OC96_WHAT DO YOU WANT, RENDER ENGINE [FIELDWORK – PLAN]

SEBASTIAN FATHI / MPHIL IN ARCHITECTURE AND URBAN DESIGN 2020-22

UNIVERSITY OF CAMBRIDGE

FIELDWORK OVERVIEW

The contemporary architect primarily uses optical tools to design, which implicitly become proxies for the worlds in which they attempt to represent. How do these tools command their own perceptual influence on design output?  The core of the fieldwork questions how site photography’s proxy optics (models, sites, cameras, render engines) influence and persuade designers to make design decisions, and how these implicit practices can be understood versus astro-imaging’s similar, but crucially explicit, optical practices. Both fields deal with the construction of reality. Despite my thesis and design proposal work centring around the Very large Telescope in Chile, this site will not be possible to visit for fieldwork due to the pandemic (as discussed with the VLT operations manager). Therefore, I will be using Death Valley USA, Joshua Tree USA, and the Mojave USA as site proxies, identified using an AI tool that detects similar geographies using satellite optics. I will spend two weeks in total over these sites, deploying physical studio models and photographing them versus shot lists comparable to astro-imaging survey methods. The collected photographs will record the aberrations and mistranslations between model and site, eventually resulting in design persuasions between model, site, camera and designer. The specific optical conditions of the proxy site and its photography are key versus the Cambridge studio environment. Multiple sites aim to generate a more reliable range of variables to the Chilean site, and the length of time over the sites balances the depth of photographic studies with budget. One further week will be spent photographing and writing on built architecture within these sites that have, by virtue of being built, transcended optics into the realm of kinaesthesia and physical site relationships, to reveal the possible kinaesthetic reality of my optical studies for my own proposal. Costa Rica is a required 14-day necessity for travel to the US, where I will remotely be conducting interviews with the building and astronomy industry in Chile within similar time zones. One night in New York will be to interview and see the work of a visual artist who explores these thesis themes.

PREFACE

Despite the design project being sited by the VLT, we will spend no time in Chile at the VLT. From correspondence/interview with the operations manager at the VLT, there are institutional blocks due to COVID, health risks due to the extreme environment over long periods of time, poor access, lack of accommodation, and the language barrier. Also, Chile is no longer admitting foreigners at the border (at least during fieldwork planning months at this time of writing) due to COVID.

Therefore, fieldwork will be conducted in three proxy sites, found in SW USA:

  1. The idea of site proxy is a major condition of architectural workflows. Conventional architectural proxies include site models, site photography, scaled site drawings, and the studio environment more generally. By using a 1:1 site proxy I am able to directly engage this further condition of architectural optical workflows. The benefits include a study into proxy translations that should become key to my design brief.
  2. It is important to note that, as mentioned, the backbone to this research depends on an understanding and simulation of astro-imaging workflows. One specific component of this workflow is that of site detection. Astro-imaging uses smaller survey telescopes (to the primary observation telescopes) that survey the sky, looking for celestial subjects similar to the parameters of the scientist’s request. The collected data is filtered through an automated software that identifies and presents these subjects to scientists for further study with larger telescopes. 
  3. As my goal is to learn from these methods, I will be adopting a similar method. I’m using a simple piece of AI that can identify areas of land using Google Maps data that are similar (in pixel complexion) to a specified target piece of land.
  4. This proxy workflow is essentially a form of optical subject analysis found in astro-imaging, for the sake of gathering a range of data and information on similar subjects that can be informative to the research goal. Survey telescopes detect a range of celestial subjects with similar software methods.
  5. These proxy sites are on average 1000m altitude, as opposed to 3000m at the VLT.
  6. Less of a cultural/language barrier.
  7. Greater accessibility to general resources such as food, water, supplies, emergency services, etc.
  8. Allows for longer term site work.
FIG.1: Descartes Labs Geo Visual Search, AI tool to detect similar geographies (proxy sites) across google maps.
WIDER THESIS RESEARCH QUESTIONS
  1. To what extent do optical design tools persuade architects to make design decisions outside of the kinaesthetic context of architectural space, and to what extent are these persuasions implicit or explicit?
  2. How does optical architectural design compare to the explicit optical tools, proxies and methods used in the similar field of astro-imaging?
  3. What can this comparison to astro-imaging infer regarding the persuasive nature of optical tools on architectural design?
  4. If there is a gap between optical design tool persuasions and kinaesthetic experience, how can these tools become consciously generative and integral to the design of the spaces they attempt to investigate and represent? 
FIELDWORK RESEARCH QUESTIONS (FWQ)
  1. How do versions of architectural site photography compare, in terms of their attempt to be proxies of their subjects?
  2. To what extent can explicitly objective methods of site photography in astro-imaging transfer to architectural site photography?
  3. What can be revealed regarding the persuasive nature of site photography on an architectural proposal’s reality(ies), by practicing explicitly objective astro-imaging site-survey techniques in architectural site photography?
FIELDWORK METHODS OVERVIEW
QUESTION (FWQ)METHOD OVERVIEW (SEE TIMETABLE FOR DETAILS)TIMING
1-ARCHITECT INTERVIEWS TO UNDERSTAND THE PROCESSES ARCHITECTS USE TO CONDUCT SITE PHOTOGRAPHY
– LITERATURE REVIEWS FOR HISTORICAL CONTEXT TO THE ABOVE.
– IDENTIFICATION OF SITE REALITIES FOUND WITHIN BOTH STUDIO PROXIES AND THE GEOGRAPHICAL SITE BY EXPLORING SIMULATION TECHNIQUES IN RENDER ENGINE, MODEL PHOTOGRAPHY AND DRAWING REPRESENTATION OF THE SITE.
– RECORD DESCRIPTIONS OF THE SITE (TEMPERATURE, WIND, INSTABILITIES, TIME, ETC) TO GIVE CONTEXT TO THE LIMITS OF SITE PHOTOGRAPHY/SIMULATION AS A SITE REALITY IN TERMS OF GEOGRAPHIC CONDITIONS.
-INTERVIEWS WITH THE RESEARCHERS AND LOCALS WHO HAVE SPENT LONG PERIODS OF TIME AT THE VLT/AREA, TO GIVE CONTEXT TO THE LIMITS OF SITE PHOTOGRAPHY/SIMULATION AS A SITE REALITY IN TERMS OF SOCIAL/KINAESTHETIC CONDITIONS.
AUTUMN 2020 – AUTUMN 2021
2– ASTRONOMER INTERVIEWS TO UNDERSTAND THE PROCESSES ASTRONOMERS USE TO CONDUCT SITE PHOTOGRAPHY
-LITERATURE REVIEWS FOR HISTORICAL CONTEXT TO THE ABOVE.
-REVIEW THESE PROCESSES VERSUS THE ARCHITECTURE INTERVIEWS IN FWQ:1
– RENDER ENGINE IN STUDIO  TO SIMULATE TYPES OF OPTICAL TOOLS, PROXIES AND METHODS FROM ASTRO-IMAGING
– SITE PHOTOGRAPHY IN-SITU USING IDENTIFIED ASTRO-IMAGING SITE SURVEY TECHNIQUES (THIS WLL COMPRISE MOST OF THE FIELDWORK).

WINTER 2020-WINTER 2021
3– REVIEW AND COMPARISON OF SITE PHOTOGRAPHS IN STUDIO VERSUS THEIR COLLECTION METHODS
– INSIGHTS ON OPTIC PERSUASIONS TO BE FED BACK INTO THE DESIGN DEVELOPMENT (METHOD NOT INCLUDED IN THIS FIELDWORK REPORT, AS IT IS A STUDIO BASED ANALYSIS THAT BUILDS ON FIELDWORK FINDINGS).
– REVIEW SHOULD BE UNDERTAKEN AS AN ASTRONOMER WOULD.
AUTUMN 2021-SUMMER 2022
FIELDWORK TIMETABLE AND LOCATION OF METHODS

Click here the timetable dates

Fieldwork method drawings (used to describe instructions and record the methods visually) are in progress.

Terminology: PM = Physical Model, MP = Model Photography, RR = Render Rig, RRO = Render Rig Output

WHEREAIMMETHOD STRUCTURE OF DAYS, PER TIMETABLERESOURCES
COSTA RICAWIDER DESIGN/THESIS AIMS WHILE COVID ISOLATING. (NOT PERTINENT TO FWQs PER SE -AS THIS IS AN EXCEPTIONAL CIRCUMSTANCE).– REQUIRED 14 DAY VISIT FOR TRAVEL TO USA, CONDUCTING INTERVIEWS WITH THOSE IN THE SAME CENTRAL/SOUTH AMERICAN TIME ZONE (CHILE) AS PART OF BROADER THESIS AIMS

– REMOTE INTERVIEWS WITH PLANNING AUTHORITIES OF ATACAMA REGION

– REMOTE INTERVIEWS WITH LOCALS OF VLT REGION (WIVES OF PINOCHET ERA AND THEIR DESERT EXPERIENCE)

– REMOTE INTERVIEWS WITH BUILDERS OF ATACAMA REGION (INSIGHTS INTO LOCAL SEISMIC STRATEGIES, RAMMED EARTH CONSTRUCTION, GC WORKS IN REMOTE REGIONS).
– INTERVIEWS WITH PLANNERS FIRST FOUR DAYS

– INTERVIEWS WITH LOCALS NEXT FOUR DAYS

– INTERVIEWS WITH BUILDERS LAST FOUR DAYS

– WRITE UPS IN BETWEEN

– TWO DAYS BUFFER FOR TRAVEL
LAPTOP, INTERNET, CONSENT FORMS
LOS ANGELESPICK UP CAMPERVANN/AONE NIGHTN/A
JOSHUA TREE
33.880500,-115.816428
(LOCATION IDENTIFIED USING AI TOOL TO FIND SIMILAR GEOGRAPHIES TO THE VLT SITE – A CONCEPT IN  ASTRO-IMAGING SURVEYING TO OBSERVE SETS OF SIMILAR SUBJECTS)
TO EMPIRICALLY EXPLORE HOW DIFFERENT TYPES OF OPTICAL DOCUMENTATION IMPART DIFFERENT VERSIONS OF REALITY ONTO PHYSICAL MODELS (PM), USING ASTRO’S CONCEPT OF ‘THE SIGNATURE OF THE INSTRUMENT’, ‘LASER GUIDE STARS’  AND ‘SIMULATED SURVEYS’ (FWQ:2). THESE OPTICAL DOCUMENTATION COMPARISONS ARE BETWEEN: THE DIGITALLY SIMULATED PM VERSUS THE DIGITALLY SIMULATED 1:1  PROPOSAL; THE SIMULATED SATELLITE’S UNDERSTANDING OF THE 1:1 PLAN LAYOUT ON SITE VS THE PHOTOGRAPHED PM ANAMORPHICALLY OVER THE SAME SITE,  THE PHOTOGRAPHED PM VERSUS THE DIGITALLY SIMULATED PM (FWQ:3)1.500 MODEL

1- COMPARE SIMULATED RENDERED 1.500 PHYSICAL MODEL (PM) TO THE  DIGITAL 1:1  IN ALL FOUR ELEVATIONS (SEE FIGURE 2). THIS IS PRELIMINARY  WORK TO IDENTIFY WHERE THE CAMERA AND MODEL SHOULD BE POSITIONED (AND KEPT IN IDENTICAL POSITION FOR ENTIRE METHOD VS GPS AND DRAWINGS) IN THE FIELD AS PER THE SIMULATIONS, BUT ALSO AIMS TO UNDERSTAND THE DIFFERENCE BETWEEN THE RENDERED PM AND THE DESIGN 1:1 PROPOSAL.

2- MARK THE PROPOSAL PLAN ONTO THE  SITE AND PHOTOGRAPH. THE PLAN SHOULD BE PREDETERMINED GPS POINTS, SHOULD DESCRIBE THE POINTS OF THE PROPOSAL ENVELOPE, AND SHOULD BE THE PLAN AS UNDERSTOOD BY THE SIMULATED SATELLITE IN  PRIOR DESKTOP STUDIES

3- MARKS MADE USING ONE HANDHELD POLE. THE POLE SHOULD BE MOVED AROUND THE SITE, AND AS WILL TAKE PHOTOS OF THE LOCATIONS USING A DX CAMERA AND TRIPOD FROM ONE PREDETERMINED LOCATION. ALL SHOTS WILL BE COMBINED IN POST TO FIND A SINGLE  ENVELOPE.

4- POSITION THE 1.500 PM AS PER 1-, AND PHOTOGRAPH. DESKTOP WORK TO COMPARE THE POSITION OF THE PM IN PHOTO TO THE  GROUND MARKS IN  1)B. DO THIS FOR ALL FOUR ELEVATIONS.

5- WITH SF USING THE VIEWFINDER, AND WITH THE PM IN PLACE AS PER 1-, HAVE AS RECORD GPS LOCATIONS OF THE MODEL ENVELOPE VIA ANAMORPHIC OVERLAY ONTO THE SITE. REPEAT THIS STEP WITH THE MODEL REMOVED, USING THE POLE METHOD AS MENTIONED ABOVE, TO UNDERSTAND THE VISUAL/PHOTOGRAPHIC COMPOSITION OF THE NEW ENVELOPE DIRECTLY ONTO THE SITE. 

6- MEASURE THE DIMENSIONS BETWEEN THE ORIGINAL MARKS AND THE NEW MARKS USING GOOGLE MAPS

1.20/50/100 MODELS

7-REPEAT FOR FURTHER 1:50 MODELS OF INTERIOR SPACES
TO BE CONDUCTED OVER ONE WEEK. 1.500 MODEL METHOD IN THE FIRST 3 DAYS, REPEAT FOR THE 1.20/50/100 MODEL METHOD IN THE LAST THREE DAYS. TRAVEL ON DAY 6/7 TO NEXT SITE. PHOTOGRAPHS TO BE TAKEN DURING DAYLIGHT HOURS, AND WHERE POSSIBLE, REPEATED DURING DAWN/DUSK HOURS FOR LIGHT STUDIES.DESCARTES LABS GEO VISUAL SEARCH, GARMIN GPS, DX FORMAT CAMERA, PHOTOSHOP, AS, (2)POLES/MARKERS, SET OF PRINTED PLANS, MASKING TAPE, UHU GLUE, LASER MEASURE, MEASURING TAPE, PENS/PENCILS, A3/A4 PAPER, CLIPBOARD, LAPTOP, INTERNET, HARD SUITCASE FOR MODELS WITH FOAM INSERTS, CAMPERVAN, (2) TRIPODS
MOJAVE
35.708888, -116.306392 
REPEAT JOSHUA TREEREPEAT POINT 5- OF JOSHUA TREE ONLYTHIS WILL BE A ONE DAY VISIT FOR TIME EFFICIENCY.REPEAT JOSHUA TREE
DEATH VALLEY
36.801908, -117.203233 
REPEAT JOSHUA TREEREPEAT JOSHUA TREEREPEAT JOSHUA TREEREPEAT JOSHUA TREE
PRECEDENT VISITS: 
A-Z WEST CA, NANCY HOLT SUN TUNNELS UT, RODEN CRATER AZ (TBD), LIGHTNING FIELDS NM, SOM’s KITT PEAK AZ,  
TO UNDERSTAND THE 1:1 RELATIONSHIP OF OCCUPANT TO CAMERA TO BUILDING TO SITE, AND THUS INFER (BACKWARDS) THE SPECULATIVE RELATIONSHIP OF MY OWN PROPOSAL TO ITS OWN SITE AND OCCUPANT, VIA THE CAMERA (FWQ:3). A STUDY IN THE KINAESTHETIC. PHOTOGRAPHY OF 1:1 PHYSICAL BUILDINGS (PRECEDENTS) WITH SIMILAR RELATIONSHIPS TO SITE CONTEXT AS THE PROPOSAL

SUCH PHOTOGRAPHY OF PRECEDENTS VERSUS A RESOLUTION TARGET(S) (METHOD BORROWED  FROM ASTRO/SATELLITE CALIBRATION TECHNIQUES). EACH PRECEDENT SHALL BE PHOTOGRAPHED AS ELEVATIONALLY AS POSSIBLE (TO MIRROR THE SHOT LIST OF THE PROPOSAL ITSELF), AND EACH SHOT WILL BE DONE TWICE TO INCLUDE THE RESOLUTION TARGET(S). THE RESOLUTION TARGET WILL BE AN ISO 12233 TEST CHART (IN BOTH REGULAR 2D AND ADAPTED 3D FORM), AS WELL AS THE 1.20/50/100 FRAGMENT PM, APPROPRIATELY SCALED TO ANAMORPHICALLY MATCH THE 1:1 SCALE OF THE PRECEDENT SUBJECT.

AS THIS IS A MORE SUBJECTIVE UNDERSTANDING OF THE REALM OF THE KINAESTHETIC, EACH PHOTOGRAPH/SET-UP WILL BE JOURNALISTICALLY WRITTEN UPON IN PROSE IN-SITU, TO GAIN A PHENOMENOLOGICALLY DESCRIPTIVE INSIGHT INTO THE RELATIONSHIP BETWEEN OCCUPANT TO CAMERA TO BUILDING TO SITE.

THREE DAYS OF PHOTOGRAPHY/WRITING, A-Z VISITED DURING JT VISIT, VISITED IN ORDER AS LISTED AFTER DEATH VALLEYCUSTOM RESOLUTION TARGETS, REPEAT JOSHUA TREE RESOURCES
UOA TEXAS, AUSTIN – SARAH OPPENHEIMER’S BRIDGE INSTALLATION OPENINGTO UNDERSTAND THE 1:1 RELATIONSHIP BETWEEN INSTALLATION TO SITE TO OCCUPANT IN THE CONTEXT OF AN EXPLICITLY OPTICALLY DESIGNED SPACE. PRIOR PRECEDENTS WERE NOT DESIGNED IN THIS WAY. SO’S WORK IS EXPLICITLY PERTINENT TO THESIS AND COMPRISES A CONTEXT OF SIMILAR/CONTEMPORARY THESIS THEMES.  REPEAT PRECEDENT VISITS. THIS WORK WILL BE CONSOLIDATED VERSUS INTERVIEWS WITH THE ARTIST TO FOLLOW.ONE DAY OF PHOTOGRAPHY/WRITING REPEAT PRECEDENT VISITS
NYC – SARAH OPPENHEIMER’S STUDIO IN LONG ISLAND CITYREPEAT UT TEXAS VISIT, TO UNDERSTAND THE OPTICAL DESIGN METHODS  OF THE ARTISTINTERVIEW, TOUR OF STUDIO, WRITE-UP OF TOUR AND FINDINGSBOOKABLE SLOT (APPROX 1HR) OVER ONE DAY INTERVIEW CONSENT FORM, PERMISSION TO DOCUMENT, PERMISSION TO PUBLISH DOCUMENTATION, A/V RECORDING DEVICES TBD ON PERMISSIONS
FIG.2: 1.500 physical model in situ, rendered. Simulates physical set-up on site. Red background is the 1:1 proposal on site, of which the foreground model aligns to as best as possible, visually, in render space. 1.500 model to be placed between camera and site to simulate this render. Alignment to camera and site is based on grid marks projected onto the model that align with the camera’s 16 square-grid viewfinder as simulated above using a custom render camera designed to simulate a DX format sensor DSLR with 35mm focal length.
ADDITIONAL NOTES ON METHOD
  • Transporting these tools/resources to sites
    • These tools will be able to fit into one backpack and one hard suitcase for easy transport.
  • Being able to repair/maintain these tools
    • These optical test subjects and models will be made from laser-cut plastics, greyboard, and 3D printed PLA plastics. They will be designed with endurance in mind, however, repairs to them can be made with glue and tape. As a backup, they can be somewhat reconstructed from card and craft supplies. Drawings will be made to ensure this, prior to my departure.
  • Not damaging the environment with these tools
    • The only contact these tools will make with the ground will be via a tripod. These are low impact tools that require no damage mitigation.
  • Not disturbing areas associated with the indigenous/natives/private.
    • Maps of such land will be made before arrival to any site, for the sake of avoidance.
  • Risk assessment not included in this fieldwork description (only included in Cambridge’s Leave-to-work-away forms).
BUDGET

See budget here.