183 replies to this topic

[Ghogiel]

Tennex, on 19 June 2013 - 09:47 PM, said:

lets see those numbers please to put this to rest.

Here are 2 mechs, that aren't the same weight class, but are whacky in size, that were assembled from the in game geometry (not the fat torso catapult model that is never used in game you see often).




These numbers should be fairly accurately measured. It's not completely 100% exact, as making the enclosed model in the space of 5mins instead of possible hours brings in a small inaccuracy of the surface, this is the level of inaccuracy >

[Tennex]

Ghogiel, on 20 June 2013 - 03:24 PM, said:

Here are 2 mechs, that aren't the same weight class, but are whacky in size, that were assembled from the in game geometry (not the fat torso catapult model that is never used in game you see often).




These numbers should be fairly accurately measured. It's not completely 100% exact, as making the enclosed model in the space of 5mins instead of possible hours brings in a small inaccuracy of the surface, this is the level of inaccuracy >

assuming these numbers are right. they are way closer in weight than they should be. according to these values (using the catapult as reference) the stalker is the size of a 70 ton mech.


but i think a previous problem with using this method, was the various random holes in the models which screwed up the volume readings.

was there anything done to address that issue in your method? I assume a enclosed model means there are no holes in it?

Edited by Tennex, 20 June 2013 - 03:44 PM.

[Tie Ma]

Ghogiel, on 20 June 2013 - 03:24 PM, said:

Here are 2 mechs, that aren't the same weight class, but are whacky in size, that were assembled from the in game geometry (not the fat torso catapult model that is never used in game you see often).

so basically the stalker is the size of a 70 ton mech

[Tennex]

Tie Ma, on 20 June 2013 - 03:45 PM, said:

so basically the stalker is the size of a 70 ton mech

yup at that size the stalker should only be 5 tons heavier than the catapult... let that sink in

should only be able to carry 5 tons more armaments than the catapult.. but instead it can carry 15 more tons. and has the armor cap of a 80 tonner but the hitbox of a 70 tonner.

Edited by Tennex, 20 June 2013 - 09:50 PM.

[Ghogiel]

Tennex, on 20 June 2013 - 03:36 PM, said:

assuming these numbers are right. they are way closer in weight than they should be.


but i think a previous problem with using this method, was the various random holes in the models which screwed up the volume readings.

was there anything done to address that issue in your method?

I explained that already somewhat. The mech geometry as used in game is full of open edges and composed of many objects with hidden and intersecting surfaces. I processed the model so there are no holes, it is a single continuous surface with no hidden inner surfaces, so can be read correctly by max, maya etc measure tools.

Tie Ma, on 20 June 2013 - 03:45 PM, said:

so basically the stalker is the size of a 70 ton mech

Or the catapult is the size of an 80 ton mech.

[Tennex]

Ghogiel, on 20 June 2013 - 03:57 PM, said:

I explained that already somewhat. The mech geometry as used in game is full of open edges and composed of many objects with hidden and intersecting surfaces. I processed the model so there are no holes, it is a single continuous surface with no hidden inner surfaces, so can be read correctly by max, maya etc measure tools.

awesome.. finally a reliable volume measurment

that definitely puts the volume argument to rest

Edited by Tennex, 20 June 2013 - 03:59 PM.

[DaZur]

Ghogiel, on 20 June 2013 - 03:24 PM, said:

Here are 2 mechs, that aren't the same weight class, but are whacky in size, that were assembled from the in game geometry (not the fat torso catapult model that is never used in game you see often).


These numbers should be fairly accurately measured. It's not completely 100% exact, as making the enclosed model in the space of 5mins instead of possible hours brings in a small inaccuracy of the surface, this is the level of inaccuracy >

In fairness the surface area and volume to be reasonably accurate we'd have to average all the objects to remove the physical nuance details (those bevels and extrusions are killer) as well as Boolean merge and subtract... In short too many bits and widgets to give a reasonably accurate assessment.
Having said that... even discounting the Cats ears (which contribute both to it's surface area and dimensions bloat)... physically the dimensions of the Stalker are too close to the Cat for anyone (even me) to argue that the scale is not off a tad.

Tennex, on 20 June 2013 - 03:58 PM, said:

awesome.. finally a reliable volume measurment

that definitely puts the volume argument to rest

Well... not entirely. But is does give it a stern disproving look...

Edited by DaZur, 20 June 2013 - 05:23 PM.

[Ghogiel]

DaZur, on 20 June 2013 - 05:22 PM, said:

In fairness the surface area and volume to be reasonably accurate we'd have to average all the objects to remove the physical nuance details (those bevels and extrusions are killer) as well as Boolean merge and subtract...

Not sure how removing parts of the mech would be more accurate. :\

[DaZur]

Ghogiel, on 20 June 2013 - 05:35 PM, said:

Not sure how removing parts of the mech would be more accurate. :\

Well... All those nifty panels, ladder rungs, bezels, bits and widgets are totaled into the surface area and volume as do the intersect of objects and or object back facing (Parts of the model that essentially are composites of several smaller objects). Not even mentioning the mess created by capping open-face objects...

All that detail comes at a cost...

[AntiCitizenJuan]

Is it possible to change the size of a mech with the games engine? Just curious.

[DaZur]

AntiCitizenJuan, on 20 June 2013 - 07:44 PM, said:

Is it possible to change the size of a mech with the games engine? Just curious.

Not Normally... (Qualifying the fact I know little about the full capacity of the Cryengine)

[Scrawny Cowboy]

AntiCitizenJuan, on 20 June 2013 - 07:44 PM, said:

Is it possible to change the size of a mech with the games engine? Just curious.

Considering how long it took em to fix the Jenner hitboxes... *shudder* If they do acknowledge that there is need for adjustment, big IF, it'll be well past launch before we see any outcome of that. : /

[Ghogiel]

DaZur, on 20 June 2013 - 07:43 PM, said:

Well... All those nifty panels, ladder rungs, bezels, bits and widgets are totaled into the surface area and volume as do the intersect of objects and or object back facing (Parts of the model that essentially are composites of several smaller objects). Not even mentioning the mess created by capping open-face objects...

All that detail comes at a cost...

Like I said that there are no intersecting geometry or back facing polygons. It is a single entirely enclosed mesh.

The ladders and dilly bobs do add up yeah, but they are part of the mech.

[DaZur]

Ghogiel, on 20 June 2013 - 07:56 PM, said:

Like I said that there are no intersecting geometry or back facing polygons. It is a single entirely enclosed mesh.

The ladders and dilly bobs do add up yeah, but they are part of the mech.

LOL! If you truly merged and subtracted, capped all open ends and welded all those objects, I tip my hat to ya.. That was no small task to help add veracity to this little discussion!

While you have those composites in hand... Think you can scale the Cat down 10% and do a side-by-side?

I've followed this whole hullabaloo regarding mech scale for a long time now and aside from the light mechs, I'm not convinced the scale between most of the mediums and the heavy and assault really amounts to a hill of beans in regards to game balance / survivability...

As I see it, an up or down of 10% in these classes does not create enough of a difference in surface area to support the premise that doing so would make certain mechs more survivable.

Visual continuity on the other hand...

Edited by DaZur, 20 June 2013 - 08:19 PM.

[AntiCitizenJuan]

DaZur, on 20 June 2013 - 08:18 PM, said:

LOL! If you truly merged and subtracted, capped all open ends and welded all those objects, I tip my hat to ya.. That was no small task to help add veracity to this little discussion!

While you have those composites in hand... Think you can scale the Cat up 10% and do a side-by-side?

I've followed this whole hullabaloo regarding mech scale for a long time now and I'm convinced aside from the light mechs, I'm not convinced the scale between most of the mediums and the heavy and assault really amounts to a hill of beans in regards to game balance / survivability...

As I see it, an up or down of 10% in these classes does not create enough of a difference in surface area to support the premise that doing so would make certain mechs more survivable.

Visual continuity on the other hand...

If the Stalker was as big as it *should* be, it would get hit more, and it wouldnt be able to aim down as easily, making it easier prey from small mechs. In theory.

[mindwarp]

Morang, on 19 June 2013 - 11:44 AM, said:

We have, indeed.


This one is from TRO3039 by Catalyst (product 35121).

If we consider human 1.8 meters tall, it gives us 9.0 - 11.9 - 13.7 - 14.9 height for the mechs.

If we consider Banshee 14.9 meters tall in the lower picture, then Leopard is 62 m long (nozzles to nose cone) and 19.5 m high (hull only, wings and wheels and bridge protrusion excluded), Union is 73 by 74 meters (less gun blisters and landing gear), Overlord is 125 tall by 92 wide (as Union).

You're about 5-10% out with your figures there. FASA actually gave us official sizes for the various dropships in Technical Readout 3057. The Union is 81.5m long, 81.5m wide and 78m high. The Overlord is 99m ling, 99m wide and 131.2m high. The Leopard is 65.6m long, 51.6m wide and 22.4m high. Which means that human is more or less 2m tall.

*edit* The Banshee is also the tallest IS mech up till about 3058. Noticeably taller than the Atlas, and head and shoulders over the Awesome. The old Ral partha mini's were closer to the intended scales than the new Ironwind ones, and even then, the Quickdraw was immensely tall - taller than an Awesome (just) with the top of it's head just below the top of the Atlases shoulder. The Trebuchet was even taller than the Quickdraw,

Edited by mindwarp, 20 June 2013 - 08:56 PM.

[Ghogiel]

DaZur, on 20 June 2013 - 08:18 PM, said:

LOL! If you truly merged and subtracted, capped all open ends and welded all those objects, I tip my hat to ya.. That was no small task to help add veracity to this little discussion!

While you have those composites in hand... Think you can scale the Cat down 10% and do a side-by-side?

I've followed this whole hullabaloo regarding mech scale for a long time now and aside from the light mechs, I'm not convinced the scale between most of the mediums and the heavy and assault really amounts to a hill of beans in regards to game balance / survivability...

As I see it, an up or down of 10% in these classes does not create enough of a difference in surface area to support the premise that doing so would make certain mechs more survivable.

Visual continuity on the other hand...

I didn't do any manually poly modelling... for what would probably be days worth, I ain't that stupid, bored or crazy. I just dynameshed it in zbrush and decimated that. Which is why there is some minor inaccuracy, looking at it though it's pretty ******* close though. No way anyone is going to faf about to do it exactly via regular poly modelling, so it's the best we are likely going to get.

Anyway scaling the cat down 10% for the volume readout.... makes it 245.73*

It is fairly noticeable visually imo too>


DAT TOPOLOGY

[DaZur]

Ghogiel, on 20 June 2013 - 08:50 PM, said:

I didn't do any manually poly modelling... for what would probably be days worth, I ain't that stupid, bored or crazy. I just dynameshed it in zbrush and decimated that. Which is why there is some minor inaccuracy, looking at it though it's pretty ******* close though. No way anyone is going to faf about to do it exactly via regular poly modelling, so it's the best we are likely going to get.

Anyway scaling the cat down 10% for the volume readout.... makes it 245.73*

It is fairly noticeable visually imo too>


DAT TOPOLOGY

Geez... I'm a dork. I never thought to Dynamesh them... good on ya!

Actually... and I hope I'm not imposing and if I am...tell me (No one else around here has a problem doing so) can you side-by-side that reduced Cat with the stalker?

This would ultimately help my detractors or prove my hypothesis...

BTW... I know there's a process of extracting these models but it there anywhere to simply grab them without all the fuss?

Edited by DaZur, 20 June 2013 - 09:12 PM.

[Ghogiel]

[DaZur]

Ghogiel, on 20 June 2013 - 09:12 PM, said:

Sigh... My detractors are going to have a field day with this.

I capitulate... resign and otherwise rescind my forwarded postulations and otherwise admit to "possibly" be wrong.