Space Syntax

Comment

Comment by Pirouz Nourian on January 26, 2022 at 10:50am

Hi Asya, sure, happy to help, I find the Termite nests very inspiring; thanks for sharing the picture! Re your questions:

1. The notion of a 3D graph is somewhat confusing because the graph itself is not a 3D space but yes, sure, if the vertices are "embedded" (mathematical term for 'drawn'/placed) in a 3D space you can colloquially call it a discrete model of the 3D space... Bottom line is that you need to produce the topology of the graph as a set of edges connecting those vertices. Since you have the models as meshes, you kind of implicitly have the graph data in the mesh as face to face adjacency graphs. If you are familiar with Python I can help you write that. Or I might even find it as I have written it before and share it. In the meantime, have a look at Ivy, it has a tool that can produce such graphs (as far as I remember).

2. once you have your graph data as a set of edges, practically a set of tuples of two integeres, referring to the vertex indices, then you can initiate an NX graph and then you are all set for computing geodesic distances

3. I suspect that NX would be/could be faster in running some analyses but I must say I did not quite fully understand what is the goal/process. If you have a publication or a proposal or a description of the process please feel free to send me through email p.nourian at tudelft.nl. I would appreciate it if you could summarise the idea in a workflow diagram (even if drawn by hand)

2. 

Comment by asyimut on January 26, 2022 at 8:15am

Dear Pirouz, 

Thank you for the great suggestions. 

1. Can I create a 3D graph representation of the topologies? Can you quickly see the figure1 in the attached paper? 

2. Then, NetworkX would recognise and compute the SP s in the 3D graph? 

3. In the meantime, I made a very heavy model in which I can calculate a limited amount of shortest paths. See the attached image. 

This is a beehive design. The combs which are inhabitable on both sides are radially distributed around a central core. There is a corridor around the inner core, as well as a peripheral gallery. Bees can also move above and below the combs. So I need the 3D dimensions in my model, or my mathematical imagination is not capable of thinking this planarly? I used the library from this link https://github.com/danilnagy/gd_tools but my model is very heavy. I made a triangulated mesh in order to define all the paths that the bees can take and defined random 20 points as start and end points to find the SPs. I guess with your system this would be faster. This image represents how I would like to think of the nest spaces and connectivity but in the honeybee nest case- which is a designed one not a natural one-

Comment by Pirouz Nourian on January 26, 2022 at 6:12am

Hi  asyimut, 

Interesting question! I would try to build a face-to-face adjacency graph of the meshes in question (e.g. using Ivy or Python) then you can use GH CPython to access NetworkX and compute all shortest paths (or centrality indicators) and do the rest in Python or GH. If you start the Python codes I can help you by giving hints on how to finish them. 

Comment by Pirouz Nourian on January 26, 2022 at 6:09am

Hi Eman Baadullah,

While running the exe installer file you need to specify a location in which the gha component will be placed. Please run it again and specify a location that you can recognise. Then you can follow the instructions of Sama to ensure that the plugin appears in your GH canvas. I hope this resolves the issue. The download link is here: https://genesis-lab.dev/products/syntactic/

Comment by Eman Baadullah on January 24, 2022 at 11:18pm

Hi.... I am using  Rhino7.14.22010.17001 and the .exe file I have run does not appear in my GH canvas. Is there another location I can download a normal .gh component file? the one I downloaded from food4rhino is incomplete.

also, I did the way Samane explained but it doesn't work.

Comment by asyimut on January 9, 2022 at 10:33am

Hi Guys! I would like to use the idea of adjacency, closeness, betweenness of areas in order to decide on some dimensional parameters of my design. So, potentially I would feed the average shortest paths of each design topology as fitness measure into a genetic algorithm, and maybe Biomorpher. So let's say I subdivide the surfaces which occur below as iterations in Biomorpher and create a graph representation as routes and nodes. Do you know any heuristic which I can calculate the average shortest paths in such way? cheers!

Comment by Samane on January 8, 2022 at 9:40am

If the plugin does not appear in your canvas, try the following:

Comment by asyimut on January 7, 2022 at 10:55am

Hello!I am using  Rhino7 and the .exe file I have ran does not appear in my GH canvas. Is there another location I can download a normal .gh component file? 

Comment by Pirouz Nourian on December 2, 2021 at 11:23am

Hi there, here is a new book chapter about going beyond analyses towards syntactic evaluations, containing some tutorials: https://www.researchgate.net/publication/354683441_Reuse_On_Evaluat...

Comment by Pirouz Nourian on September 4, 2021 at 12:47pm

Hello gaston, you seem to have downloaded an incomplete new version from Food4Rhino. You can get the last complete version from here:

https://genesis-lab.dev/products/syntactic/

Stay tuned for more updates on Genesis Lab

I hope this helps!

• ‹ Previous
• 1
• 2
• 3
• 4
• …
• 40
• Next ›
• Page