ZeroK-RTS / Zero-K

Open source RTS game running on the Spring/Recoil engine
https://zero-k.info
GNU General Public License v2.0
695 stars 209 forks source link

Skuttle Design Ticket #4380

Open GoogleFrog opened 3 years ago

GoogleFrog commented 3 years ago

Skuttle has problems. Link them here.

GoogleFrog commented 3 years ago

Capture Skuttle for reference: https://github.com/ZeroK-RTS/Zero-K/issues/4185

GoogleFrog commented 3 years ago

Skuttle as a generally usable unit suffers from being a meme that generates moments. A small number of successful Skuttles appears to be fun for the more dedicated community as a whole, and it is fun to blow stuff up so completely in singleplayer. A more reliable Skuttle would necessarily lose these uses unless it gained additional applications in addition to its current high-risk application.

A more useful Skuttle should have interaction after it happens. Skuttle is currently unique in that, once it hits successfully, the interaction between the players is over. The defender never gets to mitigate or interact with the situation. Imp and Widow solve this by starting a game of Defend The Stunned Unit(s) while Snitch often exists in multiples, often creates damaged units, and tends to imply the existence of an area-cloaker to hunt down. Scythe, Ulti and Widow start games of Find And Kill The Cloaked Unit, with the stakes at least being the efficiency of the attack.

Most effects with counterplay are less cool than a high-damage explosion.

Anarchid commented 3 years ago

IIRC Saktoth proposed a non-suicidal burrowing burning spit version of skuttle. Set units on fire with significant duration and damage instead of killing them outright.

The defender gets to choose to play Doomed Berserker Rage or Triage And Repair. Attacker gets to play Avoid The Berserker or Prevent Repairs. If the unit is also non-suicidal as in that suggestion, defender gets to play Kill The Skuttle.

Three obvious downsides:

GoogleFrog commented 3 years ago

I am aware of it. At the time the problem was that all fire currently has the same DPS. This is still a problem. Adding variable DPS requires some sort of rule for mixing different types of fire and a way to visualise the fact the the unit is on a different type of fire. We now have an additional issue: Fire is extinguished by water. Skuttle would essentially require a new type of fire, and that has a hefty complexity penalty.

In addition to your downsides:

Licho1 commented 3 years ago

Isn't the "fix" to make it untargetable in air?

GoogleFrog commented 3 years ago

I don't see what that fixes.

Anarchid commented 3 years ago

IIRC there was also an idea for the Limpet to be a literal limpet, a parasite headcrab that would attach to units and slow them down and other units could shoot it off.

What if we combine the concepts into a Frankenskuttle?

The unit is a parasitic headcrab that attaches itself to target and becomes Glint for a fixed period of time, during which it causes continuous melee damage and welding-sparks effects. The max total damage is thus known by both attacker and defender, and they can make easy decisions based on it. However, the attacker has an option to detach the headcrab at any point; the defender in turn can try to ignore or outrepair it. Multiple headcrabs can pester the same victim to stack their damage.

After the charge runs out or the host dies, the headcrab is forced to detach from the victim and become vulnerable again. If the target dies violently enough, the headcrab dies with it.

It could be cool to allow the defender to remove the headcrab with some sacrifice, but the colvol geometries probably make shooting the parasite directly an impractically tedious task to implement. Could allied units be allowed to instead cause damage to the infected unit, with a proportion of that damage applied to parasite?

Licho1 commented 3 years ago

I like that idea for limpet replacement. The many crawling bombs are pretty similar and boring otherwise. It requires extra gadget though, or do we have some reusable attach/detach logic around? Also for AI it will need to replace attack with attach or something.

Anarchid commented 3 years ago

I like that idea for limpet replacement. The many crawling bombs are pretty similar and boring otherwise.

Limpet is crowd control, this thing is antiheavy.

It requires extra gadget though, or do we have some reusable attach/detach logic around? Also for AI it will need to replace attack with attach or something.

Yes. It would need a gadget (or a very complex unitscript) and there is an engine call to attach units.

Oflameo commented 3 years ago

I like my Assault Skuttle design because it is a simple fix that gets rid of the interaction problems while allowing the unit to still serve its original purpose. After that, I hear that the whole concept of an anti-heavy bomb is bad due to lack of interaction after it fires. I suggest making it a general bomb like snitch that happens to be able to jump but not able to cloak to adjust that balance.

Arch-Shaman commented 3 years ago

What if you had the attach idea and a suicide bomb idea? Imagine a skuttle that latches onto a unit and counts down to its inevitable demise. This has the benefit of having the moment meme thing and the coolness of attaching it and the fun of a big explosion.

Oflameo commented 3 years ago

Maybe tune skuttle to be faster with a better cloak, much better vision radius, and a weaker bomb so it can function properly as a jumping cloaked scout. The lack of interaction would be an asset instead of a liability.

sprunk commented 3 years ago

That would move it closer to Widow.

GoogleFrog commented 3 years ago

That would move it closer to Widow.

My thought exactly, but maybe by this point Skuttle needs to be closer to something decent, so I wouldn't rule ideas like this on the basis of uniqueness.

sprunk commented 3 years ago

How about:

Oflameo commented 3 years ago

I am not a fan of removing things that work properly.

Oflameo commented 3 years ago

You can cut skuttle and let puppy fulfill all of the Jumpbot bombing needs without modification.

sprunk commented 3 years ago

Removing crap is the CA way but people would rage if their "surprise motherfucker" lulz unit was gone without replacement.

mojjj commented 3 years ago

i dont like where this is going...

GoogleFrog commented 3 years ago

Why? From your Skuttle reputation I'll assume because you don't want Skuttle to change. In that case you should like where it is going, because from what I can tell it is going nowhere.

mojjj commented 3 years ago

it is going nowhere.

glad to hear!

Oflameo commented 3 years ago

@mojjj You think Skuttle can't be improved at all and that it is a good unit?

mojjj commented 3 years ago

well at its current state i am having fun. this does not mean it is an unit for everyone.