Fractureme 0.4 and heaper 0.2

Wow, very nice.

I ran out of adjectives to describe these scripts …so just want to say thnx for sharing it with us…

good script but i would recomend renaming the crackme to shatterme, glassbreak, or something along those lines. if you google crack anything it is going to return unexpected results relating to piracy mostly. it is the new hack word, which for older people meant comming up with a work around for a bug, not getting into other peoples pcs, like what this forum is for.

So good it belongs in the Louvre… ok maybe not, but I like it.

!!! great work, thanks for sharing these scripts!!!

@foom: That’s the point, when I press P the script gets the error and does not work!

It was my build of Blender. I was only running 2.48. After installing 2.48a, the crack me works. Now I have crack!

Guys, thanks alot for all the happy comments. I’m happy you’re happy!

I’m sorry if anyone is having trouble with running the scripts, as I said it’s the first experimental release without user-friendliness, the order of priorities is for me now algos -> usability -> gui ->help & doc…I was just too happy with the results I got so that I had to show my progress to the community.

I’m still fighting with the polygonisation of the shards in the crackme script, sometimes normals are not correct and complicated shards do not fill well, you have to cerrect these manually. I’m also working on release of shards during simulation by other objects, like projectiles etc.

The name for crackme - I’m not natively english, but I thought cracks are what appears if stuff breaks? if not correct me, the name isn’t problem. Also I hope to make the ui for these scripts when 2.5 is out, now I want to focus on functionality, optimising meshing and simulation results. You may have noticed that crack generator gets very slow if you make the nodes value high and sometimes it even freezes(bug).

Atom - there are 2 scripts open if you open the crackme file. 1 is for setting up cracked stuff - it’s the cracker_0x script, you run it to generate a set of shards. the other script - cracker_rt is just a script of the bomb object, place the bomb near the wall and run the game engine, that’s the simplest explanation. This 2nd script doesn’t run by alt-p in script window, it just cares for the releasing and shooting of shards in game engine.

Basse, I figured out the file heap01 had ipo recording turned on, which is my mistake, it isn’t needed at all for the script, since it just makes new instances without ipo.I developed crackme in the same file and forgot to turn it off for the release. Without recording on, the script is also much much much faster. Also, if you have the objects joining working somehow automatically, don’t hesistate to post your solution please :slight_smile:

I uploaded an updated file which has this just off. object shapes also have by default convex hull by now.

looks like two very useful scripts, the results are very convincing. i only hope for some more comments at the var settings, but i like it so far.

how about using own meshes with crackme instead of just plain walls like it’s possible with maya shatter?

keep it up. :slight_smile:

updated the files and edited the first post. will be adding release updates there not to make confusion about what’s currently in, here is just discussion.
new little test here, just another boring explosion:

http://plant.ffa.vutbr.cz/~novak/dwnflz/crackme/explosion05.avi

notice the nice cracks which end in the middle of shard? those were the problem, now mostly solved…

Here are 3 presets for fun:

[CODE#wall

increasing_branching=1#if the branchingprobability changes during time
startbranchingprob=0.55
endbranchingprob=0.55
gmaxdist=0.3
d3d=0
branching=2
startpoints = 3
oneafteranother=1
startcracks=15
increasing_curvature=1
#curvature=20.0
startcurvature=55.0
maxcurvature=55.0
branchcurvature=35.0#additional!
nodes=5500
bordernodes=sqrt(nodes)*3
crackdiameter=14.5
randvorder=0#2
distribution=‘random’#‘pointbased’ #distribution of initial nodes
displace=0.00#small displacement of shard borders
cleanupnodes=0
generate_cracks=0
generate_shards=1
sectorsresolution=2
massmultiplier=1.0
blindcracks=0

sizex=4.0
sizey=4.0
sizez=0.2

#detailed glass

#animbranc
increasing_branching=1#if the branchingprobability changes during time
startbranchingprob=0.5
endbranchingprob=-0.2
gmaxdist=0.2
d3d=0
branching=2
startpoints = 1
oneafteranother=1
startcracks=12
increasing_curvature=1
#curvature=20.0
startcurvature=10.0
maxcurvature=35.0
branchcurvature=20.0#additional!
nodes=4500
crackdiameter=4.5
randvorder=2#2
distribution=‘pointbased’#‘pointbased’ #distribution of initial nodes
displace=0.00#small displacement of shard borders
cleanupnodes=0
generate_cracks=1
generate_shards=1
sectorsresolution=2
massmultiplier=1.0
blindcracks=0

sizex=4.0
sizey=4.0
sizez=0.02

#wall with 3 smaller holes impacting one after another

ncreasing_branching=1#if the branchingprobability changes during time
startbranchingprob=0.5#branching cares for the probablility a crack splits
endbranchingprob=-0.1
gmaxdist=0.2#maximal edgle length in the graph
d3d=0
branching=2#how many branches can happen each branching max
startpoints = 3#how many impacts hit the body
oneafteranother=1# if the cracks of the impacts propagate simultaneously or one after another
startcracks=3# starting cracks in each impact point
increasing_curvature=1# if curvature will change through time
startcurvature=35.0# curvature specifies maximal angle of crack edges, low values are e.g. for bulletholes in glass, higher for granular structers
maxcurvature=35.0
branchcurvature=20.0#additional curvature possibility on branch split
nodes=2500# number of nodes in the graph, don’t go too high, or crashes become a sure thing.
crackdiameter=1.5#diameter of crackc if the distribution is pointbased
randvorder=0# order of cracks distribution, the higher the more small/big shard difference from center to outer edges
distribution=‘pointbased’#‘pointbased’ #distribution of initial nodes
displace=0.00#small displacement of shard borders
cleanupnodes=0#debug option, clean up the crackmesh from node vertices
generate_cracks=1#generates a crack mesh which grows in time, e.g. for inner glass cracking without flying shards
generate_shards=1#generate shards for simulation - takes the most time in the script
sectorsresolution=2#not used yet, otimisation structure
massmultiplier=1.0#multiplier for all masses of the shards, not much usefull yet.

#size of object - if you select an object before running the script(a lying box with the thinnest side being z), it adjust sizes automatically to it, otherwise uses these
sizex=4.0
sizey=4.0
sizez=0.2]

thanks for commenting everything, but for some reason crackme is now crashing for me.

getting size from mesh
scene.link(ob) deprecated!
        use scene.objects.link(ob) instead
building border nodes
seeding weak spots in the matter
connecting borders
ohnoh
Traceback (most recent call last):
  File "crackler", line 356, in <module>
IndexError: list index out of range

Same problem here as Kai mentioned…

can someone explain a little
i downloaded the crackme 01

then what do you do with this ?

just P Key

and i see a lot of number on the upper left corner

then wath ?

Thanks for any help

Update, see first post.
new videos are here:
http://plant.ffa.vutbr.cz/~novak/dwnflz/crackme/crush.avi - 4000 objects! from image
http://plant.ffa.vutbr.cz/~novak/dwnflz/crackme/crush01.avi -projectile based release without chain reaction
http://plant.ffa.vutbr.cz/~novak/dwnflz/crackme/crush02.avi - projectile based WITH chain reaction

Kaik, sorry, that’s a bug which strangely occurs only sometimes and I didn’t solve it yet. Try to run it again, sometimes helps.Also with images this isn’t an issue, it’s just for cracks generated from the algo.

Pildanovak,
could you upload/host your scripts somewhere else too, like www.4shared.com … for some reason I can’t connect to the server in Czech Republic (I always get “The server at plant.ffa.vutbr.cz is taking too long to respond.”) … thanks, a

studioa - where are you from? it’s really strange, because I upload on our school network which has direct connection to the backbone of internet in cz. I’m sorry, but I dislike this type of upload servers, but anybody who wants can take the files and spread anywhere can do. They’re opensource of course.

pildanovak,
you can use this site for direct no frills uploads/downloads.


great work again btw.
finally I found this link of a similar process.

meta-androcto - heh, only jpegs allowed :slight_smile: it doesn’t want a zipped blend either… nevermind, some other time…

Mediafire allows for fast upload and a free account allows for uploads of up to 100 megs, you can even upload without an account. You have unlimited space and bandwidth as well.

If you get the paid account you can upload files over 2 gigs in size.

@ this moment I’m in Ohio, US. - would be great if people in charge with the scripts could upload them under http://wiki.blender.org/index.php/Scripts/

studioa,
the wiki link you provided is not the page to upload scripts.
http://wiki.blender.org/index.php/Scripts/Catalog
is the repository for external scripts.
& atm as you have to press P to run the script,
it would belong in the Games Section for scripts.
With Scripts such as this, generally I wait until it is stable, before listing.
However, I will speak to pildanovak later on irc to discuss how best to handle this script.
thanks.

pildanovak, i’ll contact you later, it may be best to move this script directly to blenderprojects.org.