Discussion on strength of blocks, particularly stone and support issues (Stone supports) to the deve

I've been absolutely loving this game, but there are 2 major flaws with the game as a whole one more easily fixable than the other, but both playing a part, or working hand in hand with each other.

The first issue:

1) Strength of the stone blocks.

The strength of the stone blocks seem to be a little out of whack, and doesn't take proper symmetry into consideration, but rather treats each "Voxel" as its own block, and only considers the surrounding blocks for "weight". This creates a rather strange occurrence in-game when building, such as foundations cracking when they really shouldn't be.

The "Weight" of a structure pushes in more towards the middle and out at the base. This is why the high Gothic cathedrals in Rome and France need "flying buttresses", or stone columns on the outside of the structure to push the weight back in at an equal force, to stop the walls from buckling out, due to being thinner walls than a normal castle would have, therefore weaker.

The problem isn't the strength of the stone itself, but the way the weight and gravity is distributed down the building itself.

This directly ties in with problem 2

2) Stone support strength problems.

Everyone knows this problem.. the moment you place a support it buckles or breaks when it really shouldn't.

The problem directly correlates with problem 1, in which the Voxels are acting as a single block, instead of individual pieces.

This can be easily seen with integrity visuals on. If you build a flat wall several times up (not the block) and build a heavy piece out, so all the walls are yellow/orange, then begin to build support stones on the same voxel as the flat wall, the support beams will already be shown as Orange, when they should be shown as Green, and the walls orange.

It's already being calculated that those supports have strain on them, when in-fact they don't. As you reach the top and attach the support to the over hanging piece that's red, your supports will break instead of relieving the stress of the weight of the overhanging block, the supports act as additional stress, not the other way around.

-------------------------------------------------------------------

This is creating a problematic situation within the game. On top of the fact, that the "weight" and stress doesn't correlate with the surrounding blocks, or blocks above or below it in ways it should.

The Arch block when doubled so both end points meet, creating a C shape, should be stronger in theory than a regular stone block of the same nature in relation to "weight" above it. The Arch spreads the weight to the sides making the 2 arched blocks push in on each other, and creating a stronger, or just as strong structure as a solid piece, but lighter due to less stone.

This is why you see the Aquaducts of Rome, and even architecture of Cathedrals with big sweeping stone arches. Like this picture, rather than solid blocks. The Strength is nearly identical but much lighter.


------------------------------------------------------

Voxels themselves should be scripted to understand what "blocks" are making up that voxel in relation to each other as well. Support beams should not be considered part of that voxel directly. (Similar to how the wooden platforms and floors work), but be considered something else entirely.

One way of possibly doing this would be if they make up the voxel, their weight could be minimized and their "force" reversed, or pushing up in relation to the voxel pushing back down on it. This would create a more realistic "push/pull" effect, as the voxels above are pushing down, the "support" stone beams are pushing back up. Only if the weight above peaks past the weight being pushed up, does the support stones break.

-----------------------------------------------------

Here are a few examples of real castles, that you cannot duplicate in-game due to the above problems of 1) Stone not being nearly as strong as it should. 2) Supports not working properly, and adding to the weight/force instead of relieving it. 3) Voxels not taking into consideration the symmetry and angle (Such as my archway explanation)



I attempted to re create this one myself actually. The stone in-game being the biggest downfall to this one. The stone is simply too brittle and finnicky and wants to break before getting half of the main structure finished. Even doubling the stone blocks for every single wall won't net you enough strength to fully create this castle without the bottom half collapsing or fracturing.



This one I actually attempted to use one of the side towers in my "Athena Temple" construction I've posted on this forums and on steam, duplicating it with only slight alterations. Again, the problem being the stone is simply too brittle. While it stands, I cannot add anything to it, as even placing fully supported and green/yellow blocks next to the walls, will make it crumble.

This problem being again, that the voxels are not taking into consideration the "structure" of near by voxels when deciding the integrity.



This one is a little more quaint, and while most of it you can create, and maybe even possibly fully recreate in-game. It still has its issues and fractures that shouldn't happen throughout the structure.

I didn't re create this one, or attempt too, but I have attempted very similar designs in my Temple Athena to no success from stones which overlap each other simply... not liking it, and as access weight is pushed on top of those blocks, they seem to "pop" out and want to break. Unfortunately overlapping blocks etc. is the only way to make these cool designs, and you need to still be able to support quite a good amount of weight above it.

I'm sure there are work arounds that can make this build able with structural integrity on, but it is more problematic than it should be for a quaint castle like this.

 

Yeah, things are a bit wonky.

Stone is really good in compression but not so much with tension, sheer, and torsion. The game, however, only seems to make stone and especially mortar fail in compression rather early compared to what it lets stand with tension and sheer.

Further, the fracture mechanic of stone blocks ignores the natural weak points of mortar joints and instead has the whole mass shatter as a unit. I'm sure this has something to do with reducing the number of calculations and debris pieces, but the result is rather odd. A cubical block of mortared stones, presumably granite or some other hard rock, shatters into big shards as if it was a block of solid glass, chert, or flint. IOW, like material that Stone Age Man would make arrowheads out of, instead of something Iron Age Man would make a castle out of. More surface crumble and less shattering would be better IMHO.

There's also the way the kinetic energy of stone projectiles stops at the 1st object encountered. If that object is still physically connected to some other block, then energy in excess of that needed to destroy the object hit radiates into other blocks and might destroy them. But if the object hit is a piece of debris, not connected to anything else, then that's the end of the story, except as that piece of debris might be moved or further broken. The projectile breaks up on impact even with a barrel, chair, or fragment of a prior projectile, instead of carrying on largely unimpeded to a solid object beyond. Likewise, roundshot always break on contact with the ground instead of skipping along.

 

Excellent post. I have been thinking along a similar line.

I wonder if you have considered the fact that the first two castles that you show are really not castles, in the sense that there is any consideration given in construction with regard to defense against siege? The current physics model uses thickness, to some extent, as a considerable factor for load bearing, so a castle that does not rely upon thick walls for structure will naturally be less solid (given the game mechanics) than one that does not.

The first structure, Neuschwanstein, was built in the mid 1800's, in a Romanesque Revival style, and as such benefited from considerable developments in construction technology, both in terms of material and in knowledge of load bearing capacity? For example, the use of the four point Gothic arch, with its superior structural integrity, is used throughout the structure of Neuschwanstein, whereas the current building structures in ME, the four point Gothic arch is nowhere to be found.

Given that we now have (as of todays update) an arch block, we may be able to experiment with buttresses, but since we have only one size, stacked buttresses are not an option.

Your second structure does fall within ME's stated time frame, but it was also not designed to be a castle, as its name implies - the Chateau de Chambord. While it was built in the 1500's there was no attempt to make this structure defensible against a siege - ie. fairly thin walls.

My feeling on this is that the game is too restrictive at this time on wall types - either a ridiculously thin partition wall, or a 'cube' which would be used as a component in a curtain wall. We would need to see the inclusion of functional buttresses, functional pointed Gothic arches, functional Gothic vaulting, etc, in order to provide support to structures that used thinner (but not ridiculously thin) walls with considerable height.

I point this out only to suggest that perhaps the problem is not with they physics so much as it is with the limited construction options available?

 

well i found out that once you make a foundation (white blocks with struct integ view on) then you can make a wall 13 blocks high, or 1 stone block piller (made from stone blocks only) 14 blocks high before collapsing



i should also note that having more then 2 block thick walls will also cause it to collapse, i tried adding a row of blocks on the bottom and it collapsed, HOW EVER the new diagonal pieces does not cause it to collapse if i try adding a row of them instead.


Battlements: 13 block high wall ONLY USE THE EDGE BATTLEMENTS, platform + center battlements will cause a callapse.

 

Structural Integrity still needs work for sure. For one, strength is entirely based on weight, so every block stacks up to 14 high before breaking. Check out this video to see a demonstration:

https://www.youtube.com/watch?v=yrSxcb9rYeU

Also, as you say, the compressive strength of stone too low. They probably set it so to prevent from building too high, but that should be limited by the structural rigidity of the structure rather than the compressive strength of it's foundation. Currently, structural rigidity isn't factored at all, so as long as you don't exceed the the compressive strength of any of the blocks you can build a 1x1 tower of progressively lighter blocks as tall as you want and it won't buckle.

This is made even more apparent by the fact that a block stack will not transfer any weight laterally to other block stacks if it is sitting on a foundation. The only way to get it to offload weight to an adjacent stack is by removing it's connection to the ground and making it hang off of the adjacent stack.

This not only makes buttresses pointless, but results in the weird situations you mentioned where adding supporting structure actually increases the likelihood of something breaking. The most common example of this for me is adding stone wall blocks below an arch, which, if the arch is particularly high or bearing any appreciable weight, will result in the wall breaking.

Finally, due to the way blocks tend to damage surrounding blocks when broken, this can result in catastrophic damage. For example, if I was building a bridge supported by stone block pillars separated by arches, adding the walls below the arches will likely result in the wall breaking, which could result in the stone block next to it shattering, which could bring down the whole damn bridge.

Essentially, fortifying things tends to make them explode, which is, uh, problematic.

 

ya but as i said try building a wall, you can only build as high as 13 blocks instead of 14

 

I do know those particular two were not meant to be "fortresses", however the developers do have some nice housing wood/whitewall decorative items in place as well, and the game does allow you to build what you want, so I don't see any reason why we shouldn't be able to build something like those either. Defensible or not.

That being said, I was not aware the first one was built in mid 1800's, so I suppose that one could be discounted, however the others are within the time era/grasp of what the game is going for, and is still unable to be reproduced even remotely in the game.

I don't expect to be able to make exact replica's that work perfectly, but if you can't even build half of what you could irl due to the blocks and mortar crumbling under such little weight, that is a problem..

 

Yeah, it's weird. It's like lateral weight transfer happens sometimes, but only when it's bad.

 

this is with struct integ turned on, and is also 2 blocks thick, same wall as i showed above, except i just created a blueprint and pasted them as segments



here is the problem with building that high thoue, that JUST 1 10x Large cannon ball (i tested this again and yes just 1, held down before throwing) basically enough to create a big hole throw the wall a big portion of the wall comes crumbling down like domino almost

 

It doesn't seem to help with the fact that it's just big blocks that is used for the calculations, so if one breaks, it kinda screws things up, perhaps some kind of octree or quadtree setup would be needed, that way, is something "cracks", you have little smaller blocks to calculate, whilst I assume this would improve things, it could potentially really hurt in terms of performance.

That said, I'm curious, are the physics GPU or CPU based?

 

Excellent post!

Beside the tuning of the current stone blocks, one thing I was thinking to help with structural integrity is to add a different sizes and shapes of large single monolithic block, like a large heavy stone slab or block. If placed correctly, they can take huge amounts of lateral pressure and could be helped further with anchoring. They can also support tremendous amount of vertical pressure if properly seated and settled.

They are used all over ancient sites to make retaining walls to support the base of high structures or prevent landslides. I too thought that the stone ramps currently in the game would help me achieve a similar effect. The fact that they don't provide much support isn't that surprising, since they are at a very high angle and not that strong structurally, but they should still help a bit with spreading outward pressure.

Back the the large monolith, these would be very heavy, so you wouldn't be able to stack them very high in survival, and would require cranes to move around. They also could be "carved and quarried" by players in the shapes they need but that would probably require voxel structural integrity to work properly. They could also make for stronger columns and make stone floors and roofs easier to make. Concrete could also be part of a simpler solution here (Although storing, moving and pouring concrete could be a challenge in itself).

 

so i was thinking that beings the blocks can "disappear" into the "ground" even partially, were the blocks really the issue or was it the ground?


so i deleted an area all the way to the bottom, and then added all the way up, and they held all the way up to about 2-3 (?) cubes from the "ground level" then they broke...i think i went 20 blocks up from the "break" but they were side by side 20 up,...just like in the pic "below ground level".

idk if this has anything to do with anything, but thought you would like to see anyway.

 

i been paying attention to all 3 ram cpu and gpu, my cpu only like 5 % or so is used while on the game while 80% of my ram is used (10gb out of 12 gb) and little of my gpu is being used

 

My point was not that they should not be buildable (word?) but that we may not yet have the pieces required to build them and maintain structural integrity. But apparently this is a moot point, as thus far, based on the evidence, weight dispersing arches , butresses, etc, appear to have no practical application, which is a shame, and it seems as though the mechanics work off of a 14 block height rule of some sort. (Compression strength rather than torsion, as has been mentioned).

One way that you might be able to build your strucutures with work arounds is by using the 'magicians tower' trick, create hidden earth/stone natural blocks & hide them (surround them with other blocks0, then use them as floating foundations.

Best of luck!

 

I'd rather not resort to "dirty" tricks to build what I believe I should be able to build normally. That's my personal preference though.

That being said, it would be kinda neat to see a dev comment on these things, so we know if they have or have not been keeping tabs on these particular issues.

 

I just want to point out that arches are not stronger than the same weight of solid stone in real life. Arches simply allow openings. They do not add some magical amount of strength over solid stone.

 

While its true that arches do not have any advantage over solid masonry with regard to support of weight via compression, their advantage is that they are much much lighter than the comparable solid mass of stone that it adequately replaces.

As such , arches should be almost as strong as a solid cube of stone (which they currently are not - they fracture rather ridiculously when you they are used as substitutes for solid cubes at lower levels of a tall structure.) And, because they are so much lighter, the capacity to support weight over a much greater height, for structures that utilize them, must be accurately represented in the Structural Integrity functionality. It is the weight to compression ration that makes them , truly, magical. The magnificent arch allows for variety of construction such that if it did not function as it did, would be otherwise impossible. No cathedrals, no grand halls, nothing really immense before the invention of structural steel / etc. One could argue that the history of architecture up until the 19th century is fact the history of the arch.

Right now they have it at about 1/2 right, 1/2 wrong.

 

I got this game recently and discovered this problem straight away. It's a fairly large deal breaker for me, because at the moment it's not engineering.
I would like to point out, not only is an arch a bad substitute for a solid block, but it is actually worse than no blocks at all... As in, you can leave a gap for a "gate" or "door" that completely square and large (and there for should be weak). Say 2 by 2 large square blocks worth. And this will work fine. Then you put some arch pieces in and some supporting walls and the thing explodes taking out all the surrounding blocks. So currently, empty air is stronger than an arch.

 

Agreed, it's rather disingenuous to resort to such trickery to build what should otherwise be workable buildings. Also disingenuous to advise another to resort to it.

Did you see the post I created about Structural Integrity not being equal for symmetrical building? (All Sides Equal =/= Equal Structure Integrity?) It seems our threads have very similar findings (albeit another poster in my thread found an even bigger problem - removing blocks below certain other blocks somehow reduced the amount of force displayed in SI).

If you click the link to my thread it is quite literally in some cases that air is better than any block at all.

 

I think this discussion should be moved to either Bug Reports or Suggestions, as it's mainly intended for the devs and I don't believe they spend much time in General Discussion.

 

I've seen a similar effect in building an inn. I had two almost totally symmetrical sides of a U shaped building and along the bottom of the U one side would reliably collapse into one of the rooms, keep in mind the bottom of the U was the perfectly symmetrical part. What I've been forced to do is break up the wooden paneling and have a few sheets of the slate roof instead of wood roof to stop it from collapsing. Possibly even more strangely is removing the mostly broken block which should be nothing but dead weight actually causes the adjacent panels to break despite the fact that the extra weight should be what causes them to break and if I place those before the breaking block it's perfectly fine. Only the act of deleting the broken block causes the collapse.

 

It seems to be that fractured blocks - as opposed to completely demolished blocks, maintain the same structural integrity of the original block. They are no weaker (and if they are connected to the ground actually become indestructible) than an undamaged counterpart in terms of compression support / resistance.

 

Perhaps it should, could we get it moved?