Space Engineers: Crash Testing The Physics Update (tutorials, tests and tips for Space Engineers) 

The problem may be that the blocks destroy a voxel that creates a ice chunk and that chunk glitches into the ship and destroys the internals at 21:58 you can see a chunk of ice stuck in the place the battery was.

On behalf of the ship turning with mass suspended under the ship by a piston: this is actually the correct kind of behaviour. Normaly a sapce engineers grid behaves in a way that, no matter where the center of mass (or in case of being inside a planetary gravity well the center of gravity) is, placing a gyroscope and a thruster for each direction will allow the ship to move freely in any direction, always rotating around its center of mass/gravity. However, as soon as another grid is added, either by means of subgrid via piston or a connector, both grids centers of mass/gravity are coming into effect, forming basicly a barycenter (a combined center of mass/gravity) around which the newly formed combined grid will rotate. If the added grid/subgrid is not equipped with an gyro and thrusters of its own, lever forces can and will be induced, forcing the combined grid to rotate around the combined center of mass/gravity. The fact that the craft is rotatiting after it bounces back directly after the impact is ilustrating this perfectly (and is best demonstrated by the 30m drop test). The heavy armor example just sets the combined COM lower then the light armor, creating a greater lever, whioch creates more leverage, which rotates the whole thing faster. This also explains why the example ship is capeable of correcting the rotation with light armor blocks but lacks the power to compensate with the heavy armore blocks. More Gyros would allow for a more stable flight pattern, even with heavy armor suspended underneath the craft. This behaviour is actually acurate and has been around since the introduction of docking. As soon as one grid is draging another grid outside itself, a rotation is induced due to the creation of a lever around a combined center of mass/gravity that can and in some cases will be placed outside both grids. The reason this fairly acurate behaviour is counter intuitive from the games perspective is that it is a game and to allow the player to create what ever ship he wants without forcing all players to merely use a limited set of symetric shapes, the center of mass in a particular grid is always in the center and the ship does not create levers for every thruster or gyro that is not on the exact edges of the craft (that would best be ball or brick shaped and no one likes those). To the drop tests themselves I have not looked into the complete update notes of the physics update, but it looks like we finally have something like structual integrity coming over from Medievel Engineers, now if we only could get composite blocks for SE XD Should have watched the whole video, the utter annihilation of the talisman shows that we now have something like structural integrity. It seems to have no efect on building things (since I have done that after the update without any changes) but finaly stuff gets completly destroyed instead of either phasing through the planet, awkwardly sinking into the ground or having just a couple of blocks destroyed before stopping completly and being nearly unharmed. Now if we could get composite blocks I'd be completly happy. And maybe a way to pancake ships (as a third option to nearly no damage with very slopw impacts (aka controled landings) or the ship completly being turned to vapor upon impact (I am nagging I know XD)) edit: err tl;dr great video, pistons create levers, nothing wierd there and we finaly have structural integrity yay!

Nice talisman meteor storm at the end splitsie ;) And again.. Interesting tests with surprising results. Due to the fact that wheel suspension now work on ships properly, I'll consider building ships with real landing gears like airplanes have.. Maybe even with the smallest wheel size.. That should look really cool and if it's functional as well, it would be a good alternative to the basic stiff landing gear.. Keep it up! Great video as always!

I love your videos because i don't have to read the Update logs

Well. Here comes the Fun Part again. I think the new update made the Crashes a bit lame but i kind of like it. Because in the Trailers of the Game you see the fighter getting damaged and flying into the hangar. Seeing that before the update wasnt common. It was more likely to get your Cockpit destroyed. Or more. Now you can actually return from combat. It makes the Combat better in my opinion. Raiding ships will we amazing again. You will have to return for repairs and restock. Maybe they shot your cargo container and you lose all your ammo but all your thrusters are still instact. -Aki

In the drops where the cockpit was destroyed instead of the batteries is an interesting result. Maybe you could protect vital components with blocks that will be destroyed first to prevent the force transferring into components like batteries.

Yes, interesting the collision and damage behavior. "Triforcing", look up Robocraft and see construction techniques a couple of years ago. Damage (from weapons hit) traveled 5 blocks, but with random progression of direction, or until no more blocks to damage. So the idea was to in a way control what got damaged and not. This by building so some blocks are touching, but with sides that can not attach (invalid). Since they can not attach, any damage then have to "move" through several other blocks if to reach the other, if created it so. If this thinking can be used in Space Engineers now, is something that can be tested.

Brilliant work sir - you are a true scientist

I love your attention to scientific method in this and all your testing videos (at least the ones I have seen). You put a lot of effort into controlling for a single key variable, and that's cool

And yet another good tutorial split, still the best of tutorials on SE:B XD

"SPACEBALLS?!" Oh, crap...there goes the planet...

very good job keep it up, and that outro :3

That piston that bounced :D Prepare for bouncing sticks vehicles.

looking at what the pistons did with the impacts, clearly Keen wants us to start making walkers, someone call Black Armor, his time is now!

blowing shit up for science

Thanks g, your videos are always appreaciated

Beautiful and epic series, so sad it has come to an end. You guys, all of you both at the centre of the streams and those behind the scenes are just amazing. You have brought much joy to us these past months! Thank you!

One other thing with your crumple zone testing I noticed - the ones with the long legs that survived had NO armor blocks directly below the batteries. All the other designs had impacted blocks below the batteries that failed.

I want to see how high untill the piston fails and mab see if it can be effectively implemented because they seem to be good shock absorbers

What I'm getting from this is the physics calcs are better handled... and that your ships may crash... and have subsystems (batteries, etc) damaged/destroyed, while the main hull lagely remains intact. I like this. This indicates some interesting, and survivable/salvageable crashes.

I feel like a decent alternative would be to build the batteries in a way so that they aren't nearly the first thing to touch the ground. Seeing as kinetic energy tends to travel from the point of impact throughout the entire structure, which may be why a piston can make hard impact from higher heights without damage. It has more length to travel before the energy hits the main body.

I would think the unarmed warheads would survive. I still would have included magnet plates and landing gear as part of the experiment. Also, would like to see more of the lanced ship dropping types. Maybe small armor blocks were too weak to use. Good vid.

The physics testing hooked me on the video, but that outro got me to subscribe. lol, good video keep up the hard work! I'm going to be testing the piston armor soon. Maybe it'll fair better than my pillared armor (similiar to the extended legs on your tests, but it has full armor surrounding all of the pillars).

I wonder if now that they have adjusted the impact system then will they also increase the speed limit? Here is to hoping.

Good video. I recommend explaining what block you used and then drop the ship before moving to the next one and repeat until the end, instead of explaining them all first and then drop them after, could save time and improve the videos flow.

Nice job on the video I'm certainly glad they fixed the wheels issue now ill be able to work on a design idea i have had for some time but had trouble with.

Great job there, kinda like a whole college paper about material's resistance on colisions, or something like that, rly nice.

when you droped the one with a piston and it bounce i was like ITS A POGO STICK XD

"It's not gonna happen mate. That's a whopping ball of burg." "Just try it. Hit the beat hard and together!" "I know how to hit things hard ok. And I'm telling you, the impact force of compact ice, UNDER PRESSURE IS!" "Just give it a go you big wuss!" "WHY!" Not exactly sure why this immediately came to mind when I saw splitsie dropping cars.

Some very interesting results with those tests, and some surprising ones as well. There are two vehicle types that I would most like to see you build and test, and that is a tracked vehicle, and a vtol ship. I was really impressed with that tracked vehicle shown in Keen's Physics Update video, and would really love to know how to build one.

YAY,, ;D I love when people do crash tests hehe. And wow.. the large grid ship was realy fun

It looks like the damage to the main ship is linked to how far away the impact was from them, the wheels and pistons have practically infinite distance because they are subgrids, except they well eventually hit the main ship. And the poles are another example, looks like you saw this too by the talisman

Haven't got to mess with the new physics until like 2 hours ago, and I can already tell that the physics has definitely been improved... I use a speed mod (1000m/s cap) and with this new physics, I no longer just bounce off the ground at those speeds. KSH really have made this game a lot more stable, thankfully. Hmm, During those tests, I'm surprised the connector didn't immediately get destroyed on the first one, since you know... You tapped a connector against a voxel in your survival series and broke it :) I would send armour designs in, but since my ships are all about function and I don't intend to crash my ships, all of my designs would just melt upon impact x). Oh, and speaking of the new physics and speed caps, me and a friend decided to push rotors to their limit, and we managed to break the game and hit 460m/s (550m/s peaking for me, due to server lag) with large atmospheric thrusters attached at length to a rotor (with a counterbalance). I suggest giving it a go, it's quite a bit of fun :D P.S: Those Talisman (Talismen, since plural?) crashes were so damn satisfying.

Hey Splitsie, keep doing what you're doing! I watch all your videos, very informative. Love Space Engineers :)

How about sensors activating pogo sticks when close to ground then?

14:42 "Team Rocket blasts off agaaaaain ~~ !"

i noticed that when voxel destruction occured, small objects with great momentum was spawned. Its possible what was happening is those objects were triggering collision with the batteries, blowing them up. still doesn't fully explain all of it, but something to consider. on a side note good video, you've earned a sub.

Initial point of contact takes the brunt unless it travels hitting weaker as the energy travels battery weaker than cockpit and armour (12 minutes in)

At 24:00 it seems to me that the damage done travels X amount of blocks into the structure that crashes. Which is interesting.

Was looking for a video like this

Could you try to build a giant cage around a ship and do the same tests?

13:09 "Hmmm, maybe that first drop was just a bad drop" Yeah, true, and I suppose people will unfortunately get those kinds of drops every once in a while. Some drops will be alright, little to no damage. Others, well, they might lose a good portion of their stuff, or even simply split in two like what was shown to just happen earlier.

That was a lot of fun (especially the ship storm at the end) and also very educational...Lots of things to try out when I can get back into SE...I am wondering how a ship would fly with pistons on all 6 axes....If that flies straight and level, that might give useful all round protection from impacts....also a really funny omni pogo ship :)

new impact physics are #odlysatisfying xD

14:42 team rocket flying off agaaaiiiiin

It's because it's spawning Ice INSIDE of a component.... When it deforms the voxel, it sometimes makes Ice, with no room for it to spawn except inside the test blocks, which is why your batteries are exploding.

I perceive two reasons why you get different results. 1. The ice is not smooth so you hit it always at an angle angle. 2. The physics engine is not consistent. My laurels are on the first. Cus math's math.

Im sorry to keep going on but half expected a boing sound added for the piston pogo haha!

watching, it might actually be the copositioning of the spawned Floating Ice Chunks impacting the other blocks. perhaps try a belly plate of heavy armor, a 1 block standoff, then another flat layer

Cool ending scene

Giant pogo stick of destruction. Landing gear fix to piston! Wheels fix to piston will not power.

I just put a cage over part to be protected and weird plates that stick out to absorb impact if I crash or land badly. Landing gears are good too. Thrusters are expensive. Landing gear on end of piston. For my next trick! I will make this space ship disappear!

Interesting to see how sometimes blocks break off instead of exploding.

you should dimension the platforms so you get a known impact velocity ... if its implemented correctly it should scale quadratically with velocity.

Btw, my interesting experience with enhanced physics: no speed limit mod works way better now so I crashed into planet with speed of about 4km per second. My ship didn't even break. It instantly disappeared. Surprisingly my game didn't crash, but I fell into planet. And couldn't get out without drilling. That might be a way to get into planet's core though.

the impression i get from these tests is as you suspect damage travels along connected blocks, the central battery fails because it takes impact from all contact blocks. the cockpit takes a bit of damage from just the front blocks, the rear takes the rear but the central takes from both front and rear so its twice the damage and it fails from the accumulation. i think you could get a bigger craft to survive as the impact would be spread over more blocks and not overlapping the damage leading to failure

My initial reaction to seeing the piston results was that’d be perfect for a helicopter

Man, idk. I like that they are trying to fix it. But you would think that when you slam a ship into a rock wall that it wouldn’t function. Before it would be totally destroyed. As Capitals jack said. I feel like we just went back to what it was. Nothing actually improved in a noticeable way.

So in essence, you could make retractable landing gear with pistons at high velocities.

Do a pair of pogo sticks on the talisman !!! I wanna see that drop !!

They should make more voxel damage so it actually looks really impressive and scars your map, also add the terminator theme song somewhere 😂

You can now do a back to the future delorean with this physics update

I think the piston was bouncing becuase, the force of the impact pushed it back a little then it pushed back out, basically creating a spring

So, locking landing gear cancels inertia? Then you may be able to make a nearly guaranteed crash arrestor by putting an auto-lock landing gear on an extended piston. The compression of the piston gives the landing gear more time to detect locking proximity and engage the lock. I'm going to have to do some testing of my own now.

Wonder if part armor/ part small wheels can protect those weak points. This looks interesting tho, thx 4 the update on the update :D

What about rotors, I like trying to make folding gear, could you test gear that folds (1 rotor each) flat to the bottom, and (2 rotor each) gear that fold flat to the sides Both fully extended and half extended, and with/without wheels Also do pistons in all directions cancel out the phantom rotation, compound it, or no effect

Should have tried an unextended piston to see if the results were the same? As if the length and absorbtion that is absorbing/bouncing is still there if the piston is retracted

Physics no longer extend with infinite force, but now a calculated adjustable force. In essence they are a realistic piston. so yes they can be shock absorbers

What about wheel suspension without the wheels attached. Also what i think is happening is not a transmission of force but the spawning of the damaged voxel blocks expanding into the space and effectively deleting what is around them. I have noticed the same damage by using a stone extracting conveyor system covering the port and turning it on.

i suspect that the damage is something like the mass and velocity at time of impact = impact value the point of impacted blocks each block has a impact value threshold (impact value above this level causes damage and the higher above the level means more damage) and each block has a impact dispersal value (so the impact value doesnt just run through your entire ship) so like a line of blocks flying straight into the ground creates a impact value of 10 all the blocks are the same and has a threshold of 2 and a dispersal of 1 and a health of 5 so the impact would cause 8 points of damage to the first block. destroying it and losing one point of impact value the next would be 7 damage and another destroyed block and lessened impact value 6 damage and another destroyed block and lessened impact value 5 damage and another destroyed block and lessened impact value and then 4 and the blocks would start surviving at least thats what i think the games doing as for the path the impact value takes could be something like from the block of impact to the center of gravity

Whoo!

I think that was surprise at 11:21

Thanks!

try the following thing, attach 6x wheel suspensions on each side of a Block, then let them fall from different highs,... dont now about the changes of the last Updates, but 4 week ago i used these method to check what is the best protection from falling, so the wheels from the wheels suspension are very strong, from hit from side,... so they are a good protection

You should Note the Center of Gravity, I find it looks like your batteries are at the focus of... test to see if that may be the "end" of force transmission?

From what I’ve been able to test it seems like most collision damage is caused by rocks being generated inside the ship when crashing into voxels which pretty much vaporizes everything beside them. Piston nonsense is a laughable side effect of subgrids not damaging ships. There are a lot of issues with damage still. Ships are receiving collision damage in internal places not touched by said collisions.

it's a bit odd that the crumble zone didn't get harmed even though it had a smaller surface witch should have made it more fragile.

Im going to comment at 5:56 because I am before watching the rest curious to see if a crumple zone could be a viable option for ablative(ish) armour.

So-batteries are extremely fragile. And I should be glad that I always mount them in fortified or external+non structural locations.

At the 20m test your ships tilted down from the front that’s why they got destroyed like that you may want to check your video to see

+Splitsie great tests - as ever. Had you considered using rotors as a shock absorbers? Much lower profile than pistons and from my tests at least, surprisingly good at damage reduction

Great Grand Finally.

Just realized that the background music is from Sins of A Solar Empire

Hmm this seems to explain why my buggy with a roll cage significantly less impact damage than my armored buggy

Very informative demo, thanks. As the batteries seem to have been the weak link on most occasions, I wonder if it would be less catastrophic if reactors were used?

I like this test. :)

can you please do a tutorial on mouse controlled rotors ?

Block collisions instead of the ground but with another block

21:00 Can we discuss this unusual dancing thruster?

It´s probably the voxels "exploding" into the components.

Supposedly they increased the range of laser antennas with this update as well haven't had a chance to check that out myself yet though. But i did notice about 30 fps and i have never had more than 20 before

20:38 could extended pistons be used as an emergency shock absorber

the pogo stick tilts because it's top heavy :D

I made a ship that had landing gear on the end of two extended pistons and while going back to the planet accidentally turned off the ship. What happened was that I landed flat on the landing gear and bounced about 100 meters in the air.

You didnt test the one component that always seem to survive my lawn dart incidents: Air vents are always found lying far from the crater, just fine.

You're like the DARPA of space engineers

Btw. 9:39 Nice. :P

And try piton without armoured blocks at bottom

The piston one was a pogo stick

That piston thing happens with my dropship that uses conveyor hinge mod