Fuel for Suc­cess

Animation Magazine - - Opportunities -

Find­ing the right job — or the right can­di­date for a job — gets eas­ier for the an­i­ma­tion and vis­ual ef­fects in­dus­tries with An­i­ma­tion Mag­a­zine’s new Ca­reer Cen­ter, launch­ing Feb. 13.

It’s no easy feat to suc­cess­fully blend live-ac­tion hu­mans and ob­jects with CGI crea­tures. If the artists veer too far into the car­toon­ish, it can be a jar­ring ride for the au­di­ence. But if the vis­ual-ef­fects crew man­ages to pull to­gether the un­real with the real in a way that sits right with the eyes of the au­di­ence, the whole story takes on a new life.

Mon­ster Trucks, di­rected by Chris Wedge, the co-founder of Blue Sky Studios, set out to do just that. Live trucks and a live-ac­tion hero are brought to­gether with “mon­sters” like Creech, and taken on a wild ride with Lu­cas Till at the wheel. Vis­ual ef­fects su­per­vi­sor Ni­co­las Aithadi was in charge of bring­ing a be­liev­able look to these un­real be­ings.

“The na­ture of the crea­ture (Creech) is squishy and wob­bly, and in this si­t­u­a­tion it was al­ready a com­plex si­t­u­a­tion to rig and to an­i­mate, but then it was made ex­po­nen­tially harder when we had to in­te­grate Creech with the car,” says Aithadi. “So what we did was cre­ate a Creech in the out­side world — what he looks like just on the ground — and a Creech that looks dif­fer­ent when he’s in­side the car.”

When it came time to de­sign the char­ac­ter, Aithadi and his crew looked to oceanic crea- tures that had the char­ac­ter­is­tics they wanted to give to Creech and his fam­ily. Given the sorts of things that he does, one an­i­mal be­came a clear choice.

“The rig­ging was ex­tremely com­plex,” says Aithadi. “We had to get a very, very flex­i­ble crea­ture that you could squish. The ref­er­ence that we used was an oc­to­pus, and we went with the idea of an oc­to­pus on the ground or in the car or squish­ing into the car. So we had the idea of this big crea­ture that could squish its body to get through the small­est space and get in­side the truck.”

But find­ing that source of in­spi­ra­tion didn’t solve all their prob­lems. The crew still had to watch how the char­ac­ter was put to­gether to make sure it didn’t take on any of the un­ap­peal­ing qual­i­ties of a sea crea­ture — things that would make it off-putting or even gross to those watch­ing the movie. So, they spent months mak­ing sure that Creech still re­tained a cer­tain cute­ness de­spite his de­cid­edly un­der­sea look.

Aithadi also fo­cused on the in­ter­ac­tion be­tween Creech and ac­tor Lu­cas Till by main­tain­ing good eye lines be­tween the two in their scenes. This made their in­ter­ac­tion feel more real and more like a re­la­tion­ship.

“It was very straight­for­ward, old school VFX,” says Aithadi of get­ting good eye­lines. “We had a bunch of props lined up. We had this big la­tex foam ten­ta­cle and a guy dressed in green pok­ing Lu­cas and in­ter­act­ing with him. We also had a low-tech bean bag and another green-clad stunt per­son would move around Lu­cas to get the in­ter­ac­tion. All of this was re­moved and re­placed with the crea­ture.”

Wal­rus In­spi­ra­tion When it came time to blend the crea­tures with the cars, color also played a cru­cial role. Aithadi and his team de­vel­oped the color scheme for Creech and his fam­ily in­de­pen­dently of the cars at first. The fe­male was given a dif­fer­ent color than the male; wal­ruses in­spired the over­all color for the CG be­ings. But the wal­rus in­spi­ra­tion didn’t stop there. The eyes and move­ment of these big, wig­gly be­ings also played a part in the char­ac­ter de­sign.

“What makes a char­ac­ter great is al­ways the eyes,” says Aithadi. “We spent a great deal of time try­ing to give the eyes depth, which wal­rus eyes have, in ad­di­tion to be­ing wa­tery, be­cause it’s very easy for the eyes to be­come

Pre­vi­su­al­iza­tion is the big thing these days.

It used to be geared more to­ward com­plex vis­ual-ef­fects se­quences, mainly be­cause big ef­fects se­quences are ex­pen­sive, re­quire lots of peo­ple, and the money peo­ple want to see what they are spend­ing their money on be­fore they spend their money.

Now the tech­nique has spread into all facets of film pro­duc­tion. The abil­ity to make cre­ative choices be­fore 150 peo­ple are sit­ting around a set, wait­ing for you to make a de­ci­sion is in­valu­able.

The norm is that fa­cil­i­ties like The Third Floor or Proof are hired to con­cep­tu­al­ize se­quences, which will then be cut into an edit for dis­cus­sion. This data is used to de­ter­mine re­quired equip­ment in pro­duc­tion as well as post-pro­duc­tion re­quire­ments.

But what about for the small­ish shoots that should go through the pre­vis process, but don’t have the re­sources to hire a pre­vis com­pany? Sure, you could go into Maya, 3ds Max, etc., and an­i­mate some cam­eras and peo­ple and props. But you risk do­ing things that phys­i­cally can’t be repli­cated on set or, even if they could, it doesn’t pro­vide you with any in­for­ma­tion be­yond how the shot should look.

CineDe­signer R2, cre­ated by cin­e­matog­ra­pher and tech­ni­cal di­rec­tor of pho­tog­ra­phy Matthew Work­man, was made to pro­vide a sim­ple way to cre­ate pre­vis us­ing real-life cam­era, light­ing and grip equip­ment.

Its a plugin for Maxon’s Cin­ema 4D (all fla­vors ex­clud­ing the lite ver­sion come bun­dled with Af­ter Ef­fects), that brings up in­ter­faces: Cam­era Truck, Light­ing Truck and Grip Truck. Each ties into a li­brary of pre­built equip­ment mod­els that cor­re­spond to the real-world stuff. For in­stance, I can put to­gether an Alexa XT sit­ting on an Arri BPL2 plate, which sits on an OCon­ner head, on a 24-inch JLF-TH off­set, mounted onto a J.L. Fisher Model 10 Dolly. Yes, it’s that spe­cific.

The cam­era is then con­trolled by slid­ers within the in­ter­face of each piece of equip­ment. You can pan and tilt with the OCon­nor. You can raise and lower the cam­era with the hy­draulic lift beam on the Fisher. You move the cam­era via the dolly — on or off tracks. And if you want to swap out pieces, it’s as easy as im­port­ing a new piece — a Mitchell 18-inch riser for in­stance — and drag­ging it in the ob­ject list so it ends up be­tween the off­set and the head. In 3D, the new riser snaps in be­tween the other two pieces. It’s pretty snazzy.

The same works for the Light­ing and Grip Trucks. The Lights are work­ing lights in the 3D view­port, so if you build a 3D set in Cin­ema 4D that rep­re­sents the set, you can do some broad­stroke light­ing de­sign. It’s not as 1:1 as I might like, but we aren’t go­ing for pho­to­re­al­ism; we are con­cep­tu­al­iz­ing and plan­ning.

CineDe­signer is about the idea is that you are de­sign­ing your shots with the re­stric­tions of the real-world in mind, which then al­lows you to plan your cam­era, light­ing and grip lists with spe­cific bud­gets in mind. Things will change once you are shoot­ing — but a solid plan is in­valu­able.

You can find more at www.cin­e­matog­ra­phydb.com

I’ve brought up Chaos Group’s VRay, which, in­ci­den­tally, won a Sci-Tech Os­car this year, and Al­le­gorith­mic’s Sub­stance Pain­ter quite a lot these past few years. Mainly be­cause I like ’em, but also be­cause they fre­quently re­lease up­dates. I like them both, but they don’t nec­es­sar­ily play to­gether nicely.

The rea­son for this is be­cause of the way Sub­stance wants to model light and sur­faces in its phys­i­cally-based ren­der setup, and the way VRay likes to han­dle it, don’t math­e­mat­i­cally jibe. As you are paint­ing in Sub­stance Pain­ter, you are get­ting real-time feed­back. You can spin lights around, change the HDR for dif­fer­ent light­ing sit­u­a­tions, etc., but then when you ex­port the maps to ren­der in VRay, they don’t look the same. Sure, you can make them both look cool, and there are work­flows to get them to be­have, but I’ve found those get a lit­tle hinky. And if you are like me, and aren’t paint­ing tex­tures ev­ery day, you can’t for­get all the lit­tle switches and knobs you have to flip and turn.

Well, Kine­matic Lab seems to think this is an is­sue, too, be­cause they re­leased a free plugin for 3ds Max users that sets up your VRay shaders to ac­cept the Sub­stance Maps, places them into the cor­rect chan­nels, adds the ap­pro­pri­ate gamma ad­just­ments.

The re­sults aren’t iden­ti­cal, but they are very sim­i­lar. At least more sim­i­lar than I’ve been able to get with­out a lot of fum­bling and tweak­ing.

So, there you have it. A short and sweet re­view for a piece of free soft­ware, writ­ten by one of the lit­tle guys, by the name of Clo­vis Gay, who just wants all of us other lit­tle guys to be able to make pretty pic­tures with a lot less trou­ble.

Check out the Kine­matic Lab site for the down­load and for lots more free and in­ex­pen­sive tools for 3ds Max. https://ho­cus­pocus-stu­dio.fr/tools. Todd Sheri­dan Perry is a vis­ual-ef­fects su­per­vi­sor and dig­i­tal artist who has worked on fea­tures in­clud­ing The Lord of the Rings: The Two Tow­ers, Speed Racer, 2012, Fi­nal Des­ti­na­tion 5 and Avengers: Age of Ul­tron. You can reach him at todd@tea­spoon­vfx.com.

Newspapers in English

Newspapers from USA

© PressReader. All rights reserved.