Torchlight 2 Catch-All

How did the Alien Swarm puzzles involve physics?

Yonder wrote:

How did the Alien Swarm puzzles involve physics?

Explode the barrel and the giant concrete cylinder thing rolls into the metal slab, knocking it over and forming a bridge.

Although it could well involve physics, I'd say because it's critical it happens perfectly every time, that was scripted, or had the physics pre-cooked like you see the big end of round payload explosions in TF2 or the bridge demolition at the start of episode 2.

Given that their building on the foundation of TL1, it might be hard (or a bad use of resources in a small developer) to add something that could only be used for general realism, than something more central to the gameplay.

Scratched wrote:

Although it could well involve physics, I'd say because it's critical it happens perfectly every time, that was scripted, or had the physics pre-cooked like you see the big end of round payload explosions in TF2 or the bridge demolition at the start of episode 2.

Given that their building on the foundation of TL1, it might be hard (or a bad use of resources in a small developer) to add something that could only be used for general realism, than something more central to the gameplay.

Besides, physics are only useful when you have box stacking and portal jumping puzzles anyways.

To be serious, though, it would probably be more bad use of resources for them than anything else to add 'physics' to the game than anything else. Especially if they're trying to translate this to an MMO where then the physics would have to be computed at one single point and then transfered to all the players viewing. That just sounds like a huge resource sink in development and live implementation / bandwidth.

Then again, I know nothing about developing software / network traffic.

The most obvious place I could see it used is ragdolls on killing an enemy, as once they're dead it doesn't really matter where the end up, the same as any other shooter for the past 6/7 years or so. The problem comes when you need to keep complex physics in sync, HL2DM (or perhaps some mod?) is still the most complex thing directly involving physics I can think of, and even then it's simple projectiles. I honestly can't think of any game with complex physics that is multiplayer (Cellfactor maybe? *sigh* physx ).

Even then, you would need to think of a way to make physics worth adding to the main hack'n'slash gameplay of Torchlight.

Scratched wrote:

Although it could well involve physics, I'd say because it's critical it happens perfectly every time, that was scripted, or had the physics pre-cooked like you see the big end of round payload explosions in TF2 or the bridge demolition at the start of episode 2.

Yeah, I'm pretty sure it's faux physics.

/subbed

I would think physics should come into play with corpses taking up space and you having to work around them or some such or using corpses to make impromptu walls or large moving projectiles (in the case of hitting it with a large blast).

Queueball wrote:

I would think physics should come into play with corpses taking up space and you having to work around them or some such or using corpses to make impromptu walls or large moving projectiles (in the case of hitting it with a large blast).

That sounds like fun..... Great for pathfinding for pets and AI.... great for your character getting hemmed-in or having your magic projectile/arrow blocked from its intended target.

Actually, i think that sounds like a really bad idea.

Arguably, I'd say in one sense it's bad, because it'd be tougher to navigate and such.

But how cool would it be to have a huge pile of corpses. =P

While it's not a realistic game, it'd add a twist.

Physics would most likely just be taken out for het mmo, so why put them in for part 2?

I am excited for this and hope they annouce more classes soon! I am so happy for the looting being done the way it is. It is about time. I hated having to focus more on clicking an item that is about to drop then playing the actual game.

I want physics for the 'eyecandy'. You don't need them in Diablo or Starcraft, but it certainly looks awesome with them. Just make it so that you can disable them, what's the big deal? Other than the MP that's the only thing I wanted in the first that was missing.

liquid wrote:

Just make it so that you can disable them, what's the big deal?

Dev time and cost mainly. At a certain point they have to start passing that cost on to the customers. I'd rather it be cheaper than have physics which I don't really care about.

juv3nal wrote:

I'd rather it be cheaper than have physics which I don't really care about.

This.

Low price > physics

They have probably 90% of the work cut out for them for the sequel. Most of the work is on the p2p and new interface/skill and the art team. It wouldn't really add that much if at all to the price. Also, the engine has a built in examples of how to combine it with other libraries such as for physics.

There are physics in Diablo and Starcraft? I've not played SC2 but there were no physics that i remember in either D1&2 or SC. All pre-rendered sprites AFAIK.

The interesting thing about p2p multiplayer is that there's a relatively large fundamental difference than with server/client connections. P2P typically deals with lower connection rates and lower speeds. Server/client really needs to scale a lot which is a different problem.

Duoae wrote:

There are physics in Diablo and Starcraft? I've not played SC2 but there were no physics that i remember in either D1&2 or SC. All pre-rendered sprites AFAIK.

Diablo 3 and StarCraft 2, sorry for the confusion.

Physics are a nice to have. Sacred 2 has it, and dead kobolds do roll down hills amusingly.

I think the problem is that good physics is quite hard to get right, and done badly it's pretty bad. Think dead bodies in Gear that suddenly weigh nothing. So to do it properly will take a fair amount of dev cycles and essentially be a distraction from the main direction.

liquid wrote:

They have probably 90% of the work cut out for them for the sequel. Most of the work is on the p2p and new interface/skill and the art team. It wouldn't really add that much if at all to the price. Also, the engine has a built in examples of how to combine it with other libraries such as for physics.

I don't know, I think you're making too many assumptions about Runic games and Torchlight 2. I'm not saying you're wrong, but I think it would take an insider to correctly determine how much time, money and other resources to implement physics.

Loved Torchlight and looking forward to TL2. I modded the snot out of vanilla TL, so importing to TL2 would probably not be an option for me even if it were available.

If you've ever played Titan Quest, much of what makes the combat so satisfying is the physics. The outrageous feeling of power you get from opening a can all over a beastman and seeing his corpse fly half way across the map is very satisfying. Maybe there are other more realistic uses for physics but 'fun' is the best reason I can think of.

The thing is, TL2 is building upon the established base of TL1 and how it does things. I doubt it's a simple change from "IF monster.health <= 0 THEN monster.play_death_anim()" to "... THEN monster.ragdoll(last_blow_vector, last_blow_force)". I'd say you have to refactor a lot of things such as collisions with the entire world (how much of it was just window dressing?), and what those corpses can and can't do, how they react with the world and other creatures in it, how do you stop the physics from looking stupid (I bet everyone's seen something clipping through the world and glitching out), do you put constraints on every physics object and character skeleton so it can't bend/stretch where it's not supposed to.

I don't think it would be quite as bad if Runic were planning for it from the beginning, but it would need a fair bit of work and rework of old systems. For a company that doesn't want to grow beyond 45 people it's a big ask for seeing varkolyns splat against a wall.

garion333 wrote:
liquid wrote:

They have probably 90% of the work cut out for them for the sequel. Most of the work is on the p2p and new interface/skill and the art team. It wouldn't really add that much if at all to the price. Also, the engine has a built in examples of how to combine it with other libraries such as for physics.

I don't know, I think you're making too many assumptions about Runic games and Torchlight 2. I'm not saying you're wrong, but I think it would take an insider to correctly determine how much time, money and other resources to implement physics.

I don't think so. Same engine, the modifications they made are there, etc. The only real new thing to add is content, update the UI, new algorithms (or however that works) for the randomized events (if they are new to them at all) and the story. Nothing technically new or spectacular compared with the first game. I could be wrong but I think it's worth an upgrade over the first title, considering how well it did.

I'm not saying it's as easy as putting another if-then-else statement, but it shouldn't be that hard for them considering the engine allows it and it's not something they have to build from the ground up.

liquid wrote:

They have probably 90% of the work cut out for them for the sequel. Most of the work is on the p2p and new interface/skill and the art team. It wouldn't really add that much if at all to the price. Also, the engine has a built in examples of how to combine it with other libraries such as for physics.

If they can do it without it taking longer or costing more, I'm all for it, more power to you.

But it seems to me that unless they think they can sell significantly more copies by adding a "has physics" bulletpoint to their marketing, they're going to have to up the price to make it worth their while. No matter how easy it is for them to do, it is still going to cost them some non-zero amount of dev-time to do it. And we've all seen shoddy implementation of physics, so factor in however much extra dev-time it would take for them to do it right.

I just think it's an unneeded cosmetic feature that would please a small percentage of their potential player base. The majority of the people who will want to buy TL2 will do so because of two things:

1) It's more Torchlight
2) Co-op

Since the MMO is unlikely to have Physics and i doubt they'd do a TL3 if they're planning to release the MMO in 2011 / early 2012 then it's just a fork in technology that is essentially a waste of time.

MrDeVil909 wrote:

Physics are a nice to have. Sacred 2 has it, and dead kobolds do roll down hills amusingly.

I think the problem is that good physics is quite hard to get right, and done badly it's pretty bad. Think dead bodies in Gear that suddenly weigh nothing. So to do it properly will take a fair amount of dev cycles and essentially be a distraction from the main direction.

Dead Space also had some goofy physics. Weightless, foam rubber corpses all over the place kind of detracted from the atmosphere. (AHHHH, the atmosphere!)

Tell me more about this railroad building stuff; why is the character building a railroad and where is it going?

The world is somewhat unexplored and there is a frontier, so the idea of all this is to tell a story and give the character a background. We don't like to do super basic, stereotypical character classes that every fantasy RPOG has. We like to put a little twist on everything. So in making our melee character tank class we thought it would be really cool if he had sort of an engineer bent to him as well. So what we've done with that is that it actually becomes a part of his combat mechanic. He builds up electric charges as he hits and those charges have a chance to express when he does certain skills and there's also certain skills that absorb those charges in his armor or his weapon to be discharged when he either gets hit (when it's in his armor) or when he hits with his melee weapon. So it sort of lets us put a twist on basic melee tanking and a new mechanic with these charges to this guy. It's really all about making character classes that people intuitively understand but have a little bit of a twist to them.

From interview over at GameShark. That was a little bit more specific info than I remember hearing before.

Scratched wrote:

I'd say you have to refactor a lot of things such as collisions with the entire world (how much of it was just window dressing?), and what those corpses can and can't do, how they react with the world and other creatures in it, how do you stop the physics from looking stupid (I bet everyone's seen something clipping through the world and glitching out), do you put constraints on every physics object and character skeleton so it can't bend/stretch where it's not supposed to.

And, more importantly, which is cooler:
1. Ragdolls.
2. Every single mob exploding in a shower of gibs when they die.

I'd go with 2.

I want option 3, whacking a mob, sending it flying through the air and into a wall where it explodes in a shower of gibs.