Edited by Solahma, 01 July 2015 - 09:55 AM.


#61
Posted 01 July 2015 - 09:54 AM
#62
Posted 01 July 2015 - 10:05 AM
Urbanmech. The leg's just need a rescale. They are too large for the torso(in fact,in game, they appear to be the same size as the torso.)
Centurion. I want to love it. But that barn sized chest just lets it go zombie all day when you are stripped of everything.
Awesome. Yeah.
Catapult. The old CRT launch tubes that don't rescale due to launchers sucks. The launchers don't need to be outside the bays.
Orion. I feel bad when I leg them.
On the clan side
Adder/Kit Fox both are slightly too wide for their size/speed.
War Hawk. I get the feeling it's just too tall/big. I have always gotten the impression it's supposed to be same size/smaller than the DW.
Mad Dog(to a lesser extent,the Timbie suffers this but less so due to superior boxes) it's way way too tall. They are known for their squat walk.
But yeah,those are the ones I noticed.
#63
Posted 01 July 2015 - 10:14 AM

#64
Posted 01 July 2015 - 11:29 AM
WM Quicksilver, on 01 July 2015 - 09:46 AM, said:
That said, there is a problem in trying to test volume cubic parameters as the models are now, I "exploded" the Nova's torso in Wings to show you something that is problematic, both because you would need to close ALL pieces like this to get a good estimate, and because you would have overlapping volumes, leading to higher numbers than what it should be.

Theres no easy way or a program to merge the pieces?
#65
Posted 01 July 2015 - 11:30 AM
Quickdraw, both height and width.
Nova, shrink height, width may be good.
Awesome, shrink width
Trebuchet, the beefiest and tallest of the IS 50 tonners.
Can't decide between Victor and Catapult.
#67
Posted 01 July 2015 - 11:37 AM
Queek Head Taker, on 30 June 2015 - 03:08 PM, said:
"b/c fixing broken things is bad for games"
#68
Posted 01 July 2015 - 11:37 AM
Tennex, on 01 July 2015 - 11:29 AM, said:
Theres no easy way or a program to merge the pieces?
More technical design programs have options for merging, mostly boolean operations. Blender I'm not familiar with enough to know what options it may have, but my initial impressions is that it's more difficult than something like AutoCAD, Solid Works, Catia, Rhino, etc.
Edit: Quicksilver ninja

Edited by Solahma, 01 July 2015 - 11:37 AM.
#69
Posted 01 July 2015 - 12:14 PM
#70
Posted 01 July 2015 - 12:55 PM
Tennex, on 01 July 2015 - 04:46 AM, said:

But there are instances where the T pose will be different from the ingame pose. Like the adder and Kit fox, which they actually raised for some reason. So keep that in mind too when using these as scaling references.


I think the mechbay uses different scaling. In the lineup the Adder and Jenner are similar, but in the mechbay the Adder is significantly larger.
#71
Posted 01 July 2015 - 12:59 PM
Solahma, on 01 July 2015 - 05:32 AM, said:
For example: the DWF and Stalker are smaller frontal profile, but have A LOT of depth (side profile) compared to an Atlas or a Banshee.
Would also be interesting to partition each silhouette into individual hitboxes and do a surface area metric comparison from the front, side, and rear (less important) of each mech (perhaps even a 45 deg from the front as well). I've always wanted to do this, but haven't found the time.
If someone did this, I could do a quick import into my CAD software and very fast and extremely accurate surface area dimension.
I feel this value has more value than a flat surface area from a 3D model which has already been done. A total Surface area and/or volume doesn't tell you what percentage of the total VISIBLE surface area will be for say... a Side Torso. If the distribution, from the front, was 15% arms, 30% legs, 40% CT, and 15% ST, THAT would be good and useful information for balancing hitboxes or at least a value for comparison.
Scale is one thing, but our eyes can easily deceive us, it would be nice to have surface area values to justify and/or dismiss claims.
EDIT: if no one does this, i'll make it a priority project for myself.
Side profile only matters if you are flanking them, eg a light. If you are heavy or assault you can only shoot the front which is shooting back at you
#72
Posted 01 July 2015 - 01:24 PM
UrsusMorologus, on 01 July 2015 - 12:55 PM, said:
It is probably simply due to how they rest chicken walkers. If you notice the torso's aren't bigger, they are simply higher placed. You can check this by looking at where the crotch is in the mechbay versus in my screenshots.
In this particular instance you can see the Adder's crotch is raised in the mechbay because it lines up with the Jenner's head in the mechbay vs near the Jenner's crotch in my comparisons. Yet the size doesn't seem to be significantly different, just the height. Thus it is simply a raised stance from having a different resting state than when it was modeled.
Edited by WM Quicksilver, 01 July 2015 - 01:25 PM.
#73
Posted 01 July 2015 - 01:41 PM
#74
Posted 01 July 2015 - 01:42 PM
UrsusMorologus, on 01 July 2015 - 12:59 PM, said:
This is not true at all. Sure, from a distance in a 1v1 situation you are partially correct. However, Mechs torso twist to spread damage, you can be caught in a cross fire. There can be a mech that is not currently engaged shooting you from the side of the enemy firing line. More often than not, you are not facing your target dead-on, full frontal profile. The side profile of a mech is as much if not MORE important than the frontal profile.
Example: Frozen City
There are many opportunities to get side-shots on mechs while still being mostly infront of the enemy. You don't have to be on some wild and wide flank to get a side shot on a mech. Like I said before, generally your opponent it going to be torso twisting to spread damage even if you are standing right infront of him. Side profiles are VERY important.
Situation 1

Situation 2

Edited by Solahma, 01 July 2015 - 01:53 PM.
#75
Posted 01 July 2015 - 01:54 PM
As an VFX artist I think I'm going to have a lot of fun with those

#76
Posted 01 July 2015 - 02:39 PM
Big Black Wolf, on 01 July 2015 - 01:54 PM, said:
As an VFX artist I think I'm going to have a lot of fun with those

Heffay's Tutorial is probably the best way to get the obj files. Unfortunately there is a process to extracting them into a more useful format and because of how many mechs there are and due to all of the dynamic geometry, all the obj files amount to around 1.5gb worth of files which means uploading would be a pain and that isn't even including the texture files.
#77
Posted 01 July 2015 - 03:12 PM
WM Quicksilver, on 01 July 2015 - 09:46 AM, said:
That said, there is a problem in trying to test volume cubic parameters as the models are now, I "exploded" the Nova's torso in Wings to show you something that is problematic, both because you would need to close ALL pieces like this to get a good estimate, and because you would have overlapping volumes, leading to higher numbers than what it should be.

Measure elements and add together? But yeah you would probably need to close all pieces. Someone who did figures from MWO on 3D printer would have this done already. I know many did this.
But even measure 2D surfaces and [front x side] I bet would tell us more about it.
And I don`t agree that side profiles do not matter. They do.
Some meks are just not noob friendly I would say, but one can take advantage for they low profile.
You can matrix dodge in those.
#78
Posted 02 July 2015 - 05:23 AM
#80
Posted 02 July 2015 - 07:36 AM
1 user(s) are reading this topic
0 members, 1 guests, 0 anonymous users