Brad Wardell's site for talking about the customization of Windows.
Published on December 13, 2008 By Frogboy In Elemental Dev Journals

Our story so far...

In Elemental there is no such thing as a knight or a wizard or an archer.  Instead, players design their own units.  If you wanted to call a unit a knight, you might take a man, equip him with some armor, give him a helmet, arm him with a sword, and pair him up with a horse.

The armor, helmets, swords, etc. are things you manufacture and thus have some control over how they look. Normally, customizing an individual unit falls only in the realm of role playing games. And what they tend to do is called texture merging. That is, they simply blend various textures together.

In Elemental, what we're doing is actually giving each item its own heft with its own physics.  The trick is to find a way to do this that still lets it run on lower end hardware so it has to be done smartly. At the same time, you want the guy with that new Core i7 with the latest nVidia or ATI card to look at it and go DAMN that's cool.

The example we have going is a knight that we've equipped with armor, a sword, a helmet, along with a horse with its own armor.  When they move, each item moves as if it were real. That is, the armor on the horse when the horse is running moves like you would expect.  The knight riding the horse moves on the horse as you'd expect and even the sword dangling from the side moves as you would expect. 

Normally, to get such an effect, you would have to model/bone/rig/animate the entire unit together.  The breakthru here is that these elements are all independent and created by the players and they just work together.  I'll try to get a little video or something to show this in action next time.

This will really make the tactical battles really compelling. Since so much of the game revolves around the premise of massive unit differentia, you will see some really breathtaking battles I think.


Comments (Page 2)
6 Pages1 2 3 4  Last
on Dec 13, 2008

At the same time, you want the guy with that new Core i7 with the latest nVidia or ATI card to look at it and go DAMN that's cool.

Nope. I just want an immense elemental game to run as smooth as an Immense, common settings GC2 game. Is that a realistic hope at this stage?

on Dec 13, 2008

Like maybe he's making fun of Spore?

on Dec 13, 2008

So the new release date is what, 2015?

I look forward to seeing the video (or any other new media you care to dribble out to us!).

on Dec 13, 2008

Nice It's good to see the engine is being worked from the ground up to support more open and strategic gameplay.

on Dec 14, 2008

This sounds awsome, but, not to be a negative nancy....I really can't imagine how were each part of each soldier is effected individually is going to run on anything but top end machines...I imagine that is soooo much processing power.

on Dec 14, 2008

...with every new dev journal I get even more addicted to this game. I didn't even know it was possible to be addicted to something you've never tasted!

lamperti
One of the issues I have with games where you design units (Usually space ships,) by putting together things you have researched and/or built is that you find yourself needing to create a new design everytime there is an incremental change in technology.  I.e. I have just researched shields that are +5 instead of my shields that are +4, so I need to design a new ship type, (Or multiple ship types,) with the new technology.  This can lead to micromanagement, a difficulity in unit uniqueness and a naming problem.  This is something that could be an issue here.  Of course it could be managed well, but these are potential difficulties I think might come up.

I second this. There definitely needs to be a good way to upgrade unit types without having to constantly build new ones. Rhishisikk's idea seems pretty good (upgrade to best), but what the computer thinks is the best is often not what you want. So for the sake of both upgrading units and for the sake of multiplayer, building/editing units needs to be quick and easy. Maybe once you've created the base unit it could just be a matter of selecting components like swords, helmets, etc from a list, and the edit places them automatically.

on Dec 14, 2008

I'll wait with any comments, positive or not, on this topic until its clearer what it actually means for the gaming experience. Flexibility in producing units is great, but at the same time: if there are no "unit archetypes" or rules at all, the result might just end up: bland. With heavy micro-management.

A comparison can be made with Dungeons & Dragons normal RPG, where I have never been a big fan of the multi-class characters. When everyone, warriors, thieves, priests, start walking around throwing fireballs, the magic of the mage is somewhat lost ... If anyone can learn priest spells, the real priests are not that exciting anymore. I'm not against some such flexibility, but it has to be rare and handled with care for when it is really appropriate.

For the graphics it sounds really cool though. I hope it will all turn out right.

 

It's the player's call to make an army of generic unit with swords, pointy hats (wizards), helmets, horses and an invisibility cloak or to make a more normal, differentiated army. In the end, it will be about balance. Yes, someone can make the first type of army with 100 master-of-everything soldiers, but i bet it will be much more expensive than an army of 25 soldiers, 25 archers, 25 cavalry and 25 support magicians.

So in the end if you actually want to win the game (and lets suppose the player is in against an human or a good AI) he will have to specialize much more his troops. Just because you have freedom to design your units doesnt mean these units will be generic, or even good, or cost effective.

on Dec 14, 2008

pigeonpigeon:  Okay, so how about this: we keep a set of templates.  For example, I have both a holy paladin and a knight template.  I want my paladins to keep their holy sword instead of getting the new flaming swords, but I want my knights to make use of the new technology.  I select UPGRADE on the knight template, move in the flaming sword, and save the template.   New knights will be built with flaming swords.  My idle production goes into making flaming swords, which are caravaned out to the units, and bring back the old weapons.  (So yes, my garrisons will upgrade first, being closest).  When a unit is in town, I can select UPGRADE TO TEMPLATE: [template name] (Or maybe UPGRADE lists the templates that match the unit's current configuration) and have it pop up with X weeks, and stick the unit into the production queue. 

This also melds into the idea of supply caravans, weapon/armor wear and tear, and other automatically supplied field needs.

on Dec 14, 2008

Maybe there should be the options to abstract unit choices? So I could equip a unit with 'sword' and it will pick the best sword available (perhaps there needs to be a way to reserve weapons). Or I could equip a unit with 'flaming sword' and it will pick the best available flaming sword, and have priority getting it compared to the guy just picking any old sword, or I could equip it with 'flaming sword IV' and it will only ever be equipped with that, and gets first pick.

(Obviously if upgrades are distinguished in the game by sticking numbers after them then it will be a disappointment.)

on Dec 14, 2008


In Elemental there is no such thing as a knight or a wizard or an archer.  Instead, players design their own units.  If you wanted to call a unit a knight, you might take a man, equip him with some armor, give him a helmet, arm him with a sword, and pair him up with a horse.

I like this method of unit construction.


The armor, helmets, swords, etc. are things you manufacture and thus have some control over how they look. Normally, customizing an individual unit falls only in the realm of role playing games. And what they tend to do is called texture merging. That is, they simply blend various textures together.
In Elemental, what we're doing is actually giving each item its own heft with its own physics.  The trick is to find a way to do this that still lets it run on lower end hardware so it has to be done smartly. At the same time, you want the guy with that new Core i7 with the latest nVidia or ATI card to look at it and go DAMN that's cool.
The example we have going is a knight that we've equipped with armor, a sword, a helmet, along with a horse with its own armor.  When they move, each item moves as if it were real. That is, the armor on the horse when the horse is running moves like you would expect.  The knight riding the horse moves on the horse as you'd expect and even the sword dangling from the side moves as you would expect. 
Normally, to get such an effect, you would have to model/bone/rig/animate the entire unit together.  The breakthru here is that these elements are all independent and created by the players and they just work together.  I'll try to get a little video or something to show this in action next time.

Why?  I know it sounds cool but why spend all this extra time on eye candy?  Will all this time spent on the physics based animation system carry over into the combat physics?  Will a hit from a large heavy two handed weapon send a small unit flying across the screen as a projectile where a normal size unit may be able to block the hit, and a large armoured unit just ignores the hit and continues with his swing?  If it is just for the eye candy I don't see that it would be worth it.

Sammual

P.S.  Watch the videos for some work others are doing on physics based animation. http://www.naturalmotion.com/euphoria.htm

on Dec 14, 2008

Why?  I know it sounds cool but why spend all this extra time on eye candy?  Will all this time spent on the physics based animation system carry over into the combat physics?  Will a hit from a large heavy two handed weapon send a small unit flying across the screen as a projectile where a normal size unit may be able to block the hit, and a large armoured unit just ignores the hit and continues with his swing?  If it is just for the eye candy I don't see that it would be worth it.

We like eye candy.  

on Dec 14, 2008

Sammual
Why?  I know it sounds cool but why spend all this extra time on eye candy?  Will all this time spent on the physics based animation system carry over into the combat physics?  Will a hit from a large heavy two handed weapon send a small unit flying across the screen as a projectile where a normal size unit may be able to block the hit, and a large armoured unit just ignores the hit and continues with his swing?  If it is just for the eye candy I don't see that it would be worth it. [...]
You're missing the point. What it does is allow for piece-together unit creation, without forcing the created units to remain static or be subject to clipping.

It's not "just" eye candy.

Edit: Or, say, allow a rider to be knocked off his bear, and have the two fly in a direction each without clipping or workarounds.

Frogboy

We like eye candy.


You know what kind of candy I like? The bear kinda candy!

on Dec 14, 2008

I personally love physics in games I prefer first order physics (affects game play) but second order physics (eye candy) is great too. If a game has a really in-depth and meaningful second order physics system it makes the game dramatically more interesting to watch and keeps the feeling of uniqueness far longer. I personally get extremely bored seeing the same exact thing over and over so when a game has a nice second order physics system that makes me want to pay attention each time a unit attacks / is struck / dies I don’t mind even if it would mean I have to wait that much longer for the game. 

on Dec 14, 2008

New computer this Feb coming up! I am good! OH I can't wait to see this game in action.

on Dec 14, 2008

Frogboy

We like eye candy.  

Everyone likes eyecandy. 

I am a big fan of dynamic, procedural, physics based animation.  Take a look at http://www.naturalmotion.com/euphoria.htm for the best useable example I have seen. 

The problem is scope and scaling.  Unless this is a core feature that will replace the standard animations everywhere I can't see it being worth the cost in terms of programming time and CPU time.  If this were a small squad real time combat game I would be jumping up and down about this.  For a turn based strategy game with easy access modding I just can't see the benifits.  For example, Stardock will test the game engine with horses, elephants, and dragons as mounts.  The physics is easy for the first two and Stardock writes in a physics 'cheat' for the dragon (Anyone ever tried to make dragons fly on paper as part of an enginering problem, it doesn't work).  How does 'Vandenburg' mod in his pegasi?  Once again the physics doen't work.  The wings have to be so large that no conventional material is strong, light, and flexible enogh.  Will modders be able to ignore the built in physics to create what they want?  If they can how does the animation get created?

Sammual

6 Pages1 2 3 4  Last