I'm repairing the repair bot - need feedback

[klauss]

Hey!

I'm repairing the repair bot.
I initially just wanted to fix the timing, since it was taking far too long to work. But then I noted it didn't fix most of the important things: jump drive, thrusters, shields, etc...

So, a list of what I think should be fixed by the repair bot, and what not:

YES:

* Life support
* Comm system
* Radar
* Hud
* Fire control
* SPEC (for VS)
* Jump drive
* Thrusters
* Engine
* Reactor
* Shields
* Cloak (perhaps not - Would it know how?)
* Computer (I'm assuming at some point in the future, there will be computerized tasks, like automated landing)

NOT:
(disclaimer: perhaps some models could)
* Armor
* Hull
* Damaged beyond quick repair (Upgrades/Damaged)
* Damaged/destroyed cargo items

And, since I'm at it:

Damage model changes:
* Mission cargo should not be damaged
* Jump drive should never be damaged beyond repair (gameplay-wise, it sucks, but perhaps it's a nice realism issue: you did survive, but you're stranded if you can't find somewhere to buy a new one).
* Ammo damage should case ammo explosion, hence more internal damage and perhaps a chain reaction with more ammo.
* Damaged guns could "fail catastrophically", resulting in destroyed guns, if triggered.


Comments... suggestions... welcome...

[spiritplumber]

I would say that the cloaking device cannot be autorepaired... past that, works for me ^-^

re cloak: WCU will NOT have a modular cloaking device, the Strakha will be in the game and maybe the player will be able to get it, but keep in mind that the Strakha is basically built AROUND the cloaking device.

http://www.spiritplumber.com/priv_remake/ Grab teh update.

[chuck_starchaser]

Last time we discussed repair bots someone suggested something very interesting: the role of spare parts. There'd be repair kits you can buy or "refill" so, while some things could be fixed with a hammer or a screwdriver, some would require replacement. The best part is, one could rummage ship wreckages for spare parts.

Another interesting concept would be to have internal and external bots. External bots take up 1 cu.m in the cargo hold, and you need peace and quiet for them to work. They'd repair sensors, weapons, engines, shields, jets, and bay launchable weapons, such as mines and decoys; but not armor.
The internal bot would repair reactor, jump drive, computers, screens, as well as internally mounted weapons, such as railguns.

[MamiyaOtaru]

NOT:
(disclaimer: perhaps some models could)
* Armor
* Hull
* Damaged beyond quick repair (Upgrades/Damaged)

For privateer, someone was mentioning that in the original, the advanced bot could repair items that had gone down to grey, and that was a difference between it and the regular bot. (I do need to confirm that though)

Armor, hull and guns should never be repaired by the bot.

Might be different for other mods, but as long as the possibility/flexibility is there to configure it that way for Privateer, I'm cool.

[klauss]

Ok, I was trying to avoid having many types of droids, since that would require a change in Units.csv, and hence would invalidate all save games.

If there's a way for that change not to invalidate save games, that would be interesting - let me know. I was reluctant to do different bot types just because of that matter, but I think I'll have to live with it - it's just a feature that must be there.

About flexibility, I'll try to make almost everything configurable either through units.csv or through vegastrike.config (I prefer units.csv - so that different ships/bots can have different characteristics).

[spiritplumber]

We will eventually do save-invalidating stuff, so we might as well do so now, I think I personally clear my save directory every day or so anyway... I still haven't played the Destinee campaign that's kinda my reward for finishing my portion of WCU.

[chuck_starchaser]

Shouldn't have to invalidate save games. The external bot would just be a new unit, I would think; and some of the things the existing (internal) bot could do before would be denied, such as fixing maneuvering jets. Then again, we could continue to have one type of bot that does it all, but I think it's a bit inellegant, in the sense that to fix things outside it has to go through an air-lock. Also, the internal bot would probably use magnets to keep itself with the wheels on the floor, and might be able to operate during combat. This is a specialized propulsion system for interiors. Whereas the external bot has to move around using jets, so it would be lightweight and specialized for free space mobility, and would probably get lost if we accelerated away. Finally, having 2 bot would be the preferred solution, so that the two can work simultaneously and get the job done sooner.

[Zeog]

The external bot could walk with some "magnetic shoes" on the outside of the ship as well. Give it some 6 or 8 legs and it looks like a spider roaming the outside of the hull desperately trying to find something he could just fix a little. It might as well have some armor on its back giving it some protection from spacefight as the looks of a little beatle

[chuck_starchaser]

Sounds pretty cool, but it would be a lot of work to model and animate if we decide to make the bot visible from the cockpit I think the most logical design would be one that uses compressed air for mobility, but deploys telescopic (antenna-like) legs that lock onto little holes on the armor plates around the usual places of work, so that it can exercise some force when necessary, without pushing itself away.

[mat_yarrow]

Mention of the repair droids crawling outside a ship's hull brings to mind the ones seen in Homeworld:Cataclysm. The problem would be in the implementation .

[dandandaman]

That list of fixable things seems awfully inclusive Unless you mean fix by "slightly improve the state of" ... otherwise it seems as if damage doesn't really matter any more, just take 5 and come back to a clean ship? At the very least efficiencies of damaged components (reactor, shield and engines especially) should not be repairable to 100% if the damage rating exceeds 15% or something... choosing to not make the reactor repairable might make things interesting too (when we get power management at least)

Dan.a

[klauss]

That's something I was considering. Doing everything now, that I got to it: the degradation stuff, and refactoring the damage model, since as it is right now it's awfully difficult to decide whether something is damaged or not, or what its full-functionality state would be. In fact, although refactoring would take a while (I wanted to release the repair bot fix quickly), it would solve many bugs (those dealing with unrepairable items, and so on).

So, perhaps, now that I'm touching that code, I could take the chance to do everything at once.

[chuck_starchaser]

I agree; best thing is for the core code to include all conceivable repairs, then cut back to taste when using it; --rather than have hard-coded limits. I'd assume different models of bots would have different capabilites, so the customization parameters have to be there, anyways.