Hi Yalda, you can use it like the grasshopper's Voronoi diagram component with a list of points and a boundary curve plus a list of areas to be used as the weights. You can check the demo file for an example. The demo file comes in the installed folder.

Thanks for explanation but I didnt have any result because of an error on radii data.As Im beginner user, I still dont know is there the logical relation between radii and boundry curve or just a restrictive...

Hi Yalda, please make sure the units of the Rhino file and the radii values are the same (e.g. meters). If you aim for large radii when the boundary is small you'll get errors because the boundary cannot fit circles that big. Please start from the example file. If something went wrong again, please attach your file to a comment here and I'll look at it. 

Hi, 

I am very new in GH. I want to use ForceDirectedGraph lines to set proximity relations with points. Because I want to use areas as constrains for point proximity in a plan organization. However, I cannot get those r values from graph. Thanks in advance.

relations.jpg

@ h 

Hi, you can get the radii of circles by making an expression like Sqrt(A/pi), where pi equals 3.1415 and A is the area of the corresponding circle. this is because the area of each circle is pi*r^2 (pi times r squared)

Yes, I know that but ForceDirectedGraph component only gives center of the circles within the graph. I wonder that it could give the areas of the circles in tangent position. I think I should calculate myself, as you told. Thanks. 

Hi Everyone!

I'm new in Space Syntax and now I have problem when trying to make the triangulation. I follow the example and also read the pdf but couldn't really find the solution.here is the screenshot:

ss.PNG

sss.PNG

anyone knows what is the problem?

thank u:)

Hi Asal, 

Could you please post the content of the error message and an image of the network that is to be triangulated? It could be that there is no possible triangulation or that the number of triangulations is smaller than 9 (9=8-1). 

BTW, I'd suggest you look at the chapters 3 & 4 of my dissertation for an overview of the processes supported by the toolkit. http://abe.tudelft.nl/index.php/faculty-architecture/article/view/n...

Hi pirouz

tnx for your quick response.

the error is:Object refrence not set to an instance of an object.

1.PNG

2.PNG

and I will read those chapters.tnx

Hi Asal, You are welcome. In the image you have posted I see that you have not provided any connections with the NEWS sides. Those links indicate your preferences for having access to one of these sides because of lighting or other reasons. Without these links it is not possible to make a triangulation. For making these links you need to draw additional lines towards the 4 sides to indicate adjacency. Note that in the end, you need to have enough links so that the graph is at least 3-connected. That means that the graph should not be torn apart by removing 3-1=2 vertices. For technical details see: http://mathworld.wolfram.com/k-ConnectedGraph.html

https://en.wikipedia.org/wiki/K-vertex-connected_graph

Please note that this is a theoretical limitation, not a bug. I hope this helps. 

Hey
Good job! 

Was wondering, is there any way of baking the results of the analysis? 
Saw a related topic before, regarding the baking process. Look through the .gh components, yet didn't find anything related to the baking.  

I do see a workaround, where You just screenshot the image/part of the graph and edit it afterwards. Anyway, was just wondering if there is a way straight  bake it, instead of using the workaround. 

// 
I find this method of spatial analysis awesome, since it lets us create a solid layer of information. The analysis supports a branch of ideas and works side-by-side with the designers own intuition. 

Again, a big thanks :) 

Hi all, as I find the question of  Tanya relevant for all, I post my answer here:

I think this can make a good didactic discussion for our users. 

My short answer is that Space Syntax or any other theory based on computational models cannot predict the behaviour of individual people. Humans are creative and their actions cannot be explained solely based on cause-effect (stimulus-response) mechanisms. In other words, people cannot be reduced to ‘causal systems’. That said, social sciences can be helpful in studying the likelihood of social behaviours. In this particular case, the underlying assumption of Space Syntax is that any environment (the network of spaces in a building or in a city) has an effect on human behaviour in terms of such things as mobility and social interactions. Space Syntax is an umbrella term for a range of mathematical constructs that seek to relate the structural (graph theoretical) properties of space to human mobility and social interactions. The effect of spatial configuration on human (social/average individual) behaviour cannot be deterministic. In any field studying human behaviour one needs to consider many factors that can nudge human behaviour in certain directions. Anyway, long story short, Space Syntax models often use the so-called centrality models to study the graphical (network) structure of space and check the ‘potential’ of a spatial configuration for conducting human movement flows or finding gradients of privacy-community, chances of social encounter etc.

I would prefer to separate two things:

• Designing a spatial configuration based on an abstract network of spatial relations (see e.g. chapter 3 & 4 in my PhD:  http://abe.tudelft.nl/index.php/faculty-architecture/issue/view/474 and http://www.sss9.or.kr/paperpdf/mmd/sss9_2013_ref048_p.pdf)
• Analysing the network of spatial relations (while designing a building or for an existing building)

As for the computational design process based on a network construct, I am working on an alternative method for producing spatial layouts that is based on a voxel-grid of space (still a work-in-progress)

Please feel free to share your insights and opinions. 

C_Disk_PZN_Corrected.gh

hi Andreas Krigoltoi; thanks for your kind words; I am happy to hear the tools are useful. Re baking (instantiating) visualization results, my new approach is to get Mesh outputs from diagram components. That is my intention but work pressure makes it difficult to realize it... There is one component I have posted before here that puts out meshes for bubble diagrams. The eventual solution (which I am working on these days) is to open source the analytic algorithms and let people develop their own visualization tools on top of them. Keep in touch. 

Hello Seghier, I would suggest using the names instead of icons (this is a GH option). This is a standard parameter container from GH (not from our plugin). You can (most probably) find it in the first menu Parameters.  

then it is an extracted "number" parameter containing real numbers (e.g. 2.34, 10.0, etc.). Use a "number" parameter instead. 

it could also be a "text" parameter container.

send a simple version of your file please. I'm not sure I understand what you are trying to do.  

Hi Pirooz

Happy New Year

This Picture attached, refer to an article, published in the Journal of Space Syntax

This article refers to software. that is as far as I know is currently not available

?Did you know that & time of publication the software

I am grateful that I guide

Hi Maryam; Happy Nowruz

I think it is done in GH by Dr. Tasos Varoudis @UCL. That is all I know. I don't know if their app is available. If you want to do visibility studies I'd suggest using SpiderWeb. 

I e-mailed to Dr. Tasos Varoudis and Dr answer to me:

This examples were done with a custom software, not available in public.

You need to use your own scripting to do something like this at the moment.

I think it is an upgraded version of the Depthmap software. Did you notice the two-dimensional to three-dimensional version upgrade it?

Hello Maryam,

I don't think it is a depthmap upgrad. If you look up the paper and take a look at figure 4 / 5 you see the standard Rhino3d Isocurve display of cubes. For some reasons people do not want to say, it was done with grasshopper....

If your up recreating 3D IsoVist analysis I suggest to find a good 3d FOV algorithm - there are some around in GH forum. Once you have determined what voxels are visible form a  voxels you can use spiderweb to run BFS and than compute what kind of integration Value you would like to compute (RA,RAA,I).

However these things get very pricy (computational wise) as visualGraphs are always very dens.

I was thinking of implementing BFS on GPU at one point, than things should be a bit faster...

Best

Richard

Hi Richard, thanks for the hints. It's funny why one sometimes feel uncomfortable mentioning GH :)). Could you please point to some of these Field-of-View algorithms implemented in GH? Have you ever tried using this one: http://www.cse.chalmers.se/edu/year/2015/course/TDA361/grid.pdf

Thanks Rich & Pirooz

I'll try my best
Hope to succeed

Hi Pirouz,

I'm having some trouble producing a functional Disk Graph Drawing that uses the NGV output from the Force Directed Drawing component. It appears that the repulsion & attraction forces are not performing correctly. I continue to get an error message from the Disk Graph Drawing component that reads:1. Solution exception:Object reference not set to an instance of an object. See posted files Norwst_02.ghNorwest_02.3dm

Hi Matthew, just tried to open your Rhino file, but it is not saved properly as a Rhino file. Can't open it. Anyhow, I am familiar with this question. The thing is, a coin graph like the one you expect from a Force-Directed Graph Drawing algorithm might not be possible to draw given the list of areas, imagine fitting a 2 Euros coin in the middle of four 5 cents coins. It does not fit. A force-directed drawing algorithm is a heuristic algorithm; meaning it is not guaranteed to work; but it often does. It is theoretically not guaranteed to converge to a solution. If you would like to see something sensible you can lower attraction and repulsion to get a better picture of the situation. You can also use the 'Spectral Graph Drawing' tools of Cheetah (only the new version available on the forum) or Spider Web for getting a picture of your spatial network.  

more info on the force directed algorithm and the whole process:

http://abe.tudelft.nl/index.php/faculty-architecture/issue/view/474

Hi Pirouz,

really appreciate your help. I am currently installing the new version of cheetah and will have a look into the Spectral Graph Drawing tool. I have read through a couple of the chapters in your phd and now have a better understanding of the Force -Directed Drawings and why I was not achieving a good result. The list of areas, spaces and connections I was using for input were a bit complicated. I will simplify and have another go. Love your work : )  

Hi Matthew, thanks for your kind words! happy to hear you find the tools useful. Perhaps you can follow the instructions/pseudocode in my PhD and make a 3D version of this algorithm with the help of Kangaroo. I can then translate it to C# and include it. 

Hi Pirouz,

I am trying to install syntactics on my desktop but can't seem to get it to work. I've tried using the installer and copying the gha file to the libraries folder, made sure it isn't blocked or anything, but still it never shows up after i open grasshopper. I've tried this several times, running as administrator and not, but still it never works. Any idea you have as to what might be causing this?

Thanks!

Hi Romero, sometimes you have to restart your machine to get to see the new plugin running in your GH; I guess you have done this already though... Have you checked if your Rhinoceros is up to date? Rhino/tools/options/update

Hi Pirouz,

I am working for construction company  under design department, i would like to ask if its allowed to install and use your plugin together with galapagos to study space planning ideas? e.g. study  possibility of building core planning iterations. Thank you.

Hi Karlo, Thanks for your message. Yes you are free to use the tools for commercial purposes. The only condition is to mention the software in your publications or cite our papers and kindly post some results here on the forum. 

BTW, if you wish to have additional tools currently missing in the toolkit you can send me a mail p.nourian@tudelft.nl 

Thank you very much for your permission. I will definitely indicate your software and credits to you once i prepare a presentation to top management on the use of this plugin.  will post on this forum any progress i have made. Thank you again if you allow me to add additional tools. 

The first 3D molecule of Space Syntax and the beginning of a new season ;)

141 iterations in 11 milliseconds

Stay tuned...

Hi Pirouz,

Thank you for your SYNTACTIC tool, I am using it to make the axial analysis. And now I have a problem. If I have a city plan or building plan to make the axial analysis for it, I want to know how to use grasshopper to generate the axis graph first?

The story and the [3D] future of SYNTACTIC: Space Syntax 4 Generative Design

https://www.slideshare.net/secret/yhBdV0QwFG0GsG

Hi Pirouz,

The presentation is indeed private (no access)! Not sure if that was intended :)

Regards,

Theodore.

Hi Pirouz,

Except the difference factor, control analysis, choice analysis, integration analysis, entropy analysis, I want know whether the plugin can calculate the value of connectivity and intelligibility.

Best regards,

Sz_tju

Hi Sz_tju, for calculating connectivity you can connect a mass addition component to the graph (whether from axial lines/cells/nodes). Connectivity is the number of neighbors of a node in a graph and in graph theory it is called the degree of a node. For intelligibility, you can calculate the correlation between node degrees (connectivity values) and the global integration, see: http://otp.spacesyntax.net/term/axial-intelligibility/. At the moment you can do this with a bit of math expressions/scripting. I will add a piece of code for that.

Best regards, 

Pirouz

Hi Theodoros Galanos, thanks for pointing out! It was a mistake indeed. Meant to be public :)

Cheers, 

Pirouz

@ Sz_tju,

Attached, you can find two components which calculate the intelligibility and connectivity.

Connectivity%26Intelligibility.gh

@Samane 

Thank you very much for your components.

@ Sz_tju,

You are welcome!

Sama

Hi Pirouz!

Just wanted to say thanks for the plugin, It really is useful to have such a tool in gH. I've been following your dissertation as I am also highly interested in the development of SpaceSyntax Methods in Architecture. I'm actually also in the process of maybe coming up with a workflow to have the initial parameters of room labels, connections and areas within a separate GUI, or say even excel. I'm using snippets of your example and I think I have run into an issue in the TutteConvexDrawing Component not translating/referencing properly to the TriangulateMeshinAllWays Component. I can't seem to understand where the problem may lie. I have attached the test files below. 

Hope to hear from you soon :)

Cheers, J

BriefAnalysisTool_gH_PreAlpha.gh

Test-Sheet_AreaInput.csv

Hi Jonathan, Thanks for your kind words about the toolkit. I am happy to hear you find them of value. I have a high education workload at the moment so I have not checked your file yet. But perhaps these explanations help (we will also look at your file) asap.

About your question regarding the Tutte graph drawing algorithm (also known as topological embedding):

The Tutte algorithm can be viewed as a special case of Spectral Graph Drawing, which is a mathematical solution for topological embedding formulated as an optimization problem. The formulation of the topological embedding (e.g. as in Tutte algorithm) is in fact quite similar to the so-called force-directed drawing that is often solved by heuristic methods like the one we have made for the SYNTACTIC plugin. You can read more about Force-Directed Graph Drawing (a.k.a. coin-graph drawing and kissing disks drawing) and Spectral Graph Drawing and Spectral Graph Theory in my dissertation. 

The functionality of the Tutte algorithm is only guaranteed for graphs that are 3-connected, i.e. graphs with more than 3 vertices which cannot be torn apart unless at least two vertices are removed.

https://en.wikipedia.org/wiki/K-vertex-connected_graph             

Speaking of the conditions for the Tutte algorithm to work properly: Practically, this implies, for instance, that there should not be rooms connected only to one other room.

Anyhow, long story short, we have decided to continue with Spectral Graph Drawing and 3D force-directed graph drawing. These algorithms are ready and with a couple of adjustments for maximum speed and stability we will release them shortly. Some conditions for these algorithms are easier to ensure, but in general if a node(room) is connected to only one space or the graph is not well connected one cannot expect a good graph drawing from neither of these methods. The other issue that is also common is that the force directed graph drawing will not work if one forces a big bubble to be squeezed in the middle of smaller bubbles. Stay tuned. 

Thanks for the prompt reply Pirouz, 

Yes, I am going through your dissertation and other related literature/papers. My knowledge of graph theory and its implementation is limited for the moment so I need to do a bit of catching up. 

In regards to my previous question, your explanation of the functionality in the Tutte algorithm have provided answers as to why the component was failing in my workflow. Within the excel spreadsheet, i have tried to implement an approach where users can define the room connections via a series of abbreviations or code - which then translates to the generation of curve connections native within gH. Some of these connections have less than 3 vertices i.e relationship between bedroom and ensuite. Anyways, it has all been laid out in the sample script that I have attached. I have also tried to implement the matrix system to add weight to the connections based on the rating system in your latest presentation - SpaceSyntax4. 

I'll continue to explore and touch base with you if I get anywhere. Cheers. 

Hello! A newbie question here - I cant seem to get the script to show up in my GH. I install the package to the GH library folder, but when trying to open the sample script it cant find the components. I checked for the "unblock" option but it isnt even there for the files. If I copy paste the files from the "SYNTACTIC" folder into the Libraries folder GH asks which of the two to load on startup, but doesnt load any irregardless of what I chose. Am I missing something really obvious here? 

Got GH 9.0076

and Rhino 5

Hello Dmitri Korolenko, you only need to copy and paste the GHA (green file) into your components' folder. And this component needs to be checked for being unblocked. Sometimes you need to close your rhino and GH and relaunch both to see the menu of Space Syntax. Please try and let us know if the problem persists.

Hi Pirouz Nourian! Cant seem to get it to recognize the component. Put the green bit in the libraries folder, no unblock option in the properties, restart, grasshopper still does not see it. If this is not a common thing it might be my machine acting up. Ill keep trying.