522 replies to this topic

[V Swoops Grym]

I haven't read all 14 pages, so this may have been said. And to be honest I think this is a great addition to the game and much needed. However, in our current year, 2013, we have inclinometers. It measures the angle of slope of the current grade you are on. You can't add this kind of dimension without giving us something in the GUI that gives us our current angle. My 2 Cents. Thanks for adding this BTW.

[Disapirro]

MrMadguy, on 28 June 2013 - 08:24 AM, said:

Overall I like the idea, cuz I understand, that current state of moving physics in game is very poor.

1) But I also don't want balance to be futher shifted towards light mechs. They are invulnerable enough even now. I've got good hardware and relatively good ping, but it's still very hard to hit them. Sometimes one spider is running among whole of your team and lives longer, then my 100 ton Atlas, while being able to kill some mechs 1vs1. Futher increase of difference in speed/manoeuvrability may cause even more problems.

Please, light/medium mechs have had to suffer far too long with being overly vulnerable without the speed cap lifted, and with the pinpoint PPC alpha B S.

[tuffy963]

Quote

Really, do you think a person of small stature, and say a chimpanzee, should have the same climbing ability?

I am not sure I understand your underlying point, but let me try to build on the OP's point...

I believe a chimp can host it's lighter body weight into a tree faster than a silver back gorilla. Of course, this example only takes us so far as the chimp and the gorilla have physical sizes that are congruent with their respective weight and power, so it is pretty clear in this example that the chimp is smaller, more agile, and faster. Unfortunately, MWO does not always follow this simple, easy to understand model. In MWO there are mechs that are presented as visually larger, but have a weight that less than a visually smaller mech. These same mechs, while larger, are also more agile than these heavier, smaller mechs. A good example here is the Quickdraw (bigger, lighter, faster) and the Cataphract (heavier, smaller, slower). So, I think the question still stands....

Why does a physically larger framed, lighter, faster weight mech make a poorer hill climber than a smaller framed, heavier, slower mech?

The lack of internal consistency is pretty annoying to me...

Edited by tuffy963, 28 June 2013 - 08:38 AM.

[Disapirro]

tuffy963, on 28 June 2013 - 08:36 AM, said:

I am not sure I understand your underlying point, but let me try to build on the OP's point...

I believe a chimp can host it's lighter body weight into a tree faster than a silver back gorilla. Of course, this example only takes us so far as the chimp and the gorilla have physical sizes that are congruent with their respective weight and power, so it is pretty clear in this example that the chimp is smaller, more agile, and faster. Unfortunately, MWO does not always follow this simple, easy to understand model. In MWO there are mechs that are presented as visually larger, but have a weight that less than a visually smaller mech. These same mechs, while larger, are also more agile than these heavier, smaller mechs. A good example here is the Quickdraw (bigger, lighter, faster) and the Cataphract (heavier, smaller, slower). So, I think the question still stands....

Why does a physically larger framed, lighter, faster weight mech make a poorer hill climber than a smaller framed, heavier, slower mech?

The lack of internal consistency is pretty annoying to me...

Point- weight alone should not be the only consideration to climbing ability. It should also include other tangible/intangible properties.

My example was just to compare things of similar size that have much different climbing abilities.

[tuffy963]

MrMadguy, on 28 June 2013 - 08:24 AM, said:

So. here is my suggestion. Physics model should be even more realistical - climbing speed penalties should depend on your engine power. Something like real phyisics:

<Traction force> = <Engine power> / <Speed> - <Friction force> - <Gravity force>
<Acceleration> = <Traction force> / <Mass>

Where:

<Friction force> = <Friction coefficient> * <Mass> * 9,81
<Gravity force> = <Mass> * 9,81 * Sin(<Climb angle>)

So max speed will be achieved, when <Traction force> = 0, i.e.

<Engine power> / <Speed> = <Friction force> + <Gravity force>

And finnaly:

<Max speed> = <Engine power> / (<Friction force> - <Gravity force>)

What does it mean?
1) More weight you have - slower you moving. That's what we have now.
2) More angle you climbing at - lesser speed you have. That's what we want to achive.
3) But!!! More power your engine have - at bigger angles you may climb with the same speed!
What will it give us?
1) You won't stuck in situation, where you have assault mech and can't fix it's mobility some way, even if you really want.
2) Heavier engines will be more valuable, cuz they'll allow you not to just travel faster, but even to overcome more difficult obstacles.

+1, great suggestions!

A more realistic model based on weight, inertia, traction/friction, and engine power will eliminate frustration with simplistic archetypes (tiny, small, large, huge)

[Disapirro]

tuffy963, on 28 June 2013 - 08:42 AM, said:

+1, great suggestions!

A more realistic model based on weight, inertia, traction/friction, and engine power will eliminate frustration with simplistic archetypes (tiny, small, large, huge)

Disagree. Bigger engine does not mean better climber, though it may help some. More important, in my mind, is does the mech structure, ie. legs, arms, center of gravity, etc. support better climbing. I think that is what they tried to do with these classifications. If I had a non off road type vehicle trying to go off road with a bigger engine, it probably will not help and may actually hurt with too much low end torque and spinning tires. However, if I incorporate nubby tires, better suspension, I can achieve much more.

Edited by Disapirro, 28 June 2013 - 08:50 AM.

[Hatachi]

Here's hoping when falling/collisions come back in the game, you will fall down to sharp hills if you try to climb down.

[tuffy963]

Disapirro, on 28 June 2013 - 08:42 AM, said:

Point- weight alone should not be the only consideration to climbing ability. It should also include other tangible/intangible properties.

My example was just to compare things of similar size that have much different climbing abilities.

Agreed!

Unfortunately, the stated changes appear to be weighted heavily toward only one characteristic.... physical size of the mech. A characteristic that PGI has taken some strange liberties with already (see the endless community feedback threads about strange mech sizing). This hill climbing mechanic could be based on so much more interesting characteristics that bring another level of diversity to the current mech inventory. A few example already stated in this thread include...

• Power to Weight ratios (mech weigh versus engine rating)
  
• Actual weight of the mech (Not all mech play at full chassis weight)
  
• Chassis characteristics (chicken legs, oversized clawed feet, long legs, center of gravity, etc)
  
• Fiction/Traction (climbing a muddy slope versus concrete stairs of the same slope might yield different speeds)
  
• Inertia of mech (included in the upcoming model at some level)
  
• Downslope acceleration

Ok, I am going to put down my drum now...

[MrMadguy]

Disapirro, on 28 June 2013 - 08:50 AM, said:

Disagree. Bigger engine does not mean better climber, though it may help some. More important, in my mind, is does the mech structure, ie. legs, arms, center of gravity, etc. support better climbing. I think that is what they tried to do with these classifications. If I had a non off road type vehicle trying to go off road with a bigger engine, it probably will not help and may actually hurt with too much low end torque and spinning tires. However, if I incorporate nubby tires, better suspension, I can achieve much more.

Yeah, system, devs offered us, is very similar to real physics, except one thing: two angles are hardcapped and it feels unfair. I agree, that ability to climb depends on sturcture at some degree, but it also really should depend on engine power too. Just imagine you have two mechs, that have exact the same model, but one is 3x bigger and heavier, then other. Why first will be permanently better climber even with the worst 100 rating engine, then second with best possible 400 rating engine? How about scaling?

<Power> / <Weight> ratio should be = Const

taking into account, that

<Structure configuration> = Const too

What is climbing? It's overcoming of gavity's sine component via traction force. And as traction force depends on power, efficiency of climbing should really depend on engine's power somehow too.

Edited by MrMadguy, 28 June 2013 - 09:53 AM.

[kuangmk11]

Catapult and Quickdraw should be moved to medium or the Cataphract and Jager upped to large. Otherwise I like it

[Fate 6]

"Tiny: Jenner, Commando, Spider
SlowdownAngle: 40°
Small: Raven, Cicada
SlowdownAngle: 35°
Medium: Blackjack, Centurion, Dragon, JagerMech, Trebuchet, Cataphract, Hunchback
SlowdownAngle: 30°
Large: Quickdraw, Stalker, Awesome, Catapult
SlowdownAngle: 25°
Huge: Atlas, Highlander, Victor
SlowdownAngle: 20°"

Are you kidding me???
Just do it by weight. Putting a Cataphract or Jagermech on the same tier as a Blackjack? This is like a sick joke. Or putting a Quickdraw or Catapult above the Cataphract and on the same level as a Stalker. And calling a Stalker as small as or smaller than the 2 80 ton assaults?

This is probably the most idiotic thing I've seen since PGI removed GD. It's even worse than nerfing the HBK-4P.

[Disapirro]

MrMadguy, on 28 June 2013 - 09:40 AM, said:

Yeah, system, devs offered us, is very similar to real physics, except one thing: two angles are hardcapped and it feels unfair. I agree, that ability to climb depends on sturcture at some degree, but it also really should depend on engine power too. Just imagine you have two mechs, that have exact the same model, but one is 3x bigger and heavier, then other. Why first will be permanently better climber even with the worst 100 rating engine, then second with best possible 400 rating engine? How about scaling?

<Power> / <Weight> ratio should be = Const

taking into account, that

<Structure configuration> = Const too

What is climbing? It's overcoming of gavity's sine component via traction force. And as traction force depends on power, efficiency of climbing should really depend on engine's power somehow too.

I think we basically agree, just a little difference on what 'climbing' is. Climbing a steep grade on a smooth service is all about engine size and weight of vehicle, and of course, having appropriate traction. I guess I envision this to be more off road, with slippery and sometimes jagged or rough surfaces that require different feet, hands, how legs bend, etc.

Edited by Disapirro, 28 June 2013 - 10:13 AM.

[Hungus]

Giant mechs with legs that can't even walk up 45° angles or step over small obstacles.

This is terrible design.

How about we add more invisible walls into the maps while we are at it.

[Blalok]

627, on 27 June 2013 - 12:35 PM, said:

uhh. lostech.

by the way, you know a wiimote plus? there's your gyro, we use that to stabilize rc-quadrocopters

^^ this. Gyros in R/C nowadays are around the size of a quarter (and as far as I can tell, have no moving parts). Link:http://www.rchelicop...opter-gyro.html

[MrMadguy]

And what about inertia? Yea, bigger mechs are dealing with bigger gravity forces, but they also have bigger inertia, i.e. accumulated energy. In terms of law of energy conservation:

<Mass> * 9,81 * <Height of obstacle> = (<Mass> * <Speed> ^2) / 2

and then we just remove <Mass> from equation

<Height of obstacle> = (<Speed> ^ 2) / (2 * 9,81)

And so effectiveness of climbing via inertia depends only on speed, i.e. engine power / mass ratio. The only things, that are affected by climb angle - friction and durability vs g-load. I.e. vertical wall will turn all your energy into internal forces - you'll be smashed via hitting it, if you're not absolutely rigid body of course.

Edited by MrMadguy, 28 June 2013 - 10:50 AM.

[Kyrs]

sound pretty fun to me,

Except that I would expect the chicken waking mech to have trouble walking on hill. It currently seem right for commando and spider vs raven .. why? raven fly up hill they dont walk like a commando.

But the ckicken walking cataphract and stalker, I'm not to shure that they should be better climer the awesome. I guess we will have to wait see. Can't wait test it out... to see if that xl400 can push that 80 tons further uphill with it speed momemtum.

[MrMadguy]

It's just interesting for me. Could anybody tell me, how max speed is calculated in this game?

<Max speed> = 3.6 * <Enginge ratio> / (<Mass> * <Friction force>)

where

<Friction force> = K * 9.81 * <Mass>

where K - is friction coefficient, which is different for different mechs and 3,6 - is mps to kph ratio?
For example Jagermech:

K = 0.0003485
<Mass> = 65
<Engine ratio> => <Max speed>
240 => 59.8
260 => 64.8
280 => 69.8
etc

If so, where can I get this friction coefficients for every mech?

Edited by MrMadguy, 28 June 2013 - 11:27 AM.

[Warge]

I think StopAngle should be raised from 45 to 60.

[Koniving]

Warge, on 28 June 2013 - 11:33 AM, said:

I think StopAngle should be raised from 45 to 60.

At that angle, most mechs with humanoid legs put their knees through their arms with certain torso rotations when MWO's 'inverse kin'whatevers (leg lifts with placement of ground) comes into play with the animations. I recall it when they first had enabled that and I also recall seeing weird things especially with Atlases and Stalkers and Alpine.

Though as a "dead stop" angle rather than a 'slow down to stop' angle as 60 is a good idea.

[Kattspya]

The slowdown when touching walls is gong to be so broken and frustrating it's going to be hilarious. The textures and collision meshes are not matching each other very well at the moment and it is already frustrating when you get snagged on invisible corners under the current system. Imagine how it will be when you only have to scrape invisible corners to lose forward speed.

Then there is the second part of the problem. Currently there is almost no sense of when your arms or body are clearing an obstacle. Now it only leads to a few wasted shots but imagine when it kills your speed in a medium or light. EDIT: 3PV just got really really useful. Small advantage mein *****...

I'm betting the wonderful testers at PGI are testing under optimum conditions so imagine what packet loss, unstable ping and high ping combined with the greatest net code of all time will do to exacerbate the problem above.

I think both mechanics sound great on paper but my faith in PGI to implement it without burning down the forums and without having to hotpatch it out is about zero. It is going to be like collisions only worse as architecture generally does not try to get out of the way.

Edited by Kattspya, 28 June 2013 - 12:18 PM.