GH_Gradual Module

結束了7個月的工作,最近會把這幾個月下來的東西稍微整理。
這篇的重點在於如何把漸變的原件控制在可模矩化的範圍,但同時仍具有漸變的效果。
畢竟若是為了漸變效果而告訴工廠你有上千種尺寸的鐵片要切時,你所要付出的時間和成本都太大了。

I'm going to organize the project I did during the past 7 months at work.
The point for this topic is trying to control Attractor effect, using divided units to make it become module units.
If you tell the factory you have thousand of sizes of medal that need to be cut, then it can be excepted that your need to pay huge cost and time.

Step1_左邊的部分與一般的attractor的步驟一樣。這邊我做的方式是把面細分成三角形後,再依據不同的方向與距離來做attractor效果。

Step1_ Left side of the graphic is similar to regular Attracor step. I divided surface into triangle sharp, and gave it different directions and moving distances to show Attractor.

三角形的Attractor
Triangle Attractor

Step2_右邊上下兩個步驟是一模一樣的,先選定你要"幾份模矩",然後把"模矩數據"重新做成一個list去取代原本的資料結構。

Step2_The lower and upper of the right part graphic are same, decided how many units you want, and rearrange the data list to replace the original data list.

原本的漸變
Original Attractor

模矩化後的漸變,這邊是做成五組模矩,若是想要更接近漸變效果當然也可以再增加模組
Become module units after the procedure. I used five module units. If you want the effect more close to the original effect, of course you can add numbers to the module lists.

將另外一個向度的三角形顯示出來,並用顏色來顯示模矩的位置
Display another direction of triangle panels, and show its location with color.




工作後發現其實很多東西不是自己一廂情願可以做到,
如何溝通和整合又更為重要。
Just find out you can't do whatever you want, how to communicate and organize also important.

--Shaun--

GH-RandomDrawing

一開始有個六角柱體，接著隨機的從側邊的六個面選一個面鏡射過去。
當然也可以只選擇其中兩個面來操作，每次產生的圖形都會也很大的不同。
上面的圖形大概是跑了將近4000的六角形柱，時間大概是15分鐘，
如果有時間的話可以試試更大的參數，但越到後面電腦越崩潰就是了。

可以看到有些地方因為重疊的關係顏色是比較深的，
而且常會有在原地打轉的問題，現在還想不到不會重複的方法。

記得以前大一給陳宣誠老師曾經對我說的一句話，
"你必須把自己丟遠點再去追"
數位好玩的地方有時也是因為電腦會帶你到你不曾想過的地方去，
有種你的腦還沒到，但你的手已經跑到很前面去了的感覺。



Generate  a hexagonal column at first, then random chooses one of  the side surface to  mirror it.
Of course you can choose two of  the side surfaces to manipulate, and you will get a variety of graphics each time.
The top graphic is running almost 4000 hexagonal columns about 15 minutes.
if you get time you can try more larger parameter, but it should be careful when the numbers of parameters beyond your computer ability.

Because of overlapping matter, graphic has deeper color at some areas,
and sometimes it will rotate in the same places more than one times, but I still don't know how to solve it.

My teacher(Eric Chen) has told me that you have to throw yourself into far away place. and trace it at the place you've never imagined before.I think that is why computer is fun.


-shaun--

GH-Burning Image

先在圖形上撒規則的點，使圖形能變成數據化，
每個點再以數據化後的數值分別向 x, y軸進行移動，
最後將搖擺的線與原本的直線Loft就完成了。

最簡單的應用是直接將數值當成半徑畫成圓，
在路上十分常見，也滿吸引人的。

Spreading regular points on the image at first, and the purpose is transforming image into data.
Next step is moving points toward x,y axis, and then image turns into data in each points.
At last, loft the original lines and the swing's lines.

The most easy application is regarded the data as radius of circle.
It's very common on the street, and I felt attracted as well.


--Shaun--

Portfolio

"DaiShuan Portfolio"

最近大概就是在忙著準備找工作用的作品集。

念了5年的建築其實回過頭來看還滿慶幸的，
畢竟有一些實體的東西可以檢視自己是否持續在往前。

做設計的時候一直有個壞習慣就是"收"不起來，
往往加了太多東西而失去原本的純粹性。
這在大五最後評圖的時候確實帶給我很大的困擾，
或許當中有想掩蓋某些不成熟的因素吧。

在作品集的當中，
可以發現我完全沒有放平面設計的圖，
因為我根本沒有那樣的圖可以放阿!
不知不覺地走向裝置和亭子的方向去了，
但這並不太表我我排斥傳統建築那塊領域，
只是能實驗新的方法和想法還是比較吸引著我。

Recently, i'm busy for portfolio, which the purpose is looking for the job.

Studied architecture for five years, i feel so lucky when i look back what i've done.
After all, i can inspect myself whether i'm keep moving forwards or not with some solid things.

I had a bad habit, when i designed something i couldn't figure out which steps i should stop,
added too much complicated elements into design, so lost pure goals at same time.
It brought me a huge trouble at final review,
perhaps, i wanted to cover some immature aspects.

i didn't put any plan's design into my portfolio,
because actually i didn't own that in the past three years!
unconsciously, i've walked toward installation and pavilion for a while.
But it doesn't mean i exclude traditional architecture field.
It's just because it deeply attract me if i can experiment new ways and thoughts.

--Shaun--

GH-Bubbles

一個簡單模擬氣泡的狀態。
與以往不同的是從原本規律的切割轉變成隨機不同大小的切割，
最理想的狀態是每個圓都與旁邊的原相切且相聯的，
但當泡泡變最大時可以發現還是有些許距離存在，
若是真的要做出來這個差距我想是不能被允許的，
還得再想想....

當然這也能運用在其他的形體上面，
謝謝阿湯大哥教我如何把面從原本的UV值做其他的分割。


A simple way simulates bubble's mode,
it's different that the cutting way transforms regular circle into random scales.
The most perfect goal is setting every circle connect the closet ones,
but the problem is you could see the tiny distance exists between two circles when the bubble turn into big ones.
I thought it's not allow if you really want to manufacture it in reality.
Still needs to thinking about it....

You could apply it in other shapes,
and thanks Tien teached me how to divide surface by changing UV values.

--Shaun--

GH-Pavilion of Voronoi

先畫出一個螺旋的線後，用Voronoi來分割
接著2D的線投射在曲面上。

有點太簡單的感覺，
不過先放著吧。



First just draw a spiral curve, and subdivide it by Voronoi,
then project into the surface.

Felt too easy, but put it waiting for next combo.


--Shaun--

GH-Net

以前在"數位迪化"展覽入口附近有個裝置，
全都是以圓形壓克力組構，
用同樣形狀但不同大小的單元去構成大的形體，
如果能在不同的小單元間有互相關聯的話應該會更有趣。

Over a few years ago, an installation near the entry of digital exhibition,
using same circle shape but different scale of acrylic to  assemble whole shape.
if i can make small units interact with each others should be more interesting.

--Shaun--