[Music]
hello so I’m an artist who creates
things by growing them but growing them
algorithmically inside a computer what
really fascinates me are the types of
forms that you see in nature
particularly the types of forms that
when you see them you know instinctively
that they’ve been created by a living
thing through a process of growth and
what is it that’s different about that
type of form compared with man-made
structures created by traditional
top-down design and construction what is
it about natural forms that resonate so
deeply with us I think there’s something
about natural organic form structure
that we’re particularly attuned to from
things we find quite beautiful to things
that we find disturbing why is it that
they used harder forms elicits such a
strong range of aesthetic reaction the
other big thing that I’m really
interested in is the idea of emergence
the idea that rich complex structure
ordered or even disordered structure can
often come from incredibly simple
processes so what I’m doing is I’m
writing code in the computer that
represents rules for how things grow so
I’ve got two our cells and cells need
food to grow and food can be created by
various different means such as light
rays coming in from outside hitting
cells and the cells they get the most
light rays create the most food and then
cells can be greedy and take all that
food themselves or they can be
cooperative and share the food with
their neighbors
and then various different things could
affect how the cells develop how they
grow such as which direction does a cell
splits in when it divides does it split
in the direction of most tension in the
structure or least tension or does the
curvature of the surface affects
direction of splitting or gravity so
lots of different things all of which
could potentially change how things grow
in interesting but also very difficult
to predict ways so what I’m doing is I’m
writing code in the computer to
basically examine explore this space of
possibilities well you could consider an
artificial biology so what you’re seeing
on the screen are some my simulations
running and they all start with a small
ball of just 42 cells and then over time
the cells split and over thousands of
time steps this develops into a
structure with over 50 million cells and
what I’m doing is I’m influencing things
in the way that maybe a plant breeders
could try to create new plant types so I
might grow generates few hundred
particle seedlings basically forms with
a million cells in them and then on
those seedlings I might look at and
finds okay they’re just a few of them
that I think look particularly
interesting like a crossbreed them with
each other to hopefully produce ceilings
which are even more interesting and if I
find a seedling I really like I come
might be able to generate randomness
around it almost like mutate its
artificial DNA to see where that produce
is no more interesting variety around it
so you can see what was what I’m doing
is more like a artificial version of
natural selection or you could describe
as being survival of the intriguing
rather than the survival of the fittest
so everything I’ll be showing you so far
all the ill the cells are the same in
the structures so just like structures
with just one cell type now I’m going to
show you what happens when you start
introducing more cells
to the structure now she these are
structures where I’ve they’re just
actually two different cell types where
I call hybrid forms so they start with
skin their small ball of cells but the
cell at the top has one set of
properties and the cell at the bottom
has a different set of properties and
then over time as the cells develop
they’ve basically the different cell
types can grow in different ways so
potentially make different geometries or
different structures or grow at
different rates
I think to be compete with each other
for space as the form grows and I think
that you can see is what you get from in
such a small addition to the system just
one extra cell in the mix are
surprisingly lifelike structures but
mostly it really surprised me I mean I
was expecting more variety but the
structures as bases real emergence
because I was expecting more variety but
it’s far far richer than I expected it
would be just introducing one extra cell
into the into the system and one idea of
things very interesting this is the idea
that how nature potentially works is by
creating them as a system that can
create rich possibilities rich order
what the experimental biologist Stewart
Kauffman calls order for free and then
evolution through natural conception can
basically choose between these
possibilities instead of having to
design everything it’s choosing between
possibilities which have Vantage for the
creature have all things that help it
survive and was following this idea
could we use similar ideas ourselves
when we as designing and creating things
so instead of designing everything for
the top down was coming up the blueprint
all the components and how they will
assemble together coming up with the
process and exploring and trying to
influence that to do the things that
we’re interested in so what I’m next
going to show you are some the recent
experiments that I’ve been doing in that
direction where when I’m trying to
create our was like VARs like sculpture
is that potentially could be fabricated
as actually as physical sculptures and
what I’ve been intended to work with is
this robotic machinery that works by
depositing layers of material at a time
and there’s some limits to what you can
fabricate with this type of machinery in
particular so that you can’t have more
than a certain overhang because every
layer needs to have at least something
underneath it to support the next layer
so they’re sort of come constraints and
limits that you can actually fabricate
with I’d so what I’m trying to do is
influence the growth in the ways they
might try to influence the plant growing
by almost having wires for the blower
plant to grow on or more or less
nutrient in different areas where you
want to encourage the plant to grow in
some places maybe discourage it in
others so what I wanted to create these
these VARs like structures with was like
a flat bottom and a flat top so I’m
starting with a tubular structure and
the cells at the top and the cells at
the bottom have to basically stay in two
planes and then their various different
influences which are hopefully going to
modify how it grows to naturally create
FAR’s like structures without me
dictating exactly what that means
so firstly cells towards the top can
grow faster than cells towards the
bottom so maybe you’re likely to get
shapes which have broader wider at the
top and the bottom then the rate of
growth can also be affected by as the
cells grow outwards to potentially limit
the growth so that it doesn’t grow too
far and then finally also introduced an
influence so based on the angle between
the cells so as the angle between the
cells gets beyond certain limits that
can slow down the growth
– hopefully naturally create structures
that would be fabric Obul with the
machinery that I was intending to use so
what I got was I expected but really not
in the way I was at almost hoping
originally what I got with these
structures that start growing these was
like really intricate branching tendril
like structures and what was happening
here there’s actually
I’ll give my code so that a cell that
sticks slightly a further out than its
neighbors tends to grab all the food so
sense be the focus where all the growth
happens and also the way that I’ve
thought about doing influencing the
growth based on the angle between the
cells I hadn’t quite thought through
carefully enough it was having the
effect of almost solidifying almost like
making sort of slight solid rock any
cells which are tending towards
horizontal so basically I’d accidentally
created rules to growth these branches
some real emergence I hadn’t
deliberately programmed it to create
these structures the the system almost
had exploited my bugs to do it now I do
think they re quite beautiful so useless
to make sculptures with least the
current machinery has tended to use but
idly just really quite beautiful in
their own way and I think also more
subtle than if I deliberately tried to
create branching structures I don’t
think I would have created things
nearly as interesting and natural and
subtle to them so one thing about being
arts this has almost become like an
accidental sort of extra art series so I
can’t fabricate these are sculptures
they can exist virtually digitally so
I’m presenting these as stereoscopic
installations so you can perceive them
three dimensionally but in a virtual
space I then went back to the code and
had a look at it and learning from what
I had seen basically fix the bugs and
change the influences and now I’m
getting forms like these which were much
more what I was initially hoping for
heavy fabrica build sculptural shapes
and I’m now actually starting to
fabricate these so these are still at
present just about 40 centimeter high
quite small sculptures but I’m hoping
now to be able to fabricate these much
bigger and I think one thing I’m seeing
is a really interesting range of natural
form in these structures clearly on
different growth rates in different
areas seems to produce interesting
variation and contrast of likes
moving us in detail and also the
structures often have I think are
potentially eerily interesting
structural properties things like
grooves like bark which potentially are
really strong when they’re fabricated so
one thing I’m learning from these
experiments is how powerful this idea of
emergence is the idea of creating things
through generative systems and
influencing the results another big
thing I’m learning is that you often
need to tread quite carefully it’s very
much a process of learning a process of
discovery finding out what happens and
appropriate ways of influencing with the
results complex systems are capable of
doing amazing things there’s also in
their nature to do the unexpected
of course this can also be great leading
you in new directions that you never
imagined were possible thank you
[Applause]
[Music]
hello so I’m an artist who creates
things by growing them but growing them
algorithmically inside a computer what
really fascinates me are the types of
forms that you see in nature
particularly the types of forms that
when you see them you know instinctively
that they’ve been created by a living
thing through a process of growth and
what is it that’s different about that
type of form compared with man-made
structures created by traditional
top-down design and construction what is
it about natural forms that resonate so
deeply with us I think there’s something
about natural organic form structure
that we’re particularly attuned to from
things we find quite beautiful to things
that we find disturbing why is it that
they used harder forms elicits such a
strong range of aesthetic reaction the
other big thing that I’m really
interested in is the idea of emergence
the idea that rich complex structure
ordered or even disordered structure can
often come from incredibly simple
processes so what I’m doing is I’m
writing code in the computer that
represents rules for how things grow so
I’ve got two our cells and cells need
food to grow and food can be created by
various different means such as light
rays coming in from outside hitting
cells and the cells they get the most
light rays create the most food and then
cells can be greedy and take all that
food themselves or they can be
cooperative and share the food with
their neighbors
and then various different things could
affect how the cells develop how they
grow such as which direction does a cell
splits in when it divides does it split
in the direction of most tension in the
structure or least tension or does the
curvature of the surface affects
direction of splitting or gravity so
lots of different things all of which
could potentially change how things grow
in interesting but also very difficult
to predict ways so what I’m doing is I’m
writing code in the computer to
basically examine explore this space of
possibilities well you could consider an
artificial biology so what you’re seeing
on the screen are some my simulations
running and they all start with a small
ball of just 42 cells and then over time
the cells split and over thousands of
time steps this develops into a
structure with over 50 million cells and
what I’m doing is I’m influencing things
in the way that maybe a plant breeders
could try to create new plant types so I
might grow generates few hundred
particle seedlings basically forms with
a million cells in them and then on
those seedlings I might look at and
finds okay they’re just a few of them
that I think look particularly
interesting like a crossbreed them with
each other to hopefully produce ceilings
which are even more interesting and if I
find a seedling I really like I come
might be able to generate randomness
around it almost like mutate its
artificial DNA to see where that produce
is no more interesting variety around it
so you can see what was what I’m doing
is more like a artificial version of
natural selection or you could describe
as being survival of the intriguing
rather than the survival of the fittest
so everything I’ll be showing you so far
all the ill the cells are the same in
the structures so just like structures
with just one cell type now I’m going to
show you what happens when you start
introducing more cells
to the structure now she these are
structures where I’ve they’re just
actually two different cell types where
I call hybrid forms so they start with
skin their small ball of cells but the
cell at the top has one set of
properties and the cell at the bottom
has a different set of properties and
then over time as the cells develop
they’ve basically the different cell
types can grow in different ways so
potentially make different geometries or
different structures or grow at
different rates
I think to be compete with each other
for space as the form grows and I think
that you can see is what you get from in
such a small addition to the system just
one extra cell in the mix are
surprisingly lifelike structures but
mostly it really surprised me I mean I
was expecting more variety but the
structures as bases real emergence
because I was expecting more variety but
it’s far far richer than I expected it
would be just introducing one extra cell
into the into the system and one idea of
things very interesting this is the idea
that how nature potentially works is by
creating them as a system that can
create rich possibilities rich order
what the experimental biologist Stewart
Kauffman calls order for free and then
evolution through natural conception can
basically choose between these
possibilities instead of having to
design everything it’s choosing between
possibilities which have Vantage for the
creature have all things that help it
survive and was following this idea
could we use similar ideas ourselves
when we as designing and creating things
so instead of designing everything for
the top down was coming up the blueprint
all the components and how they will
assemble together coming up with the
process and exploring and trying to
influence that to do the things that
we’re interested in so what I’m next
going to show you are some the recent
experiments that I’ve been doing in that
direction where when I’m trying to
create our was like VARs like sculpture
is that potentially could be fabricated
as actually as physical sculptures and
what I’ve been intended to work with is
this robotic machinery that works by
depositing layers of material at a time
and there’s some limits to what you can
fabricate with this type of machinery in
particular so that you can’t have more
than a certain overhang because every
layer needs to have at least something
underneath it to support the next layer
so they’re sort of come constraints and
limits that you can actually fabricate
with I’d so what I’m trying to do is
influence the growth in the ways they
might try to influence the plant growing
by almost having wires for the blower
plant to grow on or more or less
nutrient in different areas where you
want to encourage the plant to grow in
some places maybe discourage it in
others so what I wanted to create these
these VARs like structures with was like
a flat bottom and a flat top so I’m
starting with a tubular structure and
the cells at the top and the cells at
the bottom have to basically stay in two
planes and then their various different
influences which are hopefully going to
modify how it grows to naturally create
FAR’s like structures without me
dictating exactly what that means
so firstly cells towards the top can
grow faster than cells towards the
bottom so maybe you’re likely to get
shapes which have broader wider at the
top and the bottom then the rate of
growth can also be affected by as the
cells grow outwards to potentially limit
the growth so that it doesn’t grow too
far and then finally also introduced an
influence so based on the angle between
the cells so as the angle between the
cells gets beyond certain limits that
can slow down the growth
– hopefully naturally create structures
that would be fabric Obul with the
machinery that I was intending to use so
what I got was I expected but really not
in the way I was at almost hoping
originally what I got with these
structures that start growing these was
like really intricate branching tendril
like structures and what was happening
here there’s actually
I’ll give my code so that a cell that
sticks slightly a further out than its
neighbors tends to grab all the food so
sense be the focus where all the growth
happens and also the way that I’ve
thought about doing influencing the
growth based on the angle between the
cells I hadn’t quite thought through
carefully enough it was having the
effect of almost solidifying almost like
making sort of slight solid rock any
cells which are tending towards
horizontal so basically I’d accidentally
created rules to growth these branches
some real emergence I hadn’t
deliberately programmed it to create
these structures the the system almost
had exploited my bugs to do it now I do
think they re quite beautiful so useless
to make sculptures with least the
current machinery has tended to use but
idly just really quite beautiful in
their own way and I think also more
subtle than if I deliberately tried to
create branching structures I don’t
think I would have created things
nearly as interesting and natural and
subtle to them so one thing about being
arts this has almost become like an
accidental sort of extra art series so I
can’t fabricate these are sculptures
they can exist virtually digitally so
I’m presenting these as stereoscopic
installations so you can perceive them
three dimensionally but in a virtual
space I then went back to the code and
had a look at it and learning from what
I had seen basically fix the bugs and
change the influences and now I’m
getting forms like these which were much
more what I was initially hoping for
heavy fabrica build sculptural shapes
and I’m now actually starting to
fabricate these so these are still at
present just about 40 centimeter high
quite small sculptures but I’m hoping
now to be able to fabricate these much
bigger and I think one thing I’m seeing
is a really interesting range of natural
form in these structures clearly on
different growth rates in different
areas seems to produce interesting
variation and contrast of likes
moving us in detail and also the
structures often have I think are
potentially eerily interesting
structural properties things like
grooves like bark which potentially are
really strong when they’re fabricated so
one thing I’m learning from these
experiments is how powerful this idea of
emergence is the idea of creating things
through generative systems and
influencing the results another big
thing I’m learning is that you often
need to tread quite carefully it’s very
much a process of learning a process of
discovery finding out what happens and
appropriate ways of influencing with the
results complex systems are capable of
doing amazing things there’s also in
their nature to do the unexpected
of course this can also be great leading
you in new directions that you never
imagined were possible thank you
[Applause]
[Music]
Please follow and like us: