Jump to content

Mech's movement animation concern


163 replies to this topic

Poll: Mech's movement animation concern (212 member(s) have cast votes)

How the animation should look like?

  1. "Realistic", like, say, in MechWarrior 4. Mechs should give a giant walking machine feelin. (164 votes [77.36%] - View)

    Percentage of vote: 77.36%

  2. "Simplified", like they used to be in MechWarrior 2 (and, actually, in MWO currently). I'm totally ok with current movement. (39 votes [18.40%] - View)

    Percentage of vote: 18.40%

  3. I don't care at all. They may even "moonwalk". If they look like mechs, it's enough for me. (9 votes [4.25%] - View)

    Percentage of vote: 4.25%

Vote Guests cannot vote

#21 Nih

    Member

  • PipPipPip
  • 57 posts

Posted 27 June 2012 - 07:59 AM

View PostBarHaid, on 27 June 2012 - 07:53 AM, said:

Yeah, the moving cockpit bugs me. I would have preferred to have the cockpit seem stationary, and the view outside bump up and down. As it stands now, it looks like your chair is on springs.


This... is actually a good point. I don't know that it wouldn't make me sick, but I might like the option to see. Maybe a toggle. No reason not to have toggles! Toggles make everyone happy!

Did I mention I like toggle-able features?

#22 Fugu

    Member

  • PipPipPipPipPipPipPip
  • 527 posts

Posted 27 June 2012 - 08:00 AM

We seem to have quite a lot of engineers around here. Yesterday it was all about how an Atlas could possibly only weigh 100 metric tons.
Of course applying logic doesn't quite work for a game about giant BattleMechs.

But besides that, it's always an interesting read. Carry on.

#23 Grokmoo

    Member

  • PipPipPipPipPipPip
  • 289 posts
  • LocationWashington, DC

Posted 27 June 2012 - 08:08 AM

A very nice analysis. Hopefully as time progresses we will see the devs continue to refine the more basic systems, like movement animation.

#24 bobhound

    Member

  • Pip
  • 19 posts
  • LocationHesperus II

Posted 27 June 2012 - 08:08 AM

While it's neither here nor there for me if it's realistic or not, I think there are valid concerns raised by Undead Bane. But at the end of the day what I want to see is better animation, simply. At the moment some of the movement looks a bit stiff which just isn't very pleasing to the eye. Hopefully it'll improve!

Great suggestions!

#25 Clan Warrior

    Member

  • PipPipPipPipPip
  • 110 posts
  • LocationClan Jade Falcon

Posted 27 June 2012 - 08:09 AM

This is how real life mechs walk :) I'll take MWO mech walking over that video anytime ;)



#26 BlindProphet

    Member

  • PipPipPipPipPipPip
  • Bridesmaid
  • Bridesmaid
  • 228 posts

Posted 27 June 2012 - 08:11 AM

View PostUndead Bane, on 27 June 2012 - 07:49 AM, said:

Sure thing, i did. As i've said, i'm really a bipedal machines designer :) The problem is, that to have gyro to stabilize your machine completely it must be VERY big and heavy. So in mechs gyros are mostly used to compensate dynamic loads, not static ones.
BTW, in real life bypedal gyros are not used yet exactly because of their heaviness.


Ok going to point out that Mech Gyro's in BT, are large, and heavy. They're also more advanced than our own current technology level. Consider for a moment they're also running fairly small, self contained, safe fusion power plants to run these things as an example

#27 Undead Bane

    Member

  • PipPipPipPipPipPipPip
  • 535 posts

Posted 27 June 2012 - 08:13 AM

View PostClan Warrior, on 27 June 2012 - 08:09 AM, said:

This is how real life mechs walk ;) I'll take MWO mech walking over that video anytime ;)



Actually, it would be more proper to look at this:


As for mass/drives power Asimo is way closer to what mechs are =)

#28 Irreverence

    Member

  • PipPipPipPipPipPipPip
  • 777 posts
  • LocationFlorida

Posted 27 June 2012 - 08:13 AM

View PostClan Warrior, on 27 June 2012 - 08:09 AM, said:

This is how real life mechs walk ;) I'll take MWO mech walking over that video anytime ;)



It's hard to have a good representation of how a mech would walk since none exist, but this is pretty true, lol. With our current level of technology all we can do is speculate and watch videos like this, haha.

#29 Deaths Charger

    Member

  • PipPip
  • Elite Founder
  • Elite Founder
  • 21 posts

Posted 27 June 2012 - 08:15 AM

well if u look at the supper kolled magnets that we have to day an the basic goryos that we also have yes u make 1 thats abut 50 lbs in size. put 4 ina mech. or on some 1 then try swaying.

#30 Undead Bane

    Member

  • PipPipPipPipPipPipPip
  • 535 posts

Posted 27 June 2012 - 08:21 AM

View PostBlindProphet, on 27 June 2012 - 08:11 AM, said:


Ok going to point out that Mech Gyro's in BT, are large, and heavy. They're also more advanced than our own current technology level. Consider for a moment they're also running fairly small, self contained, safe fusion power plants to run these things as an example


Khm. Gyros will stay gyros always =) It's a big fast rotating mass, producing gyro momentum. BT universe says nothing about finding out new chemical elemets => max density and, therefore, mass of a gyro remains quite same for a size. So, there can be a rotation speed. But again, even if you take something like 50000 rpm and try to evaluate, how big the gyro should be to completely stabilze static forces, you will get a mass very close to the mass of the mech itself.

Aside from that, this is not how gyro actually operates. If you try to tilt a gyro (what gravity will try to, if you lift one leg), it will start precessing, like this:


So, the mech standing upright on one foot will look like this gyro at 1:49 of the video

;)

Edited by Undead Bane, 27 June 2012 - 08:22 AM.


#31 TKG

    Member

  • PipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 182 posts
  • LocationThe Sandhills of NC

Posted 27 June 2012 - 08:21 AM

View PostBlindProphet, on 27 June 2012 - 07:28 AM, said:

Have you taken into account that Mechs have gyroscopic stabilizers which are there to keep mechs balanced?



Its not just the gyro, it's also the neuro helmet which uses the pilot's own sense of balance to keep the mech stable on the move.


I have to agree with the guy that started this thread, In Mechwarrior 4 there was quite a bit of cockpit sway on a number of mechs, generally it was greater or lesser depending on where the heaviest weapons were mounted. The atlas proportionately had little because it's heaviest was in the lower torso where as the hunchback had a lot due to the gun being over the shoulder pretty much. I might add trying to hit things using a dasher with maximized speed is one hell of a task due to it's sway. Should it be there yes, but I'd add it ought to be proportonate to the mech model's layout rather then one general motion.

#32 Jonneh

    Member

  • PipPipPipPipPipPip
  • 334 posts
  • LocationUK

Posted 27 June 2012 - 08:25 AM

I started to read, and then got bored.

If I wanted to fight with real robots I'd cryo myself until 3049. As it is, I'm happy with the virtual slightly-less-than-perfect-real-world-animations.

#33 Dennis de Koning

    Art Director

  • Developer
  • Developer
  • 118 posts
  • LocationVancouver, B.C.

Posted 27 June 2012 - 08:27 AM

Believe it or not, there was a lot of discussion and research that went into the animation of the 'Mechs.
Many (most) 'Mechs are humanoid and so scaling cues alone are not enough to make a 'Mech look less like an environmental suit than a giant walking tank. The more animation looks and feels like natural human movement, the harder it is to suspend belief. We wanted the 'Mechs to look like machines, if this means that it has a stiffer gate, then so be it. Also, on a note of balance from the initial post; sway, for the purpose of balance is lessened the faster one moves. In other words, the slower you walk, the more sway is necessary for you to keep balanced. When running (or walking at 60kph), you (and/or a 30-100t 'Mech) will have almost no sway at all; that extra energy will be converted into forward motion instead.
As for the cockpit movement: It is attached to the 'Mech and therefore the movement is a realistic translation. This is necessary because the (exterior) architecture of some 'Mechs is visible from within the cockpit ie, Hunchback, Dragon and others coming down the pipe. Imagine the effects that more sway would have on a users aiming and desire not to puke.

#34 Canned Heat

    Member

  • PipPipPip
  • Philanthropist
  • Philanthropist
  • 50 posts
  • LocationNot Near Uranus

Posted 27 June 2012 - 08:37 AM

DARPA and Boston Dynamics are working on some amazing things now...granted MWO takes place far into the future and we cant compare now.

#35 Undead Bane

    Member

  • PipPipPipPipPipPipPip
  • 535 posts

Posted 27 June 2012 - 08:48 AM

View PostDennis de Koning, on 27 June 2012 - 08:27 AM, said:

Believe it or not, there was a lot of discussion and research that went into the animation of the 'Mechs.
Many (most) 'Mechs are humanoid and so scaling cues alone are not enough to make a 'Mech look less like an environmental suit than a giant walking tank. The more animation looks and feels like natural human movement, the harder it is to suspend belief. We wanted the 'Mechs to look like machines, if this means that it has a stiffer gate, then so be it. Also, on a note of balance from the initial post; sway, for the purpose of balance is lessened the faster one moves. In other words, the slower you walk, the more sway is necessary for you to keep balanced. When running (or walking at 60kph), you (and/or a 30-100t 'Mech) will have almost no sway at all; that extra energy will be converted into forward motion instead.
As for the cockpit movement: It is attached to the 'Mech and therefore the movement is a realistic translation. This is necessary because the (exterior) architecture of some 'Mechs is visible from within the cockpit ie, Hunchback, Dragon and others coming down the pipe. Imagine the effects that more sway would have on a users aiming and desire not to puke.


Tx for the reply! Yes, sorry for the mistake, the faster you move, the smaller the sway will be.
And yes, i do understand the idea about the cockpit view being linked to the model. However, if you have a bit of time, please, take a look at the updated post about the kinematics of movement.
The kinematics i posted really minimizes sway to being something like half a meter at 10 meters height. Actualy, this exact kinematics were used to produce a testing stand to see, how the simplified scheme works.

Also, about the fact that machine should look like a bulky machine. Yes, of course it should and I (and all the forum members, i suppose) can understand this concern. However, my guess is if you add just a slight movement according to the kinematics i have just updated the topic with, it will still look VERY bulky (as it is a simplified human model), but a bit more realistic.
IMHO, it's not what moves make a mech bulky, it's how it moves.
The real human moves VERY dinamically. Human does not have constant velocity of movement at any given moment of time. The machine, however, has, as it has a higher mass/power ratio, has limited speed of drives and so on. So, the machine's movement are more discrete, they reach top velocity fast, then stop fast too.

Edited by Undead Bane, 27 June 2012 - 08:54 AM.


#36 Lucky Strongarm

    Member

  • PipPipPipPipPip
  • Elite Founder
  • Elite Founder
  • 112 posts
  • LocationOrlando, FL

Posted 27 June 2012 - 08:48 AM

View PostUndead Bane, on 27 June 2012 - 07:22 AM, said:

Hi everyone!

After viewing some in-game video footage (official and "unofficial" alike) I, as a somewhat walking bipedal machines designer, became concerned with how unrealistic the animation of movement is (to be objective: MW:LL has completely the same issue).


First of all, are you a "somewhat walking, bipedal machines designer", or do you design "somewhat walking, bipedal machines"? If you're the first, I'm sorry about your disability. If you are the second, does that mean you're not very good at it? Do they fall over alot? If you're the second, I don't think you have any place to complain about how 'mechs walk in this game, since your own walking, bipedal machines only do it "somewhat".

But your concern is duly noted. Now, let's discuss the impossibility of jump jets generating enough kilo neutons worth of thrust to offset the gravitational force exerted on even a light mech as rapidly as depicted in the game, given their relative size. (No, please don't.)

Edited by Lucky Strongarm, 27 June 2012 - 08:49 AM.


#37 Undead Bane

    Member

  • PipPipPipPipPipPipPip
  • 535 posts

Posted 27 June 2012 - 09:01 AM

View PostLucky Strongarm, on 27 June 2012 - 08:48 AM, said:


First of all, are you a "somewhat walking, bipedal machines designer", or do you design "somewhat walking, bipedal machines"? If you're the first, I'm sorry about your disability. If you are the second, does that mean you're not very good at it? Do they fall over alot? If you're the second, I don't think you have any place to complain about how 'mechs walk in this game, since your own walking, bipedal machines only do it "somewhat".

But your concern is duly noted. Now, let's discuss the impossibility of jump jets generating enough kilo neutons worth of thrust to offset the gravitational force exerted on even a light mech as rapidly as depicted in the game, given their relative size. (No, please don't.)


Heh, very funny. "Somewhat" meant that the main type of bipedal machines i'm currently designing is exoskeleton, and it has way less concerns about the actual balancing. But I also participated in creation of an autonomous stand, that was created exactly to test, how the balance issues affect the walking machine.

As for the second part of your reply, here you go: http://en.wikipedia..../Verisimilitude
And, aside from that, the main thing, that may raise questions about JJs is... FUEL, not the landing dynamics. In CBT, JJs are not only used to jump, but also to compensate vertical velocity at landing, otherwise pilot completely ruins the mech and, possibly, himself.

Edited by Undead Bane, 27 June 2012 - 09:03 AM.


#38 Lucky Strongarm

    Member

  • PipPipPipPipPip
  • Elite Founder
  • Elite Founder
  • 112 posts
  • LocationOrlando, FL

Posted 27 June 2012 - 09:05 AM

View PostUndead Bane, on 27 June 2012 - 09:01 AM, said:


Heh, very funny. "Somewhat" meant that the main type of bipedal machines i'm currently designing is exoskeleton, and it has way less concerns about the actual balancing. But I also participated in creation of an autonomous stand, that was created exactly to test, how the balance issues affect the walking machine.

As for the second part of your reply, here you go: http://en.wikipedia..../Verisimilitude
And, aside from that, the main thing, that may raise questions about JJs is... FUEL, not the landing dynamics. In CBT, JJs are not only used to jump, but also to compensate vertical velocity at landing, otherwise pilot completely ruins the mech and, possibly, himself.


I'm just glad to hear you're not disabled!

View PostCanned Heat, on 27 June 2012 - 08:37 AM, said:

DARPA and Boston Dynamics are working on some amazing things now...granted MWO takes place far into the future and we cant compare now.


Sweet kicks on that droid.

#39 cinco

    Member

  • PipPipPipPipPipPipPip
  • Knight Errant
  • 509 posts

Posted 27 June 2012 - 09:09 AM

bi-pedal=/=human

the entire structure of the mech doesn't resemble humans one bit. it has no muscle, no ability to use inertia to propel, etc.

might as well start studying how chickens walk for the catapult. wrong, complete different mobility system. walking machines will not resemble the human gait. youtube a video of the robotic pack mule. that's really what it should look like. 4 dog legs.

View PostNih, on 27 June 2012 - 07:59 AM, said:


This... is actually a good point. I don't know that it wouldn't make me sick, but I might like the option to see. Maybe a toggle. No reason not to have toggles! Toggles make everyone happy!

Did I mention I like toggle-able features?


no. more work for the devs for a false reason. nausea is induced by shaking of the camera, ie movement to the entire environment. cockpit movement, there's no change to the environment or camera. so please don't mince arguments.

Edited by cinco, 27 June 2012 - 09:11 AM.


#40 Wiley Coyote

    Member

  • PipPipPipPipPipPipPip
  • 612 posts

Posted 27 June 2012 - 09:19 AM

View PostClan Warrior, on 27 June 2012 - 08:09 AM, said:

This is how real life mechs walk ;) I'll take MWO mech walking over that video anytime ;)



The nemesis to this design: stairs. Just like ED-209 from Robocop.





1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users