SI- system Z axis

[zingo2]

Hey All!

 

A question to the Dev and builders who have more experience in Si-system!

 

 

From the picture above White block is unstable block.

 

In this build I have implemented 4 pillars(from bedrock) in a configuration 14*14(15th being the ground{pillar})

Floor height is 7.

 

This works really good on flat surface(x axis). On Z there are 4 blocks who are white-ish but they are very stable most of the time. Problem is to the right- if i build a roof/slope/ ceiling all hell brakes loose.

 

Meaning in the picture above the first floor and the third floor forms 4 white blocks in the Z axis middle- however they are stable. You can walk on them- put a block and remove it(without collapse) etc. However in the second floor no matter what trick I use- including recalculate-/ moving SI left or right(with support blocks) the blocks wont stick fully.

 

This build has a double layer roof with long ramps. And when the concrete sets and dries and when you upgrade to steel and when you place the last block- all 4 blocks fall.

 

Under the second ceiling there are support blocks (1 row) that should increase SI stability by 1 block however in this case it doesn't work. It works on 1st and 3ed floor however.

 

* Is there a formula i can use to see how Z axis is calculated in game?

 

To eliminate the problem I'll have to go 13*13 to remove those 4 blocks

 

** Is there no way that this will work on 14*14 with height of 7?

[Kosmic Kerman]

Each face of a fully supported block can hold a certain mass. A full supported block has a solid pillar all the way to bedrock. Stone/Concrete blocks can hold 120 on each face and weigh 10 each. Thus the maximum for stone is 12 stone blocks. Metal blocks can hold 320 and weigh 20 for a theoretical maximum of 16 blocks. The game maxes out at 15 or 16 unsupported blocks (I can’t remember the max as I rarely go beyond 7 unsupported blocks for redundancy). I can’t tell for sure but it seems like you are trying to build beyond those limits.

[Pille]

Maybe you can place a few Rivet Sheet Blocks at the corners and paint them to create some additional, invisible pillars.

[Cadamier]

Hey All!

 

In this build I have implemented 4 pillars(from bedrock) in a configuration 14*14(15th being the ground{pillar})

 

I'm just curious - in what you're saying above - does that mean that adjacent to each of the corners there are pillars that go to bedrock?

[Kam R.]

Each face of a fully supported block can hold a certain mass. A full supported block has a solid pillar all the way to bedrock. Stone/Concrete blocks can hold 120 on each face and weigh 10 each. Thus the maximum for stone is 12 stone blocks. Metal blocks can hold 320 and weigh 20 for a theoretical maximum of 16 blocks. The game maxes out at 15 or 16 unsupported blocks (I can’t remember the max as I rarely go beyond 7 unsupported blocks for redundancy). I can’t tell for sure but it seems like you are trying to build beyond those limits.

 

All correct.

The max cap for overhanging is 15.

 

@OP You have to count all blocks that are not supported by the definition Kosmic posted. And you always count from the nearest supported block.

Taking the way you built this (roughly), Let's say we look at the 4th row from your support pillar using stone tier blocks then you count like this:

 

 

 

You can either build your structure out of metal tier blocks to get more Max load, oooor you can use this nice block to save a lot of space and weight:

 

 

Sorry for the atrocious paint info graphic, got no other progs at hand rn, but I think it should get the point across. Hope it helps.

[The Gronk]

SI hasn't changed since A15 as far as I know so the following vid should still be valid. It answers questions you didn't even know you had about the SI system. :-)

 

[zingo2]

Hey All! Thx for all the responses Lets see:

 

@Kosmic Kerman

 

In a way you are right- I'm definitely stressing the game engine(logic) here. In this case I'm using all polished steel. 16 is theoretical possible on a single line. For example you need to reach something high above the ground- you place your rebar block on pillar and you can safely place 15 blocks straight without falling.

 

If you want really safe- you can have 10 blocks within 2 pillars. Makes base building counting and visual appeal very nice!

 

@Pille

 

Rivet blocks in this mod have SI of 12 (i think) Steel> Rivet

 

@Cadamier

 

 

Yes. This is how 7D2D engine calculates stability. It takes the weakest block from ground(pillar)- steel or whatever- and always counts diagonally for all builds in x-y-z plane. 14*14 works because the center 4 blocks whilst unstable they are "force" held in place. And with "force" I mean 2 sides are connected to solid/stable block. Thats why in game you are able to build on these 4 without collapse.

 

However this does not apply to Z axis. From my experience now i think to get Z for steel= 16-3=13

 

@ Kam R.

 

Ahh i see. I was trying to use double layer build to throw off the SI count. Sometimes it works sometimes it does not. As with double you can shift SI by +1. Add more support add more +1.

 

In the build diagonal line has 14 blocks. Double 28 and with all shifting 4 blocks in the middle are white. Its hard to see but in second and first floor I extended the supports by +2. The offset did shift SI on 1st and 3ed floor but it had no effect on 2d floor.

 

Actually ill add the cross-section below:

 

 

@ The Gronk

 

Haha cool video. Watched it ages ago. Outdated- still valid for new builders but when it comes to complex structures- a no go.

[zingo2]

So lets round it up what we got so far!

 

The idea of the very SI system is something very interesting. It needs a section to start from. From my experience from building many complex structures. If you manage to build this section and its solid the SI "bleeds" into new sections. For example try building 15*15 bridge say 20 blocks height.

 

At first you may not succeed due to randomness however once you manage to get it up- you can add to it section by section and SI system will not collapse due to the "bleed" from the solid section. Another side effect: If your structure reaches certain stress-hold and becomes huge SI system converts some parts of your structure to "ground" and starts counting from your new "ground". When you get this building is easy.

 

Now you know how those air buildings work! Its not magic but you brake the engine logic!

 

In my example of my pyramid I started from max height 250. Built the first section and moved down floor by floor.

 

 

As you can see right side is perfect. Its rock solid and even the placing/removing block on unstable block test passed with flying colors.

 

Did I mentioned i started with 15*15? This is my second build-over....

 

 

Its just amazing when all works how it comes together.

 

And then you have pure randomness:

 

 

Where 2 or 4 blocks just wont stick. Those 4 blocks are "force" held with 12 sides- more than enough to be stable yet the engine has determined- no go buddy! This indicates SI system needs some better path finding algorithm to determine whats stable and whats not- in other words- it needs an update to eliminate the random factor in all.

 

Some die hard builders might have noticed there is a discrepancy in support pillars between 1st and 2d floor.

 

 

Nope. The bottom of the 1st floor has no supports built yet. However if they need the "bleed" effect from the unbuilt floor that would explain why those 4 blocks fall on second. However given that the right side is perfect I cant say that conclusively.

 

All in all If you have any ideas whats wrong in here I'm all ears!

 

Best bet I think ill just bulldoze it all and rebuild with 1-13-1 to eliminate those blocs forever. I know 1-14-1 works- some day- but for now given the uncertainty- its not doable in a fun way.

[Pille]

Rivet blocks in this mod have SI of 12 (i think) Steel> Rivet

 

That doesn't matter because the sheet acts as a pillar but my idea is wrong anyway because there's nothing that supports the invisible pillar.^^

 

[The Gronk]

Without giving too much away to n00b hakkorz if you check out the SDX mod system there's something in there that'll let you see the exact maths behind the SI system. You may need someone to translate computer code though.

[Alphado-Jaki]

Nice building! Unbelievable less pillar structure...

 

At first you may not succeed due to randomness however once you manage to get it up- you can add to it section by section and SI system will not collapse due to the "bleed" from the solid section. Another side effect: If your structure reaches certain stress-hold and becomes huge SI system converts some parts of your structure to "ground" and starts counting from your new "ground". When you get this building is easy.

 

As far as I know, this is the most minimum structure, which system recognize as "complex". (SI for wood is 8)

[Cadamier]

 

14*14 works because the center 4 blocks whilst unstable they are "force" held in place. And with "force" I mean 2 sides are connected to solid/stable block. Thats why in game you are able to build on these 4 without collapse.

 

 

That is the heart of the problem.

 

AHH HA!! So 'we' finally, actually get to see the structure and the issues its had. I knew there were issues with the joint where the two roofs meet.

 

Yes you can force hold blocks in place (something I learned long time ago) but when it comes time to play with a structure like that "we don't" because of that instability. OH! The video does cover the issue of when blocks are "upgraded."

[zingo2]

Hey All- Final Update for this topic.

 

Had to find some motivation to do this 3x(3ed da charm?...)

 

I did bulldoze it all at ground level. Some of you might ask- was it not better just to replace pillars by 1 preserving the understructure. Mmm in real life yes- in 7D2D Nope. As we still have SI system randomness here- I had wacko mechanics happening(i did try pillar first)- once what was stable became unstable and so forth and all hell- including floating blocks. If you have to change structure fundamentals- best bet- bulldoze it all! (saves time and errors big time)

 

So I went with 13*13 (1-13-1). As 1-14-1 was rock solid as floors go no surprises here. As for double Z axis:

 

 

An interesting observation. For the 1st to 3ed floor(from top) i had to use extra support- polish steel in picture(straps). Extra support is in the middle. Also using blocks to support extra straps from underneath while it settles(as they have the unstable block in them). For the 4th floor due to the "bleed" SI I didn't require extra straps in middle anymore.

 

Until confirmed otherwise "bleed" mechanics are a thing! You have to build something in first place- make it enough sturdy to get this mechanic working.

 

 

Nothings changed as pillars go from cross-section view.

 

 

Time to check stability. While there are some white-ish blocks they are stable. Sure you get time to time(randomness) 1 block falling off but its an easy fix! Its usually the middle one.

 

 

Finally this is why its stable. Using stability glue tool(very cpu intense) you can see why I chose 2 block slope. When it all settles and becomes sturdy that 1 block is locked in place. You can put a block on it and jump on it- it wont collapse.

 

Sure you can do with wedges as @Kam R. suggested though you loose that extra durability(especially in pvp)

 

- - - - - - -

 

So Is there a dedicated Z axis? Hard to tell- as far as I'm concerned there is no Z axis! Its really all an illusion of X and Y mechanics at play. To my mind FunPimps chose this as this gets the most buildings done and SI calculations are not that demanding on your hardware.

 

However If FunPimps do implement a proper Z axis calculations in Si-Stability mechanic in future- that would allow for some spectacular shenanigans!

[Scyris]

Just remember: every block no matter how small it looks physically takes up the space of one solid block like a wood frame etc. The AI also see's it as 1 block high.