Bay 12 Games Forum

Please login or register.

Login with username, password and session length
Advanced search  
Pages: 1 ... 7 8 [9] 10 11 12

Author Topic: [44.12] Musket-Mod v0.5c: Experimental Wall Breaching  (Read 41331 times)

Splint

  • Bay Watcher
  • War is a valid form of diplomacy.
    • View Profile
Re: [44.12] Musket-Mod v0.5b: Tesla Guns and Chemical Warfare
« Reply #120 on: September 16, 2018, 12:35:39 pm »

Oh god, does this mean I could end up accidentally knocking down trees and walls if I get careless? I mean, if so, that's grand, just because you proved it was doable.

zaporozhets

  • Bay Watcher
    • View Profile
Re: [44.12] Musket-Mod v0.5: Tesla Guns and Chemical Warfare
« Reply #121 on: September 16, 2018, 02:18:24 pm »

Oh, well, it's just that the head post talks specifically about scripts (which will not conflict with my mod in any way - I'm only modifying the raws), and I'm generally wary of straightaway including a huge chunk of someone else's work into my own upload, even with some changes applied.
I mean, it probably wouldn't hurt to specify that the raws are ok to use/modify too - if that's what you want. Then, some people may still be wary or just polite, but wha' can do about it :)

Ah, I see. Probably should have been more clear on that, I'll get it changed. It's nice of you to ask anyway, I'm looking forward to seeing what you get up to with them.

Oh god, does this mean I could end up accidentally knocking down trees and walls if I get careless? I mean, if so, that's grand, just because you proved it was doable.

That's what I'm hoping for. I need to get it to trigger cave-ins first, only figured out it didn't do a support check after posting the gif  :-[ (hoping a DFHack guru can help me if I don't figure it out for myself).
Accidentally flooding the testing arena was a lot of fun, but I just couldn't live with floating floors.
Need to check for material strength too (so metal walls are stronger etc).
The plan is to give the basic firearms and the rocket launcher to the gobbos to eliminate turtling as an unbeatable strategy, not sure where that will leave the elves as they don't seem very firearm-y to me.

Edit: Figured out cave-ins, but the cave-in itself is creating a small amount of magma which is bizarre. I'll keep trying. This is just going to lead to AoE2 style multi-walls, isn't it?
Not sure whether to suppress surface collapse announcements or not.
« Last Edit: September 16, 2018, 09:06:11 pm by zaporozhets »
Logged

Sver

  • Bay Watcher
  • An army marches on its oiling and waxing
    • View Profile
Re: [44.12] Musket-Mod v0.5b: Tesla Guns and Chemical Warfare
« Reply #122 on: September 17, 2018, 10:00:28 am »

Could you clarify a bit on the exact mechanics of the blunderbuss and the dragon pistol? Do the split shots retain the force and velocity of the weapon or is it uncertain?
Logged
DF Combat Reworked
No overpowered force transfer, no easy life without a kidney, more functional variety among the weapons and other improvements.

thefriendlyhacker

  • Bay Watcher
    • View Profile
Re: [44.12] Musket-Mod v0.5b: Tesla Guns and Chemical Warfare
« Reply #123 on: September 17, 2018, 11:36:24 am »

Could you clarify a bit on the exact mechanics of the blunderbuss and the dragon pistol? Do the split shots retain the force and velocity of the weapon or is it uncertain?
I took a gander at the code, and if I understand correctly the code just changes the fired projectile to a pellet, creates duplicates of the projectile a few times, makes a projectile out of each of the duplicates, copies the projectile properties over and adds a random factor to the origin and destination tiles.  As such, the velocity and material of each pellet will be identical to the fired projectile.
Logged
Fallout Equestria Redux - that's right, it's back

Sver

  • Bay Watcher
  • An army marches on its oiling and waxing
    • View Profile
Re: [44.12] Musket-Mod v0.5b: Tesla Guns and Chemical Warfare
« Reply #124 on: September 17, 2018, 12:20:57 pm »

I took a gander at the code, and if I understand correctly the code just changes the fired projectile to a pellet, creates duplicates of the projectile a few times, makes a projectile out of each of the duplicates, copies the projectile properties over and adds a random factor to the origin and destination tiles.  As such, the velocity and material of each pellet will be identical to the fired projectile.

Thanks!
Logged
DF Combat Reworked
No overpowered force transfer, no easy life without a kidney, more functional variety among the weapons and other improvements.

Grimlocke

  • Bay Watcher
  • *kobold noises*
    • View Profile
Re: [44.12] Musket-Mod v0.4i
« Reply #125 on: September 17, 2018, 09:39:54 pm »


Previously, I was plinking completed jobs from the linked list along with setting the timer to 0 in the case of firing jobs, but that did not cover cases where the job was not completed but still expired like the player canceling or the operator/building meeting some untimely demise. In that case the job would linger in the linked list and eventually return bogus values before giving a memory error. This prompted me to try and write a check to see if a job went gone on its own, but nothing I've found really works to that effect.

So, instead I'll just not use a link table and store only the job id, look for it on each iteration and drop the job id  if its not found. Least, that's the plan. My new place is short on internet connection which makes it somewhat of a pain to check general lua docs, forums, etc.

If that also fails I might just have to rip off your job checker and not store jobs anywhere at all :)

Anyhow, nice going on the other improvements and fixes! Hadn't noticed the bork in the targeting, I was mostly just testing the cannon by shooting it at my own dwarves with a firer added to the cannon projectile, which prevents any kind of friendly fire (boring, but convenient for testing. Also a temporary requirement of my projectile modding script, should have that fixed soon).

Also


... Cool!

If only the AI could bring guns and use them intelligently, could make turtling fortress a suicidal tactic.
Logged
I make Grimlocke's History & Realism Mods. Its got poleaxes, sturdy joints and bloomeries. Now compatible with DF Revised!

zaporozhets

  • Bay Watcher
    • View Profile
Re: [44.12] Musket-Mod v0.4i
« Reply #126 on: September 17, 2018, 10:45:45 pm »

Previously, I was plinking completed jobs from the linked list along with setting the timer to 0 in the case of firing jobs, but that did not cover cases where the job was not completed but still expired like the player canceling or the operator/building meeting some untimely demise. In that case the job would linger in the linked list and eventually return bogus values before giving a memory error. This prompted me to try and write a check to see if a job went gone on its own, but nothing I've found really works to that effect.

So, instead I'll just not use a link table and store only the job id, look for it on each iteration and drop the job id  if its not found. Least, that's the plan. My new place is short on internet connection which makes it somewhat of a pain to check general lua docs, forums, etc.

Those cases are the ones that were worrying me (and I wasn't even attempting my own list), but after messing about with synthesized fake jobs I'm a little more reassured (but still apprehensive) they're being deleted properly from the built-in job list given they're real jobs that should just be handled normally.
Annoying the list didn't work out for you, I don't really know what to suggest.

As a side note, when linking lua-created jobs into the world and assigning a worker, the job in unit.job.current_job didn't carry the same name as the one in the built-in list but dfhack.job.is_equal() reported them as being the same. Some strangeness going on there.

I'd be interested to know an average job list size after some (ingame) years of play, could end up getting expensive searching through.

If that also fails I might just have to rip off your job checker and not store jobs anywhere at all :)

Rip away, my dude. It was your idea!

Anyhow, nice going on the other improvements and fixes! Hadn't noticed the bork in the targeting, I was mostly just testing the cannon by shooting it at my own dwarves with a firer added to the cannon projectile, which prevents any kind of friendly fire (boring, but convenient for testing. Also a temporary requirement of my projectile modding script, should have that fixed soon).

If only the AI could bring guns and use them intelligently, could make turtling fortress a suicidal tactic.

Thanks, I hope they work and I credited you enough.
You're kidding me that the cloned projectiles don't have a firer ATM, right?

Just checked, crap. Also errors, double crap. Thanks for the heads up, I'll get on that now...

Edit: I remember I've tried adding it before and setting pelletProjectile.firer = projectile.firer just crashes. What am I doing wrong?
« Last Edit: September 17, 2018, 10:57:19 pm by zaporozhets »
Logged

Grimlocke

  • Bay Watcher
  • *kobold noises*
    • View Profile
Re: [44.12] Musket-Mod v0.5c: Experimental Wall Breaching
« Reply #127 on: September 21, 2018, 03:09:30 am »

You need to point the firer to the unit working the job first, else it'll try and copy a nothing. I thought not having a firer was intentional though, since it creates hilarious friendly fire incidents :)
Logged
I make Grimlocke's History & Realism Mods. Its got poleaxes, sturdy joints and bloomeries. Now compatible with DF Revised!

zaporozhets

  • Bay Watcher
    • View Profile
Re: [44.12] Musket-Mod v0.5c: Experimental Wall Breaching
« Reply #128 on: September 21, 2018, 03:36:05 pm »

You need to point the firer to the unit working the job first, else it'll try and copy a nothing. I thought not having a firer was intentional though, since it creates hilarious friendly fire incidents :)

Oh, I meant for the blunderbuss not the cannon. Nope, not intentional, just being typically thoughtless.  ;D

If anyone is interested in rocket launcher progress, I found a quicker way of accessing construction materials that doesn't require the linear search through all constructions (~88000 table elements in the testing arena, performing the search 9 times for a 3x3 "explosion" was too slow) by reindexing the entries into a new table using map pos as the key rather than build order making direct lookup possible.

It now checks for wall strength and produces appropriate debris:



Just got to set it up to insert new constructions after load, or the explosion will produce gems and beds and all sorts of other random stuff. Oh, and destroy buildings.
Logged

Splint

  • Bay Watcher
  • War is a valid form of diplomacy.
    • View Profile
Re: [44.12] Musket-Mod v0.5c: Experimental Wall Breaching
« Reply #129 on: September 21, 2018, 03:37:57 pm »

Those testers had to be some of the worst shots in dwarven history.

zaporozhets

  • Bay Watcher
    • View Profile
Re: [44.12] Musket-Mod v0.5c: Experimental Wall Breaching
« Reply #130 on: September 21, 2018, 04:41:25 pm »

Those testers had to be some of the worst shots in dwarven history.

Grand master dodgers and nothing else, they ended up getting swept along by the current to the lower part of the arena still fighting amongst the burning logs and steam. Great fun to watch.
Logged

Vordak

  • Bay Watcher
    • View Profile
Re: [44.12] Musket-Mod v0.5c: Experimental Wall Breaching
« Reply #131 on: September 23, 2018, 06:45:40 am »

Appreciated your new tile-destructive rocket launcher - to impressive, although I find that such weapons are not hand-carried.
You could not add something like thunderflash or dinamite, which was activated by a lever and also destroys tiles? - solely for engineering tasks).
« Last Edit: September 23, 2018, 06:47:52 am by Vordak »
Logged

Splint

  • Bay Watcher
  • War is a valid form of diplomacy.
    • View Profile
Re: [44.12] Musket-Mod v0.5c: Experimental Wall Breaching
« Reply #132 on: September 24, 2018, 05:29:51 pm »

Appreciated your new tile-destructive rocket launcher - to impressive, although I find that such weapons are not hand-carried.
You could not add something like thunderflash or dinamite, which was activated by a lever and also destroys tiles? - solely for engineering tasks).


It'd be more correct to say reusable ones aren't hand carried - stuff like bazookas and Panzerschrecks usually being closer to a crew-served weapon (I think one guy is doable, but massively inefficient when it comes to loading stuff like that.) Stuff like LAWs and Panzerfausts can be carried in bunches by individual soldiers, but are single-use. But that's a massive digression. :P

These are probably more like early Chinese rockets, albeit with a bit more bang than expected since I assume they can hurt solid stone walls.

I do like the idea of lever activated explosives though if doable. Short on miners? Have an engineer clear the space with a couple charges and some mechanisms. Need to deal with a particularly nasty critter? Lure it into a room filled with satchel charges and hit the lever. Need to burn stuff off above or below ground? Rig up a fire bomb to start a fire and away you go at the flip of a lever, suddenly free of all that tangled greenery (or even use such a thing to incinerate corpses without magma.)

thefriendlyhacker

  • Bay Watcher
    • View Profile
Re: [44.12] Musket-Mod v0.5c: Experimental Wall Breaching
« Reply #133 on: September 24, 2018, 05:36:52 pm »

So you want something along the lines of a support constructed from an explosive material that runs a script when it triggers eventful's onBuildingDestroyed?
Logged
Fallout Equestria Redux - that's right, it's back

Vordak

  • Bay Watcher
    • View Profile
Re: [44.12] Musket-Mod v0.5c: Experimental Wall Breaching
« Reply #134 on: September 24, 2018, 05:42:48 pm »

I do like the idea of lever activated explosives though if doable. Short on miners? Have an engineer clear the space with a couple charges and some mechanisms. Need to deal with a particularly nasty critter? Lure it into a room filled with satchel charges and hit the lever. Need to burn stuff off above or below ground? Rig up a fire bomb to start a fire and away you go at the flip of a lever, suddenly free of all that tangled greenery (or even use such a thing to incinerate corpses without magma.)

In general developing the idea - need three kinds of charges.
First, that work only in one z-level and low power (small radius)
Second - same as the first, but with directional (NSWE) action and cost much more materials.
Third - high power with multi-z-level effect.
Logged
Pages: 1 ... 7 8 [9] 10 11 12