Development Stories About the Creation of Diablo III Monsters - Community Spotlight

Diablo Pubblicato 17/09/2022 alle 07:05 da silec

asked the community on redditA similar post was made one year agoDesert Waspyuimiop via reddit: What nightmare spawned the Act 2 bees?Nathan Lutsock via reddit: The origin is pretty simple—we wanted to update the D2 dot monster, which was crazy silly and fun, and yes, create a bullet hell experience.

The best way to update the D2 swarm required us to give them a real model, which we managed with several insect models. The tech artists and the animators were the real stars, making a flexible rig that could move all the elements around and animations that made them move chaotically in a non-repetitive way. FX filled out the swarm with cosmetic tiny insects.

All the D3 monsters are meant to contribute something in the context of other monsters. The swarm works in a big group of similar swarms, but really shines as an add-on to a pack of melee monsters. Making players move around is an important goal, and those super-bright big bullet projectiles are meant to scare you out of standing still when slugging it out with Lacuni and such.

The actual wasp model was so effective that we ended up using it all over the game. The wasps shot wasps. The Witch Doctor shoots wasps. Some spells summon wasps. It’s the same wasp over and over. It’s just a super processor-efficient model that has a good, distinct shape.Mallet LordCosmic_Lich via reddit: Mallet Lords look so cool and are apparently second to the lords of hell. Is there anything you could say about them?Nathan Lutsock via reddit: Act IV was developed last, and Act IV is intentionally short. So what you have is a dev team that has worked out all their issues figuring out how to make their game, and is now turning out high-quality monsters very quickly.

With how well the monsters were going, Jay had taken a pretty hands-off approach. He simply told us that the lineup required a heavy hitting large enemy. He also mentioned that the Act IV monsters were getting complicated, and recommended we build the character around a simple mechanic: a single attack with a massive windup that will punish you for not moving out of the way. In a game with dozens of enemies onscreen at once, basic mechanics like that make it easier for the player to process what is needed of them. It’s actually more fun, a lot of the time, than more complicated monsters.

Regarding names: as the producer shepherding the monsters from first thought to completion, it was my job to give everything a production name. Final names change all the way up to launch, and we needed terms that we could use to refer to our creations until the lore guys laid down a final moniker. I had to give everything a name that contained the gist of the creature but didn’t trap the artists into a box. If I put an empty space on the monster wall and labeled it a Swamp Pummeler, that was better than calling it a VenomAxe, because if I called it a VenomAxe my concept artists were most likely to give me a green creature with a dripping handaxe. It was my job to empower their imaginations, not constrain them.

So I told people this was a Mallet Demon, and whatever he may look like, he would wind up and pound people. This name was a little more descriptive than I normally made them, but I really wanted to communicate the gameplay function.

Victor Lee was the concept lead and he always tried to turn conventions on their head. To him, long arms could be swung like mallets. He turned in a concept that I love (happens to look like an album cover) but was pretty surprised by. It was much more “giant cenobite” than I was expecting. But you hire Vic because he does dark and twisted better than anyone, not because you want to change his original concept ideas into what is expected.

I was privately not sure if it would work, but it’s my job to not quash creative exploration. Some designers were more doubtful of the design when they saw it, but I urged them to trust us. That’s producer talk for “let the art team worry about making art great.”

We had to do a couple things to make it work. The modelers and the animators worked together to get a model that could move those arms around like we wanted, and we worked with design on the timing. The tech artists gave us some killer FX on the swings and the impact and the attack started looking beefy. By the time audio came in with that “bumbumbaBUM” sound we were loving the character.

He wasn’t what anybody expected, but we loved him. We were happy to get another really tall character in the game, which we felt we could have done more often. For my part, I pushed for the Mallet name to stay as the final, because it’s a good example of a name that doesn’t not literally match up but perfectly conveys the function of the creature. Also, creatures of the Hells always have that Metal vibe to them.Treasure GoblinCpr_Cold via reddit: Goblins were kinda out of character for Diablo universe. Was it just a fun mechanic to implement some chance for exciting loot to the game or was it something in the heads of devs when they wanted to create cool monsters?Nathan Lutsock via reddit: Treasure Goblins. A personal favorite, but maybe everyone’s favorite?

You hit on something interesting here with your question. A lot of the time, monsters are born out of typical game needs: a swarmer, an ambusher, an uncommon challenge, and so on. The rhythm of the game requires certain kinds of monsters. But some monsters are born out of unique requirements.

Jay Wilson is a good monster designer. He understands the fundamentals, but he also has a good understanding of categorical need. He specifically wanted a monster that could cut across all game content and environments, fitting in literally anywhere, and providing a unique “stop all normal gameplay and do this NOW” experience for the player.

He was really open about how we accomplish this, but he set us on a path when he said, “Kinda like those little thief bastards in Golden Axe.” Once we had that personality in our heads, we couldn’t shake it. We were making a little bastard that deserved to get his butt kicked.

We asked Vic Lee to do a sketch Diabloizing the Golden Axe thief and his first drawing nailed the concept. Scrawny, greedy, an obvious ne’er do well. We considered having him actually STEAL from the players but we realized that he would be compelling enough just by virtue of his rarity. The concept of his escape to a magic realm of treasure evolved very quickly, and we planned on the Realm of Greed pretty much the first day the monster was born. It was, however, almost a surprise that it was later created, because SO MANY game ideas don’t survive the development gauntlet and actually make it to completion.

I was in love with the character early on, and helped it along every chance I could. I worked with the folks getting the plushie made and when the employee statue was developed. I have a terrific fondness for that little jerk.Soul RipperMoney_South300 via reddit: I really like the tongue demons. Just opening a door and having 25 tongues immediately latch on to you is funny.Nathan Lutsock via reddit: The Soul Ripper, AKA the Ripper (circa 2009) AKA the Thorn Ripper (circa 2006) was a monster that came a long way to finally end up in Diablo III. This creature was one of many that we saved from deletion when we moved production down from Blizzard North and restarted the game. There were a lot of monsters that we wanted to keep, even though many of them required work and I was a bit naive about how much work we would spend in modernization. They were simply creations of an older era, with poly counts, texture sizes, and rigs that didn't match what we were doing going forward. Their animation rigs were entirely redone, their models updated to the point where they were basically new monsters, and their textures were completely reinvented. I talked about the Tormented Stingers, a related monster, last year.

As with the Tormented Stingers, the Ripper was part of a family of body horror creatures. We had creatures that were borne of corpses, mish-mashes of bodies, a bloated head that crawled around, and then this guy. His most notable features were his massive, oversized eyes, which looked amazing in his animations, and was very readable from game camera, even at his original lower texture size. Beyond that, he didn't have ANYTHING. He didn't have a cool attack, his locomotion wasn't really interesting (neither creepy nor intimidating), and he didn't even really threaten you. Like some of the other members of his family, his value really came from being another freaky deaky corpse thing. I am not badmouthing the original creators of the character because he was an unfinished piece in an unfinished puzzle. He wasn't really cool in the context of the Blizzard North build, and he was less cool imported into the game as it was shaping up at Blizzard South.

But I am something of a packrat. I believe that one facet of quality is quantity. A Diablo game is one you play for possibly hundreds of hours. In an experience of that length, the number of environment changes available in the game, the number of armor looks, the number monster appearances really matter. Unless I was getting chewed out by my Lead Designer or Art Director, I was not going to throw out any assets that were halfway decent and could be salvaged for less than the cost of a new one. Sometimes, my estimates of those costs were off, and, well, "Sorry, team!" But I don't regret the success stories. And in the end the Ripper was one!

I wish I could recall which combat designer to credit with the tongue. This guy had been sitting in the game engine for a few years, waiting for a champion, and one day a designer asked me about him. It might have been Jason Bender, but whoever it was, they started asking what we could do to make it cool. We somehow got onto it having a tongue attack, about doing a Batman zip-attack, and even reeling the player in. The Tech Art team loved it, and the programmers had it working in less than a day. It was astonishing to see this creature that was very much headed for the cutting room floor suddenly take on this new life. It provided a unique play experience, and suddenly that weird crawling locomotion appeared to be a natural reflection of a lurking, prowling predator. Like we planned it!

There were moments where we were getting bounced around a lot, and we had an instinct to pull back and tame the effect. But we resisted. Sometimes you needs weird stuff, stuff that stands out and takes you out of the hypnosis of the combat flow. You need some funny stuff to alarm you and crack you up, even in a Diablo game. I know that is a hot take, but a Diablo game is still a game, and games should evoke a spectrum of feelings and experiences, or else they become monotonous.

The last thing I can say about the character is his portrait in the achievements of the game (found midpage here). Ingame, the little bastard doesn't have much of a face, or he didn't until we got this stellar close-up piece, which I would like to have framed. I’m not totally sure who did this one, but when I see the paint work of the creature’s musculature, I see vintage Josh Tallman art. Which is funny, because Josh has never painted something as cute as this purpose.Dune DervishLepoelad via reddit: Dune Dervishes were rather unique and seem like they have some kind of interesting backstory. I’d love to hear more about them!Nathan Lutsock via reddit: The Dune Dervishes were born early in development after the game was restarted in 2006. When you are at a place with a game, like we were, you need to start making statements about what you are, and what you aren't.

We were working in Act 2, which we had decided would be a desert, and we wanted to explore some boundaries on what we would do. We began work on the Sand Shark with the express intention of seeing if we could elicit some more complex combat gameplay in a Diablo game without breaking the rhythms of fast combat. We tried a lot of things that didn't work. We made the monster unkillable in certain states, we toyed with kiting gameplay... it was just too much for what Diablo combat is.

At the same time, we were making other desert enemies that would push boundaries for Diablo gameplay. Victor Lee knocked out a concept that got an immediate positive response from the team (dramatic shapes, thought-through movement, fresh design always helps!) and we knew what we wanted to do. This would be a monster who would out-whirlwind the barbarian.

If I recall right, our animator on this one was Chanon Thareechit, who was terrific, and he worked with the modeler and the tech artists to give him very specific controls. Though we had access to simulated cloth, Chanon hand-animated the draping pieces around the creature's waist. He also had very specific plans for how the character would wind its chain arms around its own torso while 'at rest', and worked out special animation rigs with our tech artists. Understanding that the big deal behind the monster was the spin, he iterated a lot on the creature's main attack. It ended up changing a number of times, and I believe that we ended up with an intro-looping spin-outro animation that gave designers that ability to change the duration of the creature's attack without requesting new animation work.

For context, this is happening in a very art-led way. Some monsters are led by a visual, others by an idea, some by a specific need. In this case, we wanted to create an experience where players would have a chance to attack a monster, but also need to change their behavior in response to cues. Not revolutionary stuff, for sure, but we were trying to find the amount of complexity that would work for us. Clearly, we had stumbled with Sand Shark, but we wanted to do more than what you saw in D2. More, that is, if it didn't make the game slow or not fun. At this time in development, we were learning what felt good to us, and what our shared vision of the combat experience was.

So Chanon's work moved over to Tech Art, who really gave that spin all the jazz. They are the ones who do the FX, and if you can make a commitment, like, "we promise this will always be spawned in the desert" (which I hate doing), then they can take that commitment and build sand FX into the monster's spin. They gave the spin all the energy and visual coding that make you immediately say, "I have to move out of there!" There is a a SURPRISING amount of education involved in the visual FX visual language, more than I can go into here, but trust me that these folks are masters and harsh critics of their own work. As an example, I was always disappointed by the bare model, which has 'nothing' under its skirts. FX creates the illusion of mass where legs would be, though no model exists. I love the finished character.

Design was very behind this character, and we were all moving the same direction with his functionality. It was about executing a simple idea well, and everyone did their part. The character was visually small, quiet, and vulnerable at first, then exploded with speed and visual noise warning you to stay away, and then spun down once more. We mostly had it on the first try, but two things came later that elevated the character.

First, we got the projectile bounce working on the spin. For various reasons, it wasn't set to reflect 100% of incoming projectiles, but it functioned often enough that you got it when you saw it in action. It really sold the idea of these blades having offensive and defensive qualities, and the creature being immune during the spin.

Second, we got the death, so, so right. Julian Love, our lead Tech Artist, had a clear idea that this character would die in a memorable way (he shared this story in his own excellent GDC talk). As the character was basically a helicopter, Julian saw its death a helicopter crash. The resulting animation is wild, and is actually a lesson to anyone making 3/4 camera action games about how to use vertical space. I would put it in the top 5 monster deaths in Diablo 3, a very esteemed position.

In the end, the creature did cause some frustration—players do not like enemies to have immune states, but I believe the cost was worthwhile. He nudges the line for what we asked of players, and he helped us decide what qualified acceptable combat for our game. He was a learning monster, and a good one. Making games is technique, and experience, and economics, but it is also art, and momentum, and morale. It was wonderful to have this creature come together and be so successful at the time that it did.Flayed OneMajinDLX via reddit: Not an interactable monster, but who are the big tortured giants in Act III on the way of Cydaea? They look like they stepped out of a Hellraiser movie. I absolutely adore their design. Any background stories or just aesthetical value?Nathan Lutsock via reddit: It’s pretty well known that there was a version of D3 being developed at Blizzard North back in 2004. I even worked on it a little. When we brought the project south, we were determined to keep everything cool that we could, and the giant hanging demons were the coolest thing in the game.

Designed up north by Victor Lee, the original hanging demon was part of a very WIP boss fight. The fight was interesting and different, but not very Diablo in the gameplay. Imagine a confusing Belial fight with less walkable area and only one spot to melee from. But we kept the model and animations with the intention of using them somehow. Ultimately, everything we kept got a massive rework that didn’t save us much, if any time. But no regrets, that thing is cool.

Not wanting to make another boss fight, we found the Demons we’re doing everything they needed just being part of the horrific landscape of the Crater. We had a hundred ideas about their purpose and function, and the writers just ran with the one they liked the best. I think my favorite concept was Vic’s, that these were giant Angelic warriors from the first celestial war, and their pain powered Azmodan’s creations.

Accedi per Pubblicare un Commento Commenti in Inglese (4)

Scrivi un commento

Messaggio Precedente

Messaggio Prossimo