Space Syntax

Comment

Comment by Ángel Romo Sandoval on April 16, 2014 at 5:25pm

Sorry I haven´t seen the comments last days because I am travelling.

I´ve download your file, Sam and, at first glance, congratulations!  

In a few days , when I came back I´m going to share new files (to resolve the spyderweb problem) and additional interesting thoughts.

Keep in touch!

Comment by Sam Florance on April 16, 2014 at 11:46am

Hi Angel and Pirouz,

If you are interested then feel free to take a look at the script I have been working on, I would appreciate any comments.  I believe it could be a useful tool for studying what the Space Syntax Kissing Disk Diagram does automatically, or perhaps studying the many possible equilibrium's of more complex relationships.  Thanks for your help by sharing your script's and components, I referenced them quite a bit with making this one.

Sam

Program%20diagram%20script.gh

Comment by Pirouz Nourian on April 13, 2014 at 12:47pm

Hi Ewa, 

Excuse my late response. I have seen the same problem before it has to do with an inundate Rhino. Could you please go to Rhino/Options/Check for Updates and make sure your Rhino is updated?  

Best regards,

Pirouz

Comment by Pirouz Nourian on April 13, 2014 at 12:45pm

Hi Angel, 

Thanks for sharing your model; It's a good example that shows how the Force Directed graph drawing component of our toolbox works, especially for those interested to tweak. However I could not manage to open the file because of conflicts of missing Spider Web components (perhaps you are using a version other than the last one?). 

Keep in touch!

Comment by Pirouz Nourian on April 13, 2014 at 12:25pm

Hi Sam, Thanks for your compliments. I am qouting from our paper in eCAADe 2013 on this component. 

"This tool contains our force-directed graph-drawing algorithm and makes a “kissing disk” drawing of the bubble diagram. This algorithm works by a set of attractive and repulsive forces (as in Equation 9) acting recursively on graph vertices, seeks a ‘relax’ situation for a graph, and reaches to a graph drawing. This tool is quite intuitive and shows in real-time bubble diagrams neatly according to the specified areas and the connectivity graph.

Equation 9

Attraction: 〖AF〗_ij=k_a ∆x_ij  for all linked (i,j) 

Repulsion: 〖RF〗_ij=k_r /x_ij   for all (i,j)

The attraction/repulsion strength inputs are denoted as k_{a and kr} 

in the above equations. If some configuration is very messy, you need to have a high repulsion first to untangle it. I have not tried Angel's method but it is very similar to the method we have scripted for this component. 

I hope this helps. 

Best regards, 

Pirouz

Comment by Sam Florance on April 12, 2014 at 6:19pm

Hi Pirouz,

Excellent set of components.  I am wondering about the attraction strength and the repulsion strength for the Force Directed Drawing component.  What do the input values dictate?  I have been inputting 13 values with many connections and I notice that if I play with the repulsion strength the space organizes a little better.  Could you tell me what these to variables mean?

Thank you for your time,

Sam Florance

Comment by Sam Florance on April 10, 2014 at 3:44pm

Excellent, thanks a lot Angel.  I will be sure to share any results I have with you.  I am not very good with grasshopper yet, but I do have the advantage of having a some time to work on it.

Comment by Ángel Romo Sandoval on April 10, 2014 at 3:30pm

Hi Pirouz, hi Sam,

I admit that when I uploaded the algorithm I though that it works correctly but later I discovered a problem with bigger diagrams. Since that day I tried to fix it but it is taking me too much time, so here is my file, if Sam o somebody wants to work over it and can improve it, welcome! I am learning python but it will probably take me a lot of time because I have no programming knowledge.

2-It would be better that the stiffness (kangaroo component) of the lines that connect no-neigbourds nodes changed in function of the distance between this nodes. It could be solve with a loop. Until now, I fix this problem applying a very high stiffness in the repulsion kangaroo component during the simulation and after I reduce it slowly.

In the file there is an explanation more detailed.

floorplanator.3dm

floorplanator.gh

Comment by Pirouz Nourian on March 29, 2014 at 6:54am

Hi Angel, Thanks for your compliments on our work. Glad to see you find this work inspiring. Very interesting paper, I need to check their algorithm. The problem of plan layout can be formulated in many ways. In any case, the geometric possibilities for a plan corresponding to a graph are endless, unless you target a specif class of plans, e.g. a plan composed of rectangular cells. In that case, it is possible to enumerate possibilities. The force-directed graph drawing algorithms, like the one we have implemented and the one you have made are not guaranteed to converge and their results are not unique in the sense that there might be many local optima/states of equilibrium that they can reach to. However, they are quite useful in many practical applications. Voronoi diagrams can also be a class of geometric "dual" cell configurations that can satisfy the connectivity graphs. I will post a small definition for using them shortly.    

Comment by Pirouz Nourian on March 29, 2014 at 6:43am

Hi all, my apologies for losing track of the posts here. Re Frano's post; we are hoping to release the new version in May. We will do our best to add the option for Space Syntax calculations with various radii (R3, Rn). Sorry I just deleted a post by mistake...

• ‹ Previous
• 1
• …
• 33
• 34
• 35
• 36
• 37
• …
• 40
• Next ›
• Page