 module) ... making sure a quarter of them is working at the same time is a little scaring to me. Not sure if OT can handle 12 such machines at once.

 ... and yes, I dont wont to pressure them too much, as they have a lot to do whyte the environment at the time. Still would like to have this more complex way of simming a vehicle set up, when they got ready whyte it so i can make a half-step by half-step how to for people to get the model flow started for the mil.-sim. dream of mine for test-purposes. ( ... while not making them angry enough to make me wake up some nice morning in the middle of siberia hawing a mad smiley tattoo on my hand, signed by the OT team and the message: "dont EWER get back !" )

[ZeosPantera]

I think a basic mechanical simulation system could work with calculated reductions / rotation proportions as long as somewhere you can define a max torque +/- and max rpm rating and that way you could actually damage fake gears, driveshafts and wheel mounts. That would be applied to all moving parts and be amazing in a game. Just do a quick monitor of 12 rotating parts / suspension components and keep track of the stresses of each. Like a damage model but much simpler and cooler.

[PytonPago]

I think a basic mechanical simulation system could work with calculated reductions / rotation proportions as long as somewhere you can define a max torque +/- and max rpm rating and that way you could actually damage fake gears, driveshafts and wheel mounts. That would be applied to all moving parts and be amazing in a game. Just do a quick monitor of 12 rotating parts / suspension components and keep track of the stresses of each. Like a damage model but much simpler and cooler.

Sounds like an idea. ...

[PytonPago]

A little step closer ... gears finally rodded ... now just to make the engine look more like real machinery, make the steering sys. and interior ... got a bad feeling about the poly-count doe.









Been a little slow lately.   

[ZeosPantera]

I would rather have the model look AMAZING and lose 30% of my frames then look poor and loose 2%.  ROCK ON!

[PytonPago]

I would rather have the model look AMAZING and lose 30% of my frames then look poor and loose 2%.  ROCK ON!

Make Cameny and co. kill some future modellers, EYY ? Just dont take me as a target, if they give you the Tiger tank, i wont to have at least a T-34 value. 


I forgot to add the trencher :

[PytonPago]

I just got to my spare wheel ... i wont to simulate later its release from a hydraulic hold. How should i rig/setup the wheel itself, so it would after the release act as an separate OT-physic object, that could bounce down a hill ? Or should i do a separate one, that would be spawned at its place once the animated one dissapears after the release ?

[Chaoz]

man I do hope Outerra can handle your level of detail, if so it'll be awesome

[ZeosPantera]

Pago you are just the right type of crazy.

If it can't Chaoz, we will force Cameni to make it handle it.

I wonder how hard it would be to code the actual changing of a tire. Tire damage occurs (essentially just lowering the radius of the tires and increasing the drag along the ground).. You would need to employ a jack (can be completely coded and just lift the truck BUT DON'T TRY IT ON A HILL!) then you would hold X to remove the busted tire.. Then do the same to get your new tire off the truck, car etc and walk it/roll it over and hold X to install. Remove jack and Blamo.. You would of course want to keep the broken tire for repair as buying a new one is expensive in the wasteland that is new Earth.

[PytonPago]

 ... well, if they land in Russia, they wont have a problem whyte spare parts to urals even after thousands of years  ... the same goes for T-34s  The changing would be nice, great idea! Seems then the disappear-spawn method would be better then, (anims. for putting them right on will be a little problem doe, have to change the axes a little to make it possible) ... that is, if the wheel wont roll down the himalayas after releasing, whyte me running for it.    (will need a little fixing for it)   ... getting the spare back would not be a problem, the holder is just a hydraulic lifter, so placing it back at it and getting it up is no problem. Will probably need a broken-tire model - would be classy if the tire could behave like mold-able cloth in a lighter version (would be a nicer look if broken on the move). But i still dont know - the tire should be a different model file all together then? It could be used even as a random object it seems ...   

[PytonPago]

Have good and a bad news ...
 .... good is, the main model is ready (the main functional parts that is), so the work is going to just make the wheels turn, transmission rods rotating, trencher rolling, and so on ...
 .... bad is, my NB is a pain lately .. for some reason it messes up the colada export out of blender (just makes the meshes stuff twisted around theyr origins for some reason), so i would like to ask someone to try do it for me (im in school-town, so till i go home it will be another 2 months to do it myself - testing season pushing too a lot) .. so export it yourselves and enjoy to play whyte it as you all like

--- freeware-modify freecence applied ! ---
        And of course OT use free.

After school season, i will make the rigging/scripting/importing job finished then as i intend, and after this transport version runs smoothly (i may still make it a back-section later), and so, when my knowledge of scripts comes up by this 4320-31 Ural, i will go on whyte the BM-21 Tornado-G project.

And the blender file of course : Ural - base version

There is a number of buttons in the cabin, the trencher has a back-part that has to run left-right to make the rope go aligned, rods as many as it gets. But the tornado will have them numbers quadrupled it seems. :-/

... P.S.: Cameny, i would ask you to move my thread to the vehicles section, if you find some spare time please. Should probably done it sooner, but cant figure it out ... im just left-handed on forums. :-/

[PytonPago]

 ... i was just bored, so i done the back-case whyte open-able sections. Will do the releases for them working, and there would be a good thing to script the sections fall, if you dont fix em (or when you try it at an bad truck angle ... will see).

 ... the back-case model : ural-4320-31









 .. how does one make the broken wheels go interact whyte the groun ? .. would it bee enough to have two tires (normal and broken) at the model and just switching it in script, when damaged ? Or will there be a need of some other way (and if they need to be separately taken in the importer) ?

[cameni]

Broken in what way? Deflated or unmounted .. probably by setting new tire radius in script, and changing some other tire properties too.