If you need assistance with your assignment 3 submission, you can email me directly.
danny.nguyen@unsw.edu.au
Good luck everyone with the final submission!
Tuesday, June 5, 2012
Wednesday, April 25, 2012
Some Grasshopper Examples to Experiment with
After marking the first assignments, I noticed some kinds of problems many students were running into, and figured it'd be handy to throw together some working Grasshopper files that you can play around with to help get your heads around how lists and trees work together, as well as generating certain kinds of repetitive geometry. In no particular order, here are the files (put into one .rar file) for you to look at, experiment with, and learn from for your own Grasshopper scripts.
http://www.gamefront.com/files/21608089/examples.rar
Tuesday, April 3, 2012
Using the Galapagos Component in Grasshopper
A Brief Description of Galapagos and How to Use It
To recap, Galapagos is a component in Grasshopper that has a unique function and way of using it. It takes a bunch of sliders hooked up to it (which Galapagos refers to as the "Genome"), and changes them by small, gradual amounts, all the while checking a number (produced by what I'll explain later as the "Fitness function") and trying to make that number bigger or smaller over time, depending on the settings you choose.
To make the Galapagos component work, you hook its "Genome" parameter up to any number of sliders (click and drag from "Genome" to the slider; OR select a bunch of sliders, right click on "Genome", and click "Selected Sliders"), and then hook its "Fitness" parameter up to a number that is produced by your Grasshopper script (click and drag from "Fitness" to the output number parameter).
To let Galapagos do its thing, you doubleclick the Galapagos component and a window pops up with three tabs. Most times, you can just leave the first tab as it is. For the short 'n' simple explanation, to run Galapagos, click on the "Solvers" tab and click "Start Solver".
To get a little more advanced, you can change the kind of solver used by clicking one of the two buttons to the left of the "Start Solver" button - "Evolutionary Solver" (default, left) and "Simulated Annealing Solver" (right). The difference between these two options is simply the process Galapagos follows to try and maximise / minimise the Fitness number.
Handy Links Explaining How the Evolutionary Solver Works
These links are the same as the ones that show up under the "Options" tab when you doubleclick Galapagos. The reason I've got them here is because they're essential towards understanding how to use Galapagos. If you don't read these, you're putting yourself at a huge disadvantage.
Evolutionary Principles applied to Problem Solving
Fitness Functions
Selection
Coupling
Coalescence
Mutations
Define "Fitness"...
Fitness Pressure
On Getting Lucky in Higher Dimensions
Lastly, I'm working on getting some simple video tutorials together explaining how to use the Galapagos component. You can expect to see them posted here some time over the next week, depending when I get time to create them.
To recap, Galapagos is a component in Grasshopper that has a unique function and way of using it. It takes a bunch of sliders hooked up to it (which Galapagos refers to as the "Genome"), and changes them by small, gradual amounts, all the while checking a number (produced by what I'll explain later as the "Fitness function") and trying to make that number bigger or smaller over time, depending on the settings you choose.
To make the Galapagos component work, you hook its "Genome" parameter up to any number of sliders (click and drag from "Genome" to the slider; OR select a bunch of sliders, right click on "Genome", and click "Selected Sliders"), and then hook its "Fitness" parameter up to a number that is produced by your Grasshopper script (click and drag from "Fitness" to the output number parameter).
To let Galapagos do its thing, you doubleclick the Galapagos component and a window pops up with three tabs. Most times, you can just leave the first tab as it is. For the short 'n' simple explanation, to run Galapagos, click on the "Solvers" tab and click "Start Solver".
To get a little more advanced, you can change the kind of solver used by clicking one of the two buttons to the left of the "Start Solver" button - "Evolutionary Solver" (default, left) and "Simulated Annealing Solver" (right). The difference between these two options is simply the process Galapagos follows to try and maximise / minimise the Fitness number.
Handy Links Explaining How the Evolutionary Solver Works
These links are the same as the ones that show up under the "Options" tab when you doubleclick Galapagos. The reason I've got them here is because they're essential towards understanding how to use Galapagos. If you don't read these, you're putting yourself at a huge disadvantage.
Evolutionary Principles applied to Problem Solving
Fitness Functions
Selection
Coupling
Coalescence
Mutations
Define "Fitness"...
Fitness Pressure
On Getting Lucky in Higher Dimensions
Lastly, I'm working on getting some simple video tutorials together explaining how to use the Galapagos component. You can expect to see them posted here some time over the next week, depending when I get time to create them.
Galapagos Video Jeremy Showed in Class
Project Teaser from Nate Holland on Vimeo.
Video description on Vimeo:
This animation condenses my architectural thesis project from an entire year of research and design into four short minutes.
Video and project by Nate Holland.
Sunday, March 25, 2012
Guitar Hero in Grasshopper and Circle Packing
This video shows what can be done in Grasshopper with a little ingenuity. The designer uses a program to capture audio of him playing his guitar, then uses a Grasshopper plugin called gHowl - available here - to display that audio information as spheres in Grasshopper.
This is a long video, so skip towards the end to see the cooler stuff this designer produces. This video shows various kinds of circle packing done in Grasshopper, feeding in different parameters to determine the size, shape, and orientation of the circles. The most interesting results in my opinion are when he starts using pixel colours in images as the parameters.
Wednesday, March 21, 2012
MVRDV's Iterative Lego Skyscrapers
If you're wondering about the best way to display your iterative process in your posters, check out how MVRDV displays their 676 Lego towers!
http://www.fastcodesign.com/1665603/mvrdvs-winy-maas-builds-676-lego-skyscrapers
http://www.fastcodesign.com/1665603/mvrdvs-winy-maas-builds-676-lego-skyscrapers
Flying Bird Man - Real or Hoax? You Decide!
To keep with the flying theme from the previous post, here's a recent internet video sensation that has yet to be definitively proven as a fake.
What do you think?
What do you think?
Monday, March 19, 2012
Saturday, March 3, 2012
Friday, March 2, 2012
Grasshopper Tips and Tricks
Here are some quick tips to streamline your workflow in Grasshopper.
1) Quickly Adding Components:
To quickly add components to the canvas, doubleclick the canvas, and type in the component name. Similar to a Google search, Grasshopper shows you results it thinks you're looking for as you type. Eg, typing XYZ will turn up "Point XYZ" without typing in the whole component name.
2) Renaming Components:
To change the name of a component placed on the canvas, rightclick its name, and at the top of the menu that appears, you can type in a different name. You can also rename input and output parameters in this way. As your flowgraphs get more and more components in them, changing the names of components becomes helpful to remind you what that component is supposed to be producing. (Eg, if you're using a Polyline component to make a site boundary, you might rename it "Site Boundary Polyline".)
3) Component Errors (Orange and Red Components):
If a component goes orange, there's a problem with its inputs, but it can still produce some output. If a component goes red, there's a problem with its inputs that stops it from producing output. To find out what the problem is, hover your mouse over the little speechbubble above the component, and some text will appear telling you what's wrong.
4) Panels to Help Bug Finding:
Panels can be very handy for bug fixing, since they take an input and convert it to text for you to read. Doubleclick the canvas and type in "panel" to create one quickly. If a flowgraph isn't working the way you expect it to, take a guess at what components aren't behaving the way you want them to, and connect their output to a panel. A lot of the time, you'll get a list of outputs that you can look at to see what's wrong. (Sometimes things get a little more complex than that, but we can cross that bridge when we come to it.)
5) Hiding and Unhiding Components:
Most components you'll use will produce some graphic that appears in the Rhino viewport. To hide and unhide the geometry, you can select one or many components on the canvas, and hit Ctrl+Q to toggle their visibility. If you select a mix of components that are both hidden and visible, hitting Ctrl+Q will make the visible ones hidden, and the hidden ones visible.
6) Connecting Multiple Outputs to One Input:
You should know by now that you can connect an output to an input by click-and-dragging from one node to another. However, this method only lets you connect one output to one input. To connect multiple outputs to one input, click-and-drag, then hold shift. If you're doing it right, you'll see a little green arrow and a plus sign appear next to the cursor.
7) Some Different Ways of Disconnecting Outputs:
To disconnect an output from an input, hold Ctrl while click-and-dragging between the input and output. A little red arrow and a minus sign should appear next to the cursor. Alternatively, you can right click an input, click Disconnect in the dropdown menu, then click the input you want to disconnect. Alternatively again, you can right click an input, click Manage Collection, and use the popup dialogue that appears to do the disconnecting.
8) Easily Finding Out How to Use a Component or its Parameters:
Right click the component name or the parameter name, and click "Help" at the bottom of the menu. If you're doing this for a component, a window will pop up inside Grasshopper explaining what the inputs and outputs are, as well as a quick description of what the component does. For a parameter, a window will pop up inside Grasshopper explaining about the its "type". Some of the parameters have an overly technical explanation, so you're better off getting your information from the component's help.
1) Quickly Adding Components:
To quickly add components to the canvas, doubleclick the canvas, and type in the component name. Similar to a Google search, Grasshopper shows you results it thinks you're looking for as you type. Eg, typing XYZ will turn up "Point XYZ" without typing in the whole component name.
2) Renaming Components:
To change the name of a component placed on the canvas, rightclick its name, and at the top of the menu that appears, you can type in a different name. You can also rename input and output parameters in this way. As your flowgraphs get more and more components in them, changing the names of components becomes helpful to remind you what that component is supposed to be producing. (Eg, if you're using a Polyline component to make a site boundary, you might rename it "Site Boundary Polyline".)
3) Component Errors (Orange and Red Components):
If a component goes orange, there's a problem with its inputs, but it can still produce some output. If a component goes red, there's a problem with its inputs that stops it from producing output. To find out what the problem is, hover your mouse over the little speechbubble above the component, and some text will appear telling you what's wrong.
4) Panels to Help Bug Finding:
Panels can be very handy for bug fixing, since they take an input and convert it to text for you to read. Doubleclick the canvas and type in "panel" to create one quickly. If a flowgraph isn't working the way you expect it to, take a guess at what components aren't behaving the way you want them to, and connect their output to a panel. A lot of the time, you'll get a list of outputs that you can look at to see what's wrong. (Sometimes things get a little more complex than that, but we can cross that bridge when we come to it.)
5) Hiding and Unhiding Components:
Most components you'll use will produce some graphic that appears in the Rhino viewport. To hide and unhide the geometry, you can select one or many components on the canvas, and hit Ctrl+Q to toggle their visibility. If you select a mix of components that are both hidden and visible, hitting Ctrl+Q will make the visible ones hidden, and the hidden ones visible.
6) Connecting Multiple Outputs to One Input:
You should know by now that you can connect an output to an input by click-and-dragging from one node to another. However, this method only lets you connect one output to one input. To connect multiple outputs to one input, click-and-drag, then hold shift. If you're doing it right, you'll see a little green arrow and a plus sign appear next to the cursor.
7) Some Different Ways of Disconnecting Outputs:
To disconnect an output from an input, hold Ctrl while click-and-dragging between the input and output. A little red arrow and a minus sign should appear next to the cursor. Alternatively, you can right click an input, click Disconnect in the dropdown menu, then click the input you want to disconnect. Alternatively again, you can right click an input, click Manage
8) Easily Finding Out How to Use a Component or its Parameters:
Right click the component name or the parameter name, and click "Help" at the bottom of the menu. If you're doing this for a component, a window will pop up inside Grasshopper explaining what the inputs and outputs are, as well as a quick description of what the component does. For a parameter, a window will pop up inside Grasshopper explaining about the its "type". Some of the parameters have an overly technical explanation, so you're better off getting your information from the component's help.
Wednesday, February 29, 2012
Examples from the First Tutorial
Here are a couple of quick examples in Grasshopper from the first tutorial class!
Subscribe to:
Posts (Atom)