18 replies to this topic

[Vexgrave Lars]

We are getting excellent artwork from the Developers and here in the forums.

As I work on 3d drawings 'm finding that the Art does not support the mechanics often in my renders. For example.

I did a drawing of the Marauder.. which was cool, but when i tried to redraw the leg in the original fashion, or "canon" design. I found the primary knee joint and the thigh don't leave enough clearance to flex the leg to 90 degrees forward without some healthy modification, nor a 15deg. clearance to either side.

Yes.. i know .. its Robotic Myth, and a game. but still, as I'm learning the armatures and animation this is getting frustrating. I know I can "cheat" but well.. that's cheating...

So to the rest of you, do you get frustrated when you find problems like this and want to correct them? How do you stay close to canon when the mechanics support a "pose" but not an expected function of a limb or device on your Mech?

Making the world happy.. just not possible. But I'd settle for some advice.

[HanaYuriko]

Straighten the leg some to make the model stand taller and give it more mobility for any poses or walking animations. The leg doesn't need to bend a full ninety-degrees for it to work. Otherwise your other viable option is to modify the leg and knee joint so it does work.

[Phelan Adam Davion]

today I read my Battletech Source Book again.... 512 sites full of everything you can just know about the 31st century..... with plenty pictures in it..... and while watching all those original drawings I just wanted to tear my hair out!

On every picture there was something illogical or irrational.
The old Battletech art is just supposed to look "cool" in the 80`s style.....

...but it`s not even technical or logical.
Thats why I really love the new Mechs here presented for MW:O cause I have a look on every single detail and it just makes sence.
It`s great to see those 80`s toys gettin "alive".

Make the Marauder move!

[Garth Erlam]

I'll ask Alex to drop by.

[Vexgrave Lars]

Thanks guys, glad its not just me

[Paul Inouye]

You also have to realize that the creators of the original art pieces were not mechanical engineers or industrial designers.

[CoffiNail]

I love the fact it looks like FD takes his time working on the MWO TRO art Freaking half the 80s TRO line art looks like it was rushed and done in a few minutes.

Plus what Paul said. Look at the Centurions feet in the TRO vs FD's redesign. The flat, solid looking feet of the TRO makes it look like it would have trouble walking on un even ground. The realistic look of the FD design makes it look like the feet will adjust as the mechs feet hit the terrain and gives it a much more stable look.

[Vexgrave Lars]

Paul and Garth, thank you for the time and responses.

Your artist have gone way above and beyond. I concede its easy to be critical of the past and that we shouldn't diminish the work of our "Mechfathers". I guess I haven't found my envelope edge of "canon" and "can function" quite yet.

In anycase, thank you for your efforts as they are truly helping to carry the genre forward, my 6 year old son started playing Mechwarrior4 with me the other day, after I showed him the media page.. he's stoked.

We are all looking forward to the day we can ride our mechs into battle in MW:O.

Humbly,

[Strum Wealh]

Paul Inouye, on 27 February 2012 - 01:56 PM, said:

You also have to realize that the creators of the original art pieces were not mechanical engineers or industrial designers.

Though, it seems that Shōji Kawamori, who did the designs for the Macross series (among others), does actually have some engineering background:

Quote

After contacting the design company, Kawamori was offered a designing job on the second Yamato movie. He was only a freshman at Keio University at the time, studying engineering (his goal was to get into aeronautical engineering). By the time Kawamori was a sophomore in college, he had already worked on several anime projects before Studio Nue finally approved his Macross idea. Kawamori worked on Macross all throughout college before he eventually dropped out to work on anime full-time.

(source)

Quote

Watch for the split second cameo of the real-life XB-70 Valkyrie bomber — the very plane whose picture inspired Kawamori as a child to pursue aerospace engineering all the way to Japan's top private university.

(source)

[flyingdebris]

well, I'm not the one that does the 3d around here, but i find that accounting for clipping issues and other assorted problems beforehand in the 2d before it reaches 3d often involves lots of iteration. I often go back into a design I'm working on and change stuff all the time to make it's joints make a bit more sense.

In so far as sticking to the canon art, throwing out the canon joint appearance and replacing it with something similar but more believable and appropriate is pretty much the first thing i do. Some people may care just how closely the hips or knees match the art, but a lot more will care if it's clipping like crazy or results in a really awkward walk, so yeah don't be afraid to overhaul those old joints.

in creating mech joints in general, I don't usually know how animators will actually make it move, but i do try to incorporate into the designs enough range of motion for a particular joint where they would have the necessary leeway to do what they want.

Additionally, if you go through the trouble of making orthographic turn arounds of a design, it's a lot easier to see the problem spots that you might not notice in the 3/4 view.

and lastly, if it still ends up not making sense in 3d, then yeah, just fix it. no shame in it. The art itself isn't sacred. If it doesn't work it doesn't work. Just fudge the details til it does.

[Ironhawk]

Garth Erlam, on 27 February 2012 - 01:27 PM, said:

I'll ask Alex to drop by.

Paul Inouye, on 27 February 2012 - 01:56 PM, said:

You also have to realize that the creators of the original art pieces were not mechanical engineers or industrial designers.

flyingdebris, on 27 February 2012 - 05:34 PM, said:

well, I'm not the one that does the 3d around here, but i find that accounting for clipping issues and other assorted problems beforehand in the 2d before it reaches 3d often involves lots of iteration. I often go back into a design I'm working on and change stuff all the time to make it's joints make a bit more sense.

In so far as sticking to the canon art, throwing out the canon joint appearance and replacing it with something similar but more believable and appropriate is pretty much the first thing i do. Some people may care just how closely the hips or knees match the art, but a lot more will care if it's clipping like crazy or results in a really awkward walk, so yeah don't be afraid to overhaul those old joints.

in creating mech joints in general, I don't usually know how animators will actually make it move, but i do try to incorporate into the designs enough range of motion for a particular joint where they would have the necessary leeway to do what they want.

Additionally, if you go through the trouble of making orthographic turn arounds of a design, it's a lot easier to see the problem spots that you might not notice in the 3/4 view.

and lastly, if it still ends up not making sense in 3d, then yeah, just fix it. no shame in it. The art itself isn't sacred. If it doesn't work it doesn't work. Just fudge the details til it does.

FD's comment is insightful to say the least. Made me recall the process of working 4 times on those Urbie legs. He's absolutely right. By the way, PGI = coolest dev team of them all.

[Mordhar]

I do not have much to add, previous posts cover almost everything.

Only some tips. It is how I make models, and I do not pretend it is the best way to work.

Try to make very simple model first, just a couple of boxes linked together, to make sure it have enough joints to support any movement you need. If it does not (and usually this is the case) look where you can add them.

And while working on your model do not forget to rotate some parts once in a while to make sure they have enough space to move without clipping.

Making a model that not only looks “cool”, but also can really work is a very hard task. You will probably spend as much time on making it functional as on modeling itself.

[GraySho]

I ran into the exact same problems when modeling my ShadowHawk, Stinger, PhoenixHawk, Thug, etc. TRO art is not logical nor functional.

I once read an insight article about the MW4 intro movie, where they mentioned the remodeling of the Vulture/Maddog just to have more liberty when animating in Messiah, and making hip movement possible. The results where very believable and convincing, so I´m all for the redesign that´s going on here on the MWO art team to make Battletech finally a tiny bit more "realistic" and believable.

The tip with low poly models is really great (if you know a bit about 3D modeling) to figure out how joints and geometry can work out.

Edited by GraySho, 28 February 2012 - 12:47 AM.

[BLaaR]

Same goes for trying to bring the models to paper. I found out that many a time, I need to alter some one or other part to make it work. When you built the model, you quickly found out the weakness of the mech. Most of the time, and this is more evident in the chicken walkers, they tend to be less stable than the humanoid counterparts. I usually end up adding weight to the feet to keep the chicken walkers upright, where as the humanoid's tend to stand easily on its own.

I think animating the humanoids should be easier than the chicken walkers, but then again I am no 3D artist.

Another weakness I found was usually the "hip joints". Way to small at times, and I had to make them bigger.

I love the new art of MWO, functional and realistic.

[Xathanael]

Who is FD?

[Eximar]

Alex Iglesias is "flyingdebris", or FD, the person that gives us all these gorgeous 'Mechs.

http://mwomercs.com/...-alex-iglesias/

[Bishop Steiner]

Firstly, love the nod to heavy metal under your siggy.

Secondly.... YES!!!!!!!!!

Tis a reason i am constantly redesigning mechs, trying to achieve usable joints and believable proportions. I am far from perfect at it, and wish i knew how to render, or had a partner in crime who did, as it would help me improve my overall images.

And yes the canon unseen marauder may look neat, but to make it actual move is near to an engineering impossibility.

[Steel Raven]

Looking at Models and existing game animation helps, many time artist change the look to something that makes more sense to themselves which you can chalk up to artistic license.

[ThePhil]

OP. We artists stand upon the shoulders of giants so we may reach the things they did not. Refining a design to increase functionality of the item were it to exist in real life is something that takes considerable time. This is why we continually see improvements on old technology. It takes time. I hope that the designs and illustrations (though I barely feel comfortable calling them illustrations at this point) will serve as a launching point for another handful of artists in the future.