


Mech's movement animation concern
#81
Posted 28 June 2012 - 02:11 PM

#82
Posted 28 June 2012 - 02:14 PM
#83
Posted 28 June 2012 - 02:17 PM

Don't people discuss football and stuff even if they are completely unable to affect the overall result?
#84
Posted 28 June 2012 - 02:25 PM

battletech did this so maybe it will be here...

EDIT this picture is taken from battletech univers book
Edited by Blaze32, 28 June 2012 - 02:27 PM.
#85
Posted 28 June 2012 - 02:31 PM
#86
Posted 28 June 2012 - 03:52 PM
Skadi, on 28 June 2012 - 02:11 PM, said:

My main post, btw, holded a question about if this kind of animation is somewhere final or just a draft for now.
And, btw, we've got ourselves an answer from Dennis de Koning, that it is very close to be final (the general idea). Look at page 2.
Actually, the following thread (and my updates to the initial post) are about trying to explain, that swaying may be from minimal to none, and the "realistic" effect can be achieved by just changing the way the legs and pelvis of the mech move.
BduSlammer, on 28 June 2012 - 02:31 PM, said:
Can you please post a link and time, where? Because neither in official videos, nor in leaked one i saw an atlas, commando or jenner moving properly.
BarHaid, on 28 June 2012 - 01:24 PM, said:
Not exactly. The cockpit moves up and down + shakes. Actually, it should also sway a little.
But the main idea is about how legs and pelvis should move in order for mech not to look like a plastic stiff doll, cockpit movement being a consequence of their movement =)
#87
Posted 28 June 2012 - 04:07 PM
Well spotted I knew I saw something strange watching the video but I could never quite pin down exactly what it is!
Good work!

#88
Posted 28 June 2012 - 04:52 PM
#89
Posted 28 June 2012 - 07:35 PM
Undead Bane, on 28 June 2012 - 12:14 PM, said:
Sorry, but still no. Even if the gyro would be super-duper-mega heavy + rotating at 1mil rpm, it still would not allow the mech to stand absolutely upright on one foot.
Gravity will just create a momentum and make the gyro (and mech with it) precess, like shown in a video with gyro i've posted here.
Here is the video again. Start looking at 1:47, that's where the situation with CM off the supporting area, that is described in initial post:
The main idea of a massive gyro is compensation of dynamic loads, like hits, steps (meaning the loads, produced from foot's impact with the ground). They, unfortunately, cannot compensate static ones.
But I wasn't referring to static loads, gravity wins every time. With a vector component (excuse me if this is a bad use of the term), the term "momentum" comes into play and the moving mech would tend to resist gravity long enough for the transfer of load/balance to the opposing leg.
Edited by TLBFestus, 28 June 2012 - 07:36 PM.
#91
Posted 29 June 2012 - 09:27 AM
Rag doll physics implemented as well would be cool. I recall reading in one of the older novels a mech getting nailed in the right arm (I think by a gauss rifle) and the momentum forced a torso 360. A radical example, obviously if players could do this issues would arise, but a mech going rag doll at death would be cool.
Also, it was always an eye sore for me to see a mech destroyed mid jump, land on his legs, stand perfectly still for a second, then fall over and blow up.
#92
Posted 30 June 2012 - 12:21 PM
grimzod, on 28 June 2012 - 08:25 AM, said:
We have no 13 meter tall combat robots today. Please, when were trying to promote suspension of disbelief by asking for realistic movement in a game about 3049 mechs, don't bother with our robots do this so in the fake future robots won't do that.
Human size robots are considered big by today's standards and that's what I was referring to. I'm not talking big as in Mechs. I was meerly creating a point of reference. But anyway mechs designed like the ones in mechwarrior WOULD move stiffly. By no means would they move smoothly or gracefully.
Edited by Irreverence, 30 June 2012 - 12:32 PM.
#93
Posted 01 July 2012 - 02:00 PM
just made a quick mw4 vid. best example of good mech movement. running has different mech animation/movement than walking and the mechs sways a little bit to achieve balance. the mechs here really looks alive and controlled(movement/balance) by mechwarriors using their neurohelmets. when they are hit, it really rocks the mech because mechs don't have shields, only armor.
Edited by zer0imh, 01 July 2012 - 02:06 PM.
#94
Posted 01 July 2012 - 02:13 PM
#95
Posted 01 July 2012 - 02:41 PM
syngyne, on 01 July 2012 - 02:13 PM, said:
it's the other way around...
look at the mech movement of mechs here...
they have very stiff movement in relation to the terrain. look at the movement of the light mechs, they are floating especially when running(check out 7:15, the light mech is literally floating and observe the upright position all the time. you can't pull of that speed on that running posture. running animation is same as walking animation which is truly wrong. the only good animation is when their legs are crippled or when they get destroyed.
the neurohelmet...
http://www.sarna.net/wiki/Neurohelmet
the reason why mechs move and have balance/posture.
if we be more strict. every mech has it's own unique posture because every pilot's way of balance is different.
p.s.
a BATTLEMECH is not a TANK.
a BATTLEMECH is not a ROBOT.
a BATTLEMECH is a machine that is controlled by a pilot using a NEUROHELMET to give posture, balance and movement(running and walking). that's why mechs should move like humans do but in a larger/heavier scale.
Edited by zer0imh, 01 July 2012 - 03:03 PM.
#96
Posted 01 July 2012 - 03:01 PM
BlindProphet, on 27 June 2012 - 07:28 AM, said:
This would of been my reply. A good example would be the M1A1 Abrams tank (in real-life) targeting system. Even though the tank is moving forward or backward, the turret can pivot and lock (on target), keeping the barrel perfectly stable even over rough, moving, bumping terrain. As far as your HUD, terrain would be moving but target would remain locked and in the center of your screen.
#97
Posted 01 July 2012 - 03:13 PM
Ducks Guts, on 01 July 2012 - 03:01 PM, said:
Well, M1A1 utilizes different type of gyro, if I am not mistaken.
A bit of gyro background.
There can be actually two types of gyro: massive balance gyro (used mostly in sattelites nowadays) and gyro sensor (used pretty much everywhere).
Massive gyro is used to apply actual momentums and forces to the structure. It's really massive, as it is to produce precession force/momentum enough to move the structure. In BT mechs utilize this type of gyro. Also, this type of gyro makes the structure, that it is installed in very stable against hits and other short dynamic loads.
Gyro sensor is a small, fast turning gyro, that's floating free in it's basket. When the structure rotates, the gyro does not change it's position (against the ground/any other support), but it does change position against the structure it's built in. Basing on the difference produced the countermeasures are done - for example, actuators move on a robot, planes on the wings move on a plane, thrusters are fired on a sattelite. BT mechs use pilot's middle ear as such gyro. Our middle ear is such gyro, sort of, it allows us to understand, how we are moving and what is our position currently.
And, neither typo of gyro neglects the need for support under CCM - it does not allow the structure to float in air.
That's why I do understand and like MW4 animations and have questions to MW:O animations.
Edited by Undead Bane, 01 July 2012 - 03:14 PM.
#98
Posted 01 July 2012 - 03:17 PM
Undead Bane, on 01 July 2012 - 03:13 PM, said:
A bit of gyro background.
There can be actually two types of gyro: massive balance gyro (used mostly in sattelites nowadays) and gyro sensor (used pretty much everywhere).
Massive gyro is used to apply actual momentums and forces to the structure. It's really massive, as it is to produce precession force/momentum enough to move the structure. In BT mechs utilize this type of gyro. Also, this type of gyro makes the structure, that it is installed in very stable against hits and other short dynamic loads.
Gyro sensor is a small, fast turning gyro, that's floating free in it's basket. When the structure rotates, the gyro does not change it's position (against the ground/any other support), but it does change position against the structure it's built in. Basing on the difference produced the countermeasures are done - for example, actuators move on a robot, planes on the wings move on a plane, thrusters are fired on a sattelite. BT mechs use pilot's middle ear as such gyro. Our middle ear is such gyro, sort of, it allows us to understand, how we are moving and what is our position currently.
And, neither typo of gyro neglects the need for support under CCM - it does not allow the structure to float in air.
That's why I do understand and like MW4 animations and have questions to MW:O animations.
So, it sounds like one of two choices: base it on middle ear of pilot or middle ear of mech. No? If the pilot is the mechs middle ear, how is the difference depicted? Same difference yet POV would be different ie: pilots head would move more/faster than a mechs.
Edited by Ducks Guts, 01 July 2012 - 03:22 PM.
#99
Posted 01 July 2012 - 03:24 PM
#100
Posted 01 July 2012 - 03:50 PM
Ducks Guts, on 01 July 2012 - 03:17 PM, said:
hm. Ok, let me explain, how a system, described in BT could work, if it was made.
Given:
We have a neurohelmet, a VERY heavy thing, that rests on the pilot's shoulders and reads pilot's middle ear signals (+ several brain signals also). Pilot is strapped to the mech's chair, he cannot move freely.
We have myomers, the mech's muscles-drives. They are controlled by mech's computer.
And we have a power gyro in the mech's CT.
Question: how can this possibly work (given that the mechanics were developed by people who knew at least something about engineering)?
Ok, here we go. In the first place, mech's computer has numerous "standart" movement pre-recorded to it's memory - pretty much like the human's reflexes. Like if you start to fall down to left side, you try to move your leg there to have a support + use hands to make a balancing impulse for CCM. Also, it has more complex movements recorded, like if a pedal is pushed, start stepping.
Mech is not controlled directly by brain. It is speciall written in gamebooks, that it didn't work out (and this is, actually, quite realistic) at all.
So what happens when a mech starts falling down? The system reads the signal from human's middle ear about the movement, that the system recognizes as "falling down" and gets a direction from it. Additionaly, it recieves a command from a pilot "prevent falling down". So, the mech starts to use it's built in reflexes to countermeasure the fall to an already known side.
Where does the power gyro comes in play? It depends. If it is a static gyro, that is unmovable against the chasis, then it is just a factor that reduces the velocity of a fall, damps hits and so on. But more likely, it's a dynamic gyro, it's basket is moveable by some kind of drives. Then, aside from the static gyro functionality, it also is a part of reflex "prevent falling down". Like your (human) hands - you wave them. Mech will turn it's gyro in a way to produce precession momentum, that counters the momentum, that makes the machine fall down. But, of course, as you (human, again) cannot most of times regain balance by simply waving your hands, the mech cannot do that only with his gyro - it needs to move legs, torso etc.
Edit: typos
Edited by Undead Bane, 01 July 2012 - 04:04 PM.
1 user(s) are reading this topic
0 members, 1 guests, 0 anonymous users