4 replies to this topic

[Nightbird]

Referring to the image below, which is used to justify the rescale was done correctly, it is clear that the profile was used to scale instead of volume. To explain this simply, when you double the profile of a sphere (x2), you increase the volume by x2.83. When you triple (x3) the profile of a sphere, you increase the volume by 5.20x. When you quadruple (x4) the profile of a sphere, you increase the volume by 8.00x. When you quintuple the profile of a sphere, you increase the volume by 11.18x. While the volume of a mech is not as easy to calculate to that of a sphere, the real difference is small due to a mech being a collection of spheres, cylinders, and rectangular prisms.

Why does volume matter? Because volume is how we intuitively understand mass. Something with double the volume should have double the tonnage. However with the current scaling, a 100 ton mech has 8x the volume of a 25 ton mech, intuitively it feels like a 200 ton mech. Likewise, a 90 ton mech compared with a 30 ton mech feels like a 150 ton mech.

This is a hard question given how much effort went into the current rescale, what do you guys think, do the mechs feel right now or still wrong?

Edited by ironnightbird, 09 July 2016 - 06:22 AM.

[Wintersdark]

Navid did an interesting analysis via pixel counting with front/top/side and while we're all aware of how that's a flawed method of determining volume, it's about the most practical method we have at this time.

The results largely supported volumetric rescale.

What is very clear however is that the did NOT scale by front profile alone. Not even close.

[Nightbird]

I can see that Navid did a good job in the volume scaling, and his graph shows that a 100 ton mech has something not far from (> in fact) 11x the volume of a 20 ton mech. The image below is from his post:



Surface pixel counting is fine, however you need to take the cubed root (exponential 3/2) of whatever pixels you counted to get to the volume. Well, not exactly the volume since it is unit-less but it would be proportionally correct at least. PGI clearly took the front, size, top profiles and without this volume adjustment scaled based upon that.

You may consider this to be necessary game design, surface area is what you shoot at so ignore volume right? In my opinion, the representation of the weight of the mechs is far more important. What is the first thing you ask about a new mech? What tonnage it is. This game is all about the mechs, if it's OK to make a 100 ton mech the volume of a 200 ton or heck based on Navi's chart a 300 ton mech, what is sacred?

Edited by ironnightbird, 09 July 2016 - 06:24 AM.

[Nightbird]

If anyone doesn't understand the math, just do it intuitively. Humans are actually quite good at spacial perception and imagination.

The issue becomes more egregious and easier to see the bigger the tonnage separation is. In your mind, cut apart a 30 ton mech and fit it into a 90 ton mech, can you fill that 90t mech with 3 30t mechs or do you need 5 30 ton mechs? It gets worse when you compare a 25 ton mech with a 100 ton mech, or a 20 ton mech with an 80 ton mech, you should be able fill the larger mech with 4 smaller mechs so why does it take more around 8?

My suggestion is: make the mech look like how much they weigh. Mech size is too important to use as a balance point. If this has the effect of making mechs like the marauder (hate), crab/king crab, dire wolf, warhawk (lol), timber, ebon, jenner, raven, dragon, etc.. better because their design prioritizes minimizing front profile, then so be it, let the meta change. At the very least, mech design will actually matter instead of having well designed mechs being inflated for "target size" balance.

Edited by ironnightbird, 09 July 2016 - 06:20 AM.

[EvilCow]

I think volume is fine and the game overall improved after the change.