02_006_003 Catapult Issues

Catapults have many problems that make them utterly useless at present. I've been doing extensive testing, trying to make a working catapult, and have at this point given up. Here's the situation.

Starting Point
I tried making a catapult based on the one shown in the trailer. The salient points of this catapult are:

• The arm is only 4 blocks of wood between the torsion spring and the cup.
• The crossbar that stops the arm's swing is behind the vertical posts so is actually above the front part of the torsion spring, preventing the arm from going fully vertical.

None of the other construction variables (horizontal distance between rope drum and torsion spring, that the torsion spring is slightly above the rope drum, etc.) matter at all. Of these 2 variables, the 1st is essential to avoid self-destruction but makes a useless catapult, and the 2nd is no longer possible, as explained below.

PROBLEM 1: Crossbar Issues
First off, you absolutely have to have a crossbar to stop the arm's swing. If you don't have one, even a short arm will break on firing. Second, if you mount the crossbar as shown in the trailer, so it overlaps the front of the torsion spring, upon firing the arm starts vibrating like mad, eventually shaking the whole catapult apart. So, the only option is to mount the crossbar in line with the vertical posts that hold it up, just in front of the torsion spring. This allows the arm to get fully vertical on firing. See pics below for how you have to mount the crossbar.

PROBLEM 2: Pulling Arm Back Far Enough
If you can't pull the arm back far enough, the rocks roll out of the cup before you fire, often falling on and damaging the catapult. So, I built a test rig to see what variables affected the ability to pull the arm back. The test rig had 4 rope drums at different distances from the torsion spring, and I then varied the length of the arm and the position of the rope ends as well. Here's my basic test rig with an arm 4 blocks long like the trailer catapult.



No matter what rope drum I used, or whether I attached the rope to the rope drum or the arm first, the result was always the same. The catapult arm would hardly go back at all. The rear rope drum made it go very slightly farther back, but not enough to keep the rock from rolling out.



I then moved the rope end from below the cup to the back of the cup, effectively increasing the arm length by about 2 blocks. This really didn't make a difference, either.



So then I added a timber behind the arm so its end stuck out beyond the cup and put the rope end on that. This made the arm about twice its original length in terms of leverage, but still kept the cup in the same place.



This solved the problem of being able to get the cup level enough to actually load the catapult with a rock. So did replacing the original 4-block arm with arms of 6 and 8 blocks in length. No problem loading the catapult.

UNFORTUNATELY, all such catapults fail on firing. There is so much more force in the torsion spring when pulled back far enough to load the catapult that 1 of 3 things always happens when you fire, depending on the weight of the catapult. #1 and #2 are for heavy catapults with double beams and lots of weight in the back, such as my test rig here. #3 is for lighter catapults, with only 1 timber in any frame component.

1. The rock collides with the cup during the initial part of the swing, destroying both the rock and the cup before the arm hits the crossbar.
2. The arm stays intact until it hits the crossbar, but then snaps from the sudden stop. The cup and rock then collide and explode in midair a very short distance downrange.
3. If the catapult is light, then the force of the spring can move the entire catapult insrtead of breaking the arm. The entire catapult remains intact but does a front flip and goes bouncing off across the ground forever until you chase it down and delete it. The rock, meanhile goes into huge, high trajectory suitable only for indirect fire weapons. It will go over the highest spires of the demo castle from anywhere nearby.

NOTE: I also made a different test rig with the rope drums mounted inside the frame as with the trailer video catapult. The difference in height between the torsion spring and the rope drums made no difference at all in how far back you could pull the arm.



PROBLEM 3: Ineffective Catapult Ammo
Assuming you do manage to get a shot off from your catapult, whether it destroys itself or not, the rock has zero effect on castles. It just bounces off with no damage to either itself or the castle.

BOTTOM LINE
I think the underlying problem is that the torsion spring is too strong. The only way to get it back far enough to load the catapult is to use an arm of at least 6-8 blocks in length between spring and cup. But the force created by compressing the spring that much destroys either the arm or the entire catapult at the completion of the firing stroke. And then, of course, the rocks do no damage.

 

My catapults have been working just fine except for the ammo not doing damage. Definitely a bug. If you want to pull the arm all the way down you should build the drum closer and lower than you are now. Also the longer the arm the less likely it is to blow up when you shoot.

 

@Manaravak:
We're getting entirley different results (other than the rocks bouncing off harmlessly). An arm longer than 4 blocks between spring and cup either breaks on firing (if the catapult is heavy) or causes the entire catapult to flip and go bouncing off into the distance (if the catapult is light). I've also found that vertical distance between spring and drum has no effect on how far you can pull back the arm. The only thing that matters there is arm length.

 

How to make it work:

• The vibration if crossbar is where you want it can be prevented by placing it higher - this might necessiate a longer arm (which works just fine) (this also makes spoon less likely to break, just like having it at all like you said)
• You can place a few small blocks just below the spoon to prevent ammo from falling out - this also allows firing with less force even if you do manage to pull the arm back enough
• You can make your own 'spoon' using wooden blocks, this usually works better than the default one (probably weights more though)
• I heard the rope attachment order affects how far you can pull the arm. Im definitely able to pull the arm far enough if I do it right.
• You have to use the big projectiles to destroy stuff (Maybe the devs intend the small one for wooden buildings or non-buildings?)
• (Also, you can add weights to the arm to make it heavier which might help reduce the forces)

So it IS possible to make a catapult that works (of course!), but you need some tricks to do it.

What I would wish the devs would do:

• Make blocks not so bouncy to reduce the vibration if the crossbar is placed low
• Adjust the acceleration of the arm to be constant with the torsion spring to minimize breaking (It might be that way already, but it could also accelerate proportional to how fast you pulled it back, which is the worst case really)
• Make the default spoon a bit stronger, it seems to be weaker than just using wooden bars.
• Modify the default spoon so it has a small notch in the 'arm' part to prevent ammo from rolling off so easily (or just modify the collision geometry or make it deeper or whatever) - I cant see any reason why not to do this
• Add some special block that you can put on the part that impacts the crossbar and softens the impact - Ive experimented by holding the crossbar with a torsion spring so its flexible (which works), and hanging it on ropes might also work.
• Fix the rope placement order bug that prevents pulling the arm back far enough.
• Allow the torsion spring to be pulled back more than 45 degrees (90 or even full 360 - I see no reason against it and it creates more engineering possibilities. In reality it might be a bad idea but this is such a small thing that nobody should care)

 

@Waterlimon:
I'd really like to see some pics or video of a catapult that actually works. Over this weekend, I spent about 10 hours fiddling and experimenting with catapults and ultimately reached the conclusions above. These conclusions are pretty much the opposite of what you and others are saying, but I stand by them because they're repeated results for me. It's like we're playing different versions of the game (see thread title). I have SI on if that makes a difference. So please show some pics.

The direction you attach the rope makes no difference. It did in the previous version but does not in 02_006_003. Not an issue.

Yes, you can modify or replace the stock rock cup with wooden blocks. With small rocks, this does allow loading at less pull-back angle, but does no other good. During firing, the blocks just provide something else for the rock to explode on prior to hitting the crossbar.

As to arm length, there are 2 issues. A short arm such as in the trailer video is the only structurally sound option. However, you can't pull it back far enough to load a rock in it, so short-arm catapults doen't work. With a long arm, you can load the catapult but the catapult almost always fails on firing the 1st shot. If the frame is heavy, the long arm will automatically break upon hitting the crossbar, regardless of the height of the crossbar (see pic below). OTOH, if the frame is light, the entire catapult somersaults and destroys itself.



So no matter how you slice it, catapults don't work. At least without doing some strange trick I haven't managed to discover in all my trying. But even if they can be made to work with some bizarre configuration of parts, that just shows how broken they are because you certainly can't build anything that looks realistic.

 

I have had success. While at first I blamed mechanics, it was largely poor design on my part.
If you machine jumps like a rabbit when you dry fire, sadly it will make a poor siege weapon. Attaching the rope from the drum to the hook in that order works best for me.

Here are some photos showing that it can be done. The bucket only breaks if I derp on the loading and the projectile bounces around a little too much before settling.

 

i found out its better to just not place a cross bar i get more favorable results that way

 

anyway it doees work better with out the crossbar but it looks nicer with it here is a pic of my catapult, it even has its own aiming device tords the middle *the H shape thing" u build a log higher or lower depending how far u want it to go and it doesn't brake



its a stationary, it can be created using a blueprint i made with ctrl+b
and no i dident use your design i just used the multi spindles cause i was tweaking seeing if it makes a difference.

the thing i did copy from yours thoue is the way you placed the spoon not on the tip but like sitting under/ontop the timber


btw here is a link to the twitch stream video i have of it https://www.twitch.tv/evedaniel/b/625468328

 

Hi, thanks for you report we fixed soon

 

starting today i seem to be having a new issue with the catapult, last night i could keep dropping cannon balls with out them braking anything while i was trying to put them in the spoon, HOWEVER today when i tried, a small slight drop when missing the spoon destroys anything it touches

 

OK, I finally made a working catapult:



It lobs big rocks and has a gunner's platform from which the whole thing can be operated while NOT in fly mode and without moving much if at all. This gives it a rate of fire of about 1 round every 10-12 seconds, most of which is spent making sure the rock goes in the cup properly. Now all we need is a ladder that attaches to dynamic objects so you can get on the platform without flying in the 1st place .

It's heavy enough not to have much recoil so you can ride out the shots on the platform no problem. And the big rock makes the long arm accelerate slowly enough that it doesn't break. However, it does have a lifespan. It'll do 20-30 shots before something eventually breaks (assuming you haven't already screwd it up yourself already dropping a rock during loading or failing to hit J before clicking) before something fails. Usually the torsion spring explodes sending the whole arm flying. Why?

While it lasts, it's reasonably effective. It's more of a howitzer than a gun so the rocks don't hit that hard or go that far. They do best against round towers and tend to bounce off flat surfaces. However, it's quite accurate (thanks to low recoil) so you can hit the same spot several times in a row and eventually break through.



In the above pic, you can see a rock about to enter the hole already made. When I checked the damage later, there were 3 rocks in a pile at the bottom of the stairs. The rocks penetrating the outer wall did very little damage to the spiral stairs.

However, despite all this apparent success, there are still problems. Note that it doesn't have wheels. I tried for a long time to make a wheeled version but the catch blocks and axles always exploded the 1st time I fired it, sending the whole rig tumbling.

Also, this cat seems to be a special case where all the forces are in balance. It will NOT work with a small rock. Whether I try the stock cup or build my own, the light weight of the small rock lets the arm accelerate so fast it always breaks on the 1st shot. So to bulk up the projectile weight, I experimented with multiple small rocks at once in the big rock cup.



No better luck here. This thing is jsut incapable of being turned into a small rock shooter without a major overhaul, if even then. The issues I reported above with small rocks still seem to be in play. Because the arm moves so fast with a small rock, it will break or flip the cat with a long arm and be unloadable with a short arm.

Finally, I noticed something odd. After I'd been shooting a while, the big rocks started getting "english" on them. That is, I'd put them in the cup and they'd quit moving for a split second, then they'd start spinning in place at very high RPM. Shooting spinning rocks produced highly variable results. Sometimes they flew as normal but had a tendency to ricochet unpredictably even off round towers. Other times, if the rock had backspin, it was leave the cat very early and only go a very short distance at very low speed. And they also had a tendency to destroy the cat during firing.

Anybody ever seen the balls spinning like that?

 

About catapults and long arm. After yesterday update(i think cause i cannot test it before but no one metioned it yet) arm sometimes just fly throught wood that should stop it. Mostly when arm is long, crossbar is placed high and arm can accelerate quickly(like when launching small projectile or empty) no matter how thick crossbar is arm will just get throught(it can stuck inside too but rarely) without damaging itself or crossbar.

 

I do not use crossbar to stop an arm in my catapults. I use another rope attachet to arm and frame. It is more relayable and can be used to pricise and stady aiming.

 

Seen the same issue. Also had troubles when I moved my catapult (with wheels on), fired a shot and then it suddenly turned into a static object and couldn't be moved anymore. Anyone had the same issue before?