COMPUTER AIDED DESIGN OF MING CHAIR
Dr. Zhang Kun, Prof. Wang Bo Wei
Tongji University, Shanghai,
China
e-mail:
doctorkun@hotmail.com
Abstract
Elegant and ingenious Ming chair has gained an important place in the
world of furniture, there are many beautiful chair styles and this paper
proposed an idea that by the aid of computers, we can study the possible
combinations of existing styles and may design new and beautiful chairs. This
paper also realized several interesting deformations based on the combination
results. The deformation is not just for fun but can be seen as a approach to
stimulate the designer’s creativity.
1. Ming Chair
Ming dynasty (1368-1644) is a period during which Chinese furniture
manufacturing has reached the zenith of aesthetic. The flourishing of furniture
manufacturing was due to the stable political situation, overseas trade and
demanding for fine furniture. Ming furniture was made mainly in Su Zhou City
and its outskirts. Su Zhou City is located near Chang Jiang River and is not
far from sea, so it is easy to import precious timbers from south-east Asia,
these precious timber such as rosewood, cedar, ebony are very good raw
materials to make furniture. By the cleverness and exquisite craftsmanship of
ancient Chinese people, precious timbers were changed into elegant, ingenious
and unornamented furniture.
Ming furniture includes chairs, beds, tables, cabinets, racks and
screens. This paper studies chairs. Ming chair is divided by Chinese scholars
into four categories: high-back chair, horseshoe chair, folding chair and
armchair. The difference between high-back chair and armchair is that high-back
chair has no armrest. The name of horseshoe chair comes from the shape of crest
rail and armrest, these two parts merged together and the shape looks like a
horseshoe. Folding chair’s front legs and
back legs crossed and the chair can be folded. Figure 1 shows the four kinds of
Ming chairs: from left to right, they are high-back chairs, armchairs,
horseshoe chairs and folding chairs.
Figure 1. Four categories of
Ming chair
Chair is a very important furniture, it supports people in the daytime,
allowing people to read, write and talk in a comfortable posture. It could be
say that chair is a symbol of civilization, so it is essential for mankind to
produce good chairs both in quality and aesthetic. In modern society, it is
easy for us to produce good quality chairs because we own assembly lines, and
we can find good timbers. But it is not easy to satisfy the need of aesthetic,
since different people have different aesthetic that is a matter of philosophy,
culture and history. The fact is Ming chair has a good reputation for its
beauty and is loved by people around the world. We believe that Ming chair
contains both good quality and high level aesthetic, it is a source of
creativity for modern furniture design.
2. Computer aided design of
Ming chair
A chair can be divided into smaller parts, for example, crest rail,
armrest, splat, stile and so on. This paper’s aim is to study
these sections, find their features and reunite them, and then to create new
styles of Ming chair.
The approaches consist of three steps. The first step is to disassemble
chairs, study every style of each part. Of course, it does not mean to
disassemble a true chair, it is a work done in the mind. Through this step we
learned that Ming chair has beautiful contour lines, there are straight lines
and curved lines. Curved lines have two main shapes: S-shape and C-shape. We
believe that curved line was designed both for practical purpose and decoration
purpose. A curved splat is more suitable for people to lean against, and a
curved crest rail is more beautiful than straight one.
The second step is to build 3D models for these chair parts. 3D model
is good for visualization, it looks like a real one and user can view the model
from every angle. We used 3D modeling software to build these models. Although
there are many tools for modeling, experience and “craftsmanship” are also
needed because
the shapes ought to imitate the Ming chair best. The completed 3D models are
stored in a database which will be accessed by a program later.
The last step is assembles a new chair by gluing all parts of the chair
together. This task is finished by computer program, the user only need to
select which style he wants for every part; for example, user can select a
curved armrest or a straight armrest. This paper has developed a program that
fulfills the last step. Figure 2 shows the interface of the program.
Figure 2. The program
interface
This interface provide five chair parts for selection, they are crest
rail, stile, splat, armrest and armrest bracket. Each part has two or more
styles represented with icons. User can choose only one style of each part, and
every part has a default style. The program will automatically “glue” these parts
together according to the user’s selection. It
can be seen from the interface that there are two possible crest rails, two
stiles, two splat two armrests and three armrest brackets, so, the possible
results is 2 ×2×2×2×3, that is 48 results. If there are more styles for each part, there
will be more possible combinations. For simplicity reason the program does not
consider the selection of other parts. Figure 3 shows four glued chairs.
Figure 3. Four “glued” Ming chairs
To make the results more interesting, this program adds four
deformations to the chair model, they are scale, taper, bend and stretch. The
degree of each deformation can be controlled by parameters. Scale deformation
has three parameters, they are x factor, y factor and z factor, x, y, z means
three directions in the coordinate system. The factor was given a random value,
so the result is different each time. The first chair in figure 4 shows the
scaled result.
Figure 4. Chairs after deformation
Taper deformation has three parameters: primaryaxis, amount and curve.
In the program, primaryaxis is z, curve is -1.5 that gives the outline a curvy
look, and amount is a random value that ranged from 0.5 to 3.5. In fact, parameter
curve could also be set to random, but the result would be more unpredictable.
The second chair in figure 4 shows the tapered result.
Bend deformation could produce interesting result. In figure 4, the
third model looks like a rocking chair. The bend degree is controlled by the
parameter angle. There are other parameters to control the deformation such as
direction, but our program just ignored them and used the default value.
The above three deformations method have a common ground that they affect
the whole chair model, while stretch deformation only affect parts of the
chair. In the program, stretch deformation only adds to legs and stiles. Like
the last chair shown in figure 4.
3. Conclusion
We have seen what this program can do, in fact, we can expect this
program to do more things. For example, let program select the chair parts
automatically without the user’s interaction,
and produces a sequence of new style chairs. User can choose the most beautiful
chairs from the sequence, or, in the future, the program will have the ability
to judge which chair is the most beautiful one. Then there emerges a question:
does computer can do the design work instead of human being?
We know that computer is becoming more and more clever, it can play
chess with the champion. We believe that one day it may do design works by
itself, this is really a challenge dream. Please look at the chair in figure 1,
we noticed that the side pedals are higher than front pedal, the back pedal is
higher than side pedals, it just like a stairs. This kind of pedals has a very
interesting Chinese name “Bu Bu Gao”, which means
step more and more higher. It is a hint that person, who sit on this chair is
likely to make progress step by step, for example be promoted to higher and higher
positions. We believe that the studies of computer designs will also get
progress step by step in the future.
References
[1] Celestino Soddu, “La Citta’ Ideale”
Generative Codes Design Identity, GA2002 Proceedings, 2002
[2] Wang Shi Xiang, “Appreciating of Ming
Furniture”, San Lian Publisher of HongKong, 2001