First test results

Last week I’ve done the first test. I’ve made an animation of my design, and showed it, and four different pieces of music, to a number of four people, and asked them to guess which piece would have been the basis for the architecture.

Surprisingly, three out of four chose the same piece! But, they all chose the wrong piece. Which is quite interesting, as the chance of three people choosing that same number, is about 1 in 20.

The pieces were:
Dave Brubeck – Take Five
Beethoven: Mondscheinsonate
Miles Davis: Summertime
Chopin: Nocturne Op. 9 No. 2

Three out of four people have chosen Take Five. The fourth chose Summertime. Both are jazz songs, while the beethoven music is classical.

Based on the argumentation for their choices, I’m changing the ‘rules’ I’ve used for the design. Most prominent elements for choosing the jazz song, is the harsh diference in lighting, and the complete randomness of the columns in some places, associated with improvisation. Also, the composition was perceived as too full, associated with a higher tempo. These aspects will be changed for the next model.

Time for testing

It’s been a while since my last update.

I have designed the space, and modeled it in Maya. The lighting is still a problem, I can’t get it the way I want it.

Rudy has given me some helpful pointers on how I might improve my interpretation. The volumes decribing the melody aren’t as eyecatching as I’d imagined, this could be improved by making them bigger, but that would limit the supposed usability (I’m imagining it could be some sort of exhibition room). Colour might help to make them stand out more. 

An other pointer was the bass line, in the floor. The height difference could be to local, to much focused around the theme objects. Maybe it should be sloped more gently.

Though I think these changes would improve the design, I will do the first test with the current model.

I will finish the lighting as soon as possible, and then go for it. I’ll upload the animation later.

translation

I wasn’t too happy with my presentation last Monday. The presentation itself should’ve been more clear on what I have tried and didn’t like, instead of skipping it entirely. Because I didn’t really show anything, visually, it was difficult to comment on it.

Today, the goal for today is to design a translation, based on a set ‘ method’, which for now only exists in my head. Though it may change during the modeling, I will write a short description, explaining how and why I link the specific elements.

The piece is built on four aspects. First, there is the motif, which runs through the entire piece, stringing the different parts together. It stays the same for a long time, but at the end and around the middle of the piece there is a distinct variation.  The motif is something which is quite regular throughout the piece/space, and becomes dominant when it is varied upon. In architecture, this can be associated with a pattern of windows, columns, etcetera, for example the repetition of window frames in our studio. Though it’s quite obvious, it doesn’t catch the eye. But if someone were to take out several frames, and put in something on big frame, that specific change would certainly be more apparent. In the translation, I will use a grid of columns. thin ones, so they are visible, but won’t dominate the space (hopefully). Then, where the musical motif varies, the grid will be altered.

The second element is the theme/melody. The melody is the most pronounced in the music. It’s what one hums when one hums music. It is also not a constant line, but several somewhat separated parts. The melody in archtiecture can be several volumes, objects, in space, which are the eyecatchers, the things we remember. The exact form and size of the volumes is a difficult subject. I hope to figure this out during the modeling.

The third aspect is the bass line. The bass line accentuates the themes, and variations in the motif. In the architecture, it should do just that, accentuate variations in the column grid, and the theme objects. This could be done with a small, but noticeable, change in floor height.

The final aspect is the dynamics, the loudness of the music. The piece starts really soft (pianissimo), then gradually gets louder (crescendo) and softer again (decrescendo). This happens several times. When the notes are played louder, especially the higher tones which form the melody get more pronounced. In architecture, light has a similar effect. Hard direct light makes shapes more distinguishable then soft diffuse light. So as the music varies in loudness, the space will have a varying lighting.

Some of these aspects aren’t quite clear yet, but they should, during the modeling.

Mondscheinsonate

Hypothesis: When architecture and music are based on similar formal rules,  this connection between the two stays recognisable.

The idea is to take a specific piece of music, which is composed on a set of rules. These rules will be analysed, and an architectural form will be based on the same set of rules. In order to somewhat transfer the mood of the music, determinative elements will be selected as parameters, which influence several architectural elements. This relation will not be a direct one, as two elements will not have the exact same meaning in music and architecture, but serve as an indication.

To test the hypothesis, a questionaire will be used, in which people are presented the design, and have to choose from a number of musical pieces the piece on which the architecture has been based.

The selected piece of music is the Moonlight Sonata, by Beethoven. The sonata form has a strict set of rules, which make it suitable for a translation. A further analysis on the sonata form and the chosen piece is needed, in order to specify the determinative elements.

stop translating, you fool

Today I’ve been trying to use the first three (or so) seconds of Für Elise as a base for a composition of reliëf and light in a facade. It’s based on the pitch and length of the individual notes. The facade is first devided into two, as the music is a pair of almost identical groups melodies. Each of those melodies consists of three parts: one string of short notes with a longer one, and two, again almost identical, smaller groups. Then, each part is devided again, so that every area is a representation of the length of an individual note. The reliëf is based on the pitch of the notes. A lower tone lays deeper into the facade, a higher tone jumps forward. The wall opposite to the composition emits light, so higher tones are brighter. The entire process is done a second time, with a slightly different division. The third one is the most literal of all.

The scenes where then rendered (in SketchUp, with Podium). The results is are interesting pictures, which otherwise don’t appear to have any relation with the music, except maybe the rhythm which is somewhat noticable. 

So what do we learn from this? When one has the music, he might be able to recognise the link. Or he may not. Either way, it doesn’t really ad anything. It confirms that this (above) isn’t the way to approach this research.

I’ve been taking the recognisable part to literally. I don’t want to make architecture in which people feel ‘oh look, it’s like a sonate of Beethoven’. After all, baroque art and architecture don’t have such an explicit link either. What they have in common are the underlying principles.

So what I am looking for is not a translation, maybe not even an interpretation, but a way to extract those principles and based on those principles, create some architectural space. The link between the architecture and the music then won’t be as clear as I originally anticipated, but when compared to the same space based on other principles (different music) there should be a noticable difference.

the scale of light

The last couple of days I’ve read some of the literature suggested by Truus, mostly about light zones and the ‘scale of light’. Especially the scale of light intrigues me, as it’s set up, a scale, resembles a tonal scale, or, associated with poetry, the different vowels. This reminds me of an example a friend of mine gave me on the connection between vision and sound:

One is presented two doodles of characters. One is pointy, like a star or sun, while the other is round. When asked which of both would be Kiki, and which one would be called Bobo, about 90% (if not more) of the people choose the sharp character is Kiki (keekee), and the round one is Bobo. This is because the sound of the ‘i’ is considered a sharp sound, while the ‘o’ is a round one.

The same might apply for light. Using the same names, I personally would associate Kiki with hard shadows and direct lighting, while Bobo would be more diffuse and soft shadowed. Or in music, Kiki would be a higher tone. I guess the i in Kiki is indeed a higher tone than the o in Bobo. A higher tone would be consistent with a stronger shadow. Continuing on the lighting, a louder tone would be associated with more luminancy, while a softer tone would be less. In a spatial form, for example a curving wall or a series of spaces, a louder tone could be associated with more reliëf, or a larger space.

Though I do not intend to end up with a literal translation, I might try to do this with a light scale and music, just to see what the visual effect will be. It would certainly make a cool windows media player visualisation.

 

Considering the previous post about Xenakis, a good way in using music/poetry could be to analyse it’s basic elements (rhythm, pitch, etc) and use those to make an architectural form. Not as a direct input-output relation, but as two separate works, based on the same underlying principles. The architecture for instance wouldn’t have the exact same  sequence in pitch, but it could have a similar distribution in high and low tones and abrupt or fluent transitions between them.

A problem which I haven’t mentioned yet, but which I have discussed with my instructor Rudy, is the element of time. In music and (spoken word) poetry, time is an important factor. In architecture it is not, as it is static mostly. A solution could be to make the architectural experience dynamic. A series of spaces, which is experienced in a sequence, can introduce the element of time in architecture. A continuously transforming space can do this as well. But when introducing the element of time in the architecture, there is also the problem of ‘freedom’. When one has time in both architecture as music, it will be more difficult to make architecture which is inspired by the music, rather than a direct translation of it.

density

I’ve read an interesting article about music and architecture. Iannis Xenakis was a composer as well as an architect, and experimented with numerical proportions in music and architecture. For the facade of the monastery of la Tourette he designed the glass panes in the facade. By using a set of window widths, he ran into a problem:

too limited a number of elements results in an arid and predictable composition, too many elements make it impossible to aesthetically control the resulting configurations.

…. he considered the problem on a more general level, above the individual elements, by replacing the concept of rhythm by that of density (in the ense of ‘number of events per time or length unit’). Rather than considering the individual distances between the upright casings, he now demarcated zones in the facade where a higher or lower number of casings per length unit would be rewuired and then decided how the transition between these two states would occur: fluently or abruptly. - Muecke, Resonance: Essays on the Intersection of Music and Architecture, p25-26

This compostion wasn’t directly related to a specific piece of music, but more to the underlying principles of composition of music and architecture. I will do a small test on what happens if I try to base it on a specific piece of music. (I probably won’t come anywhere near my research goal, but the idea of linking density to rhythm might prove useful)

research proposal

Hypothesis:

the mood of music (and poetry) can to some extent also be achieved by means of architectural devices such as (but not limited to) composition of space and light.

 

I will do some background research into all elements of architecture, music and poetry which I believe are important in setting the mood, and explore ways to ‘translate’ these from music to architecture. Translation of music to architecture has been tried numeral times by various architects and designers. Usually, an algorithm is used to translate these 1 on 1. By doing so, the goal of my research, translating the mood, is lost, as the link between the music and the architectural form is only recognisable when one knows the algorythm.

The method I will try to develop is not so much focused on a 1 on 1 literal translation, which can be translated back again, but focuses more on the, let’s say, interpretational side. The architectural form will be inspired by the music or poetry. The result will therefore not be a strict formula which acts as a black box that automatically creates a form based on an input, but a set of guidelines on how several elements might be ‘translated’ into architecture. The term translation should maybe be replaced by a term such as transformation or interpretation. During this research, I will primairily be focusing on composition of space and light, because of the limited amoud of time, but other devices won’t be thoughtlessly discarded or ignored.

 

To test the hypothesis, an interpretation method will be developed, which will be used on ideally several examples, say seven. Through a questionary, people will be presented the seven inputs and outputs, and asked to sort out the pairs. Then, the average score will be calculated and this will be used to determine how succesful the test is. A low score, up to 3 out of 7, can be achieved by random guessing, 4 is already a lot harder to guess, and chances on 5 or 7 out of 7 are very slim. 

Probably there won’t be enough time to do more than one interpretation. To test this, people can be showed the output, and asked to select the right music/poem out of a set. Or vice versa, the music/poem can be showed, and the corresponding architectural form has to be selected.

things I HAVE to read

A few titles of books I think need to read (partly), or at least look at.

Resonance: Essays on the Intersection of Music and Architecture -  Mikesch W. Muecke, Miriam S. Zach

Pamphlet Architecture 16: Architecture as a Translation of Music – Martin, Elizabeth

Poetics of Architecture: Theory of Design – Antoniades, Anthony C

Semiotic discipline in architectural design – Turrin, M

Music Notation and Terminology - Karl Wilson Gehrkens

Light-zone(s): as Concept and Tool – Madsen, Merete

literature

Today I’ve had a meeting with Karel. We discussed some ways to approach this topic. It is clear that I need to do a lot of research. I have to study how different effects in music are annotated. This will also lead to a better understanding of comparable elements in poetry. Simultaneously a lot of research into literature and precedents has to be done on the elements in architecture which can be compared to the elements in music and poetry. The basic ingredients of poetry/music have to be defined, and linked (when possible) to architectural elements. Once a more thorough understanding of these components is achieved, there can be made a selection on which element(s) are going to be researched more specifically.

So tomorrow will be a day spent mostly in the library.

I will also send an email to Alex Dragulescu, to ask about his method for translating text into ‘architecture’.

I’m a little bit lost in how to go on from this point. I’ve started on Michela’s text, which is very interesting, though difficult to fully comprehend. I’ve gotten two tutors assigned to my research, and if I understand the email correctly, on monday it will be decided which one of them will stay my tutor.

I’ve sent an email to both to make an appointment on friday.

presentation day

So today everybody presented their research proposals, and the instructors and tutors commented on the subjects, formulation of questions, possible approach, and others.

Because of André’s comment on that the earlier topic is more interesting to research, on which I agree, and though I feared it would be an item too broad to research in this course, I decided to switch back, and hope for the best.

Here’s a link to my presentation 

http://content.yudu.com/Library/A13lah/artinarchitectureyli/

I’ve gotten some useful comments on how to approach the task up hand.  

One was to avoid the infinite trial and error on translation methods. Instead I could perhaps start with a poem, and design the end result myself, the artitectural form, and then try to find a ‘system’ by comparing them. After that, it would be interesting to test if the system can be applied to other poetry as well.

Another was to concentrate on one specific element, for example rhythm, as the topic is too broad to research everything during this course.

I have some trouble recollecting all the comments I’ve gotten. Next time I’ll try to take an mp3 player to record everything, if that’s okay by the teachers of course.

perhaps a better research question

While listening to some spoken word poetry I realised, my question would probably be a recipe for failure. I can’t even distinguish a clear line in rhythm and composition when listening to it. Furthermore, the rhythm and composition aren’t the elements defining the emotion, at least certainly not the most important ones. I guess I was too preoccupied with the promise of a visualised poem, that I didn’t realise the magnitude of the problems I’d encounter. So I’ll drop my research questions and come up with some new ones, which are more doable and researchable.

 

Instead of trying to translate art to architecture, a more usable and doable research will be to test how the elements which architecture and art have in common can be used in architecture to communicate.

I’ve mentioned earlier how rhythm or rather a change in rhythm can accentuate a place. In a same way can it contribute to a different, let’s say feel, between spaces. I’ve also mentioned how proportion and composition of spaces influences the perception of that space. Through subtle (or less subtle) changes it can make a distinction between spaces. Separating them without physically putting a division between them. A common example is defining two separate area’s in a living room by raising the floor of one area a little.

An element which I haven’t mentioned yet is ‘klank’ (not sure of the English word, maybe timbre). In architecture differences in klank can be differences in lighting. Like some tones are sharper than others, light can differ from hard direct light to completely diffused indirect light.

 

Let’s say there is a program which needs to be placed. Each function has a need for a certain amount of space, volume, and there is a certain hierarchy in them.

• How can composition of spaces and rhythm and timbre be used to make a distinction between the several ‘areas’ without physically closing them from the others? This would focus on the perception of differences in composition and proportion between adjoining spaces.
• How can these elements be used to communicate the relationships between spaces? This might be done through the use of comparable rhythm and timbre in spaces which are related to each other.

A hypothesis would be that areas can be separated from  each other when there is a difference in height of ceiling/floor, lighting, repeating of elements like construction or openings, while at the same time these can imply a relation between area’s which aren’t related to each other in a topological way.

 

A way of testing could be to design a space where these instruments are used in the described way, and determine through a survey if people can recognise the several areas as separated, and which areas are related to each other.

 

This is still quite a broad topic. Hopefully tomorrow will help towards a more confined research question.

possible research question

Regarding André’s comment on how most translation/transcription/mapping techniques ‘fail’ from an over-obsession with notation, Axel’s comment on how challenging it is to translate the interpretational/emotional side of ‘art’ in a mechanical approach, but also Michel’s example of an apparently successful translation, I’d like to try this myself.

To try to come up with a translation method (or several, if that’s doable regarding the time span); make several translations,  and then see how ‘successful’ the method(s) is by testing if one can still link the output with the original input.

 

Is a translation of rhythm and composition of music/poetry to an architectural form sufficient for expressing the mood of the initial input?

• How can rhythm and composition be mapped/extracted? (is it better to analyse an audio track, or a notation on paper; what elements are determinative, since an analysis up to individual notes would probably go beyond being useful for a translation, and into the realms of obsession)
• How can the extracted data be used to create a three-dimensional shape? (what form is most suitable, for example a spatial arrangement, a sculptural mass, openings or relief in the planes enclosing a space; how do the separate elements defined in the first question effect the three-dimensional form, is there a hierarchy in more and less influential elements)
• Is the output of this method still recognisable as a translation of the input? (can you still link the three-dimensional form to the original piece of music/poetry; without seeing/hearing the input, is the overall mood of the input tangible in the output)

 

For researching these questions, a selection of either some pieces of music or poems will have to be made. In these pieces, rhythm and composition will have to have an important role, as there obviously are enough examples of poetry or music where rhythm is almost entirely absent. So they will have to be suitable for a translation.

Then there will have to be made a list of aspects on which they will be analysed, and a way to analyse them, preferably mechanical, like a script, though a manual method could suffice since it has to be done a limited number of times.

The greatest challenge is the translation from the derived data to an architectural form. This might be a script in Rhino or Maya, where the data on composition and rhythm are the parameters. Important here is that it shouldn’t be a literal transcription of the input. It isn’t supposed to be just another form of notation, which would be meaningless, but an object/space, which in based on/inspired by the initial piece of music/poetry. In the end, the goal is to achieve an architectural form, which should have a spatial quality. The translation therefore might not be as mechanical or straightforward as said, but more of a two way interaction, where the input doesn’t enter a black box which generates the output; there could also be room for reflection during the translation. On the one hand there is the data about composition and rhythm, on the other hand architectural ‘tools’ which can be used to achieve the certain atmosphere, and it isn’t possible to just say ‘this data means that this and this tool is used in this and this way’. As rhythm and composition are a part of a poem, it can only provide data for a part of the end result. The translation part needs a lot more attention during the rest of the course.

A way of testing the results would be a survey in which people need to match the output (renders or models) to the input (audio track or printed text). Before giving the input, matching the output to a certain mood might be interesting as well.

 

The research hypothesis would be that the mood/emotion of a piece of music/poetry which is expressed through composition and rhythm, is transferable to an architectural form.

 

Chances are none the methods will give a satisfactory result. Then there will have to be an evaluation on the shortcomings of the methods, the reason of failure.

 

An example of an input could perhaps be a poem in which there is distress, causing a hectic change in rhythm, overal shorter sentences, and short and long sentences close to eachother. Individual words, separated from the surrounding lines by a pause or a change in pitch or volume. Opposed to that is a more relaxed state of mind where there is a more even rhythm, less sudden changes in pitch, or length of sentences, less extremes.

In order to have a chance of being able to match the output and input, I’ll have to make sure the different pieces of music or poetry used for input are quite different from eachother.

note to self

It is clear that I am interested in composition, rhythm, and proportion, and the way these can be used in art and architecture. Now, how do I find a researchable problem on this subject, one that is suitable for using an analysis software, and doable in the time span of eight weeks.