Pages

Sunday, December 29, 2019

BACK IN DUBAI



Let’s start with a pretty picture that represents the model as it is right now.  Then we will recap over the past 3 or 4 weeks, back in Dubai.  It’s taken a little while because I went straight down with flu after the 24-hour, triple-flight back to the desert.  Also had to fit in AU Middle East which meant cutting my talk down from 1 hour to 20 minutes.  But in between all that the model has also progressed.







We upgraded to the latest version of Revit 2020, so the new “internal origin” icon pops up in all the views. (eye roll).  Meanwhile Francois now has access to the BIM360 hub so I arrived back to find he had updated several window families.   Also Alfredo had added one by Marc Zappia to the transept gable.  Looking forward to more work in that area Marc.







I started looking at the ZigZag vaults.  I am assuming that Alfredo will come back to these eventually and do a better version, but for the moment I just want to do a roughing out that captures how they grow out of the columns below.  Almost all our work is going through several iterations and these interlocking triangles are particularly difficult to figure out.  


I am testing a slightly different approach, taking a unit of 5 triangles, which repeats three times around the apse, and starting with the ribbed arches that make gothic vault technology what it is.  Adjust these until they sit properly upon the supporting columns and pilasters.  Make sure that there are arch/ribs to define all three sides of all 5 triangles.  Embed these nested families in a conceptual mass container file and start to add connecting model lines and surfaces using the “point world” tools.  





I am using a mass family rather than generic model adaptive because I want to host it on a level and have an offset parameter.


By the way, I have to share this cute little gif that Francois made while working on window tracery for the project.  





I also decided that the columns need an upgrade and while in TruView, noticed that there is one with a different form along the aisles as you enter the chancel.  It’s basically square with clusters of ribs at the corners.  This was a chance to have a go at a stone-by-stone analysis.  It’s not something we can consider for the whole model, but maybe just this once.










Then came AU Middle East, which was really great, and an image that I shared in the build-up to that.  This shows the zig-zag vaults pretty much complete (for the current iteration)  The ribs have lost the colour coding which emphasised the hierarchy of the different arch types. That was just a temporary move.







Just to recap, I used a three-point adaptive to define the curved ridges where three curved surfaces of the triangular vaults meet.  This is my idealised version of these vault types.  In  practice the masons seem to have improvised a bit from vault to vault.  The surfaces themselves are 9 or 10 point adaptives which rely on the lowest points being at an intersection of two arches that touch each other quite precisely.  I didn’t achieve that level of perfections, but it’s not too bad.






There are two more types of “ZigZag”.  Both have three triangles, but the geometry is slightly different.  The strategy for covering the arches 3 point ridges and multi-point surfaces is identical.  






So back to the columns, and my first attempts to develop some elaboration around the capitals. I’m starting with just two variations, the reality is much more than that, but two will at least give us the impression of different treatments.  One is based on a crocket motif, the other is more a series of leaves.  Both of those themes are prominent in the Notre Dame capitals.







I’m alternating these two capital types down the length of the nave for the moment.  I think the extra detail definitely adds something the nave elevations.








A bit of work on the ribbed arches into the side chapels and adjustments to neighbouring elements to try to get all the “rib-branches” sprouting from the symbolic “tree tunks” of the columns and pilasters below.







I ventured into the realm of material textures, feeling that the blandness of the cream stone we have been using so far was overdue for an upgrade.  I wanted to give the feel of coursed stone.  It’s a bit too regular right now, but we will come back to that.  Also wanted to go for more of a yellowish grey, slightly streaky stone which would be closer to the original.  Personally I think it’s a big improvement.





Compare these images from earlier in the same weekend.  The illustration from Hunchback of Notre Dame is there because Alfredo had bought a lavishly illustrated version of that novel and I was sharing some of the earlier graphics.  I think this one may be by Victor Hugo himself.  He was quite an accomplished artist in this kind of atmospheric, almost expressionist mode.





More views of the model in its current livery.  I prefer the colder stone ambience that it has acquired.  The light shafts that Enscape3d provides look great also.  The scene has a certain gravity now.  Somewhat lacking in scale because Daniel has removed the people (I think)  He has a plan for how to populate the scene in a better way apparently.  So let’s see how that develops





He asked for some examples of how Viollet Le-Duc drew people, so I hunted down a few from the download of his dictionary that Alfredo found.  You have to love the way that architects stylise the human form to blend in with their creations.  He definitely had that “medieval feel” nailed down. 







Sunday, December 22, 2019

AU 2019. 02


How do I explain that AU exceeded my expectations? To be honest I had some misgivings. Vegas is a strange place and huge gatherings of people don’t excite me. Plus... I am suspicious of technology hype/worship, which is inevitably part of such events. But for me AU2019 was an overwhelmingly positive experience, partly because of things I learned/sessions I attended, but mostly for the interactions with people: old friendships reinforced, new friendships begun.


Day 2 began with rain, which made the short walk from my hotel more interesting. On my phone, some messages to remind me that my presentation had made an impact. 




I attended the AEC keynote. Tantalising to see an animation from the new point cloud that Autodesk helped to organise for the French Government. I doubt that we have any chance of gaining access to this, but still live in hope that we may get our hands on the scan done by Andrew Tallon … one day.  It would be useful to cross-check and adjust our setting out.  Possibly we could use it to generate low-poly mesh objects of some of the sculptural features.  Ideally we would find a way to bring these into Revit with mapped bitmap textures to represent the finer detail.  This is something that both Archvision and Enscape3d have mastered so we know it can be done.






Possibly the highlight of Nicholas Magnon’s presentation was his interaction with Spot, the robot dog created by Boston Dynamics.  I had broken Twitter silence a couple of days earlier.  I have removed Twitter and Facebook from my daily routine, but the brevity of a Twitter post remains useful during a conference.




I guess you could spend the whole three days wandering around the exhibition floor.  I thought I had got the balance about right, but skimming through the list of exhibitors on the way home, I realised that there were lots more booths I should have liked to visit.  It was good to run into Anthony Hauck on the last day.  Allegion & Enscape are well established partners.  BIMtrack is a potential portal for linking our model to historical data.  Janice was a chance meeting while grabbing lunch and a fascinating conversation.  Mental Canvas is a digital pencil I would love to play around with.  Oh to be super rich.






Alfredo nagged Ryan into doing a LinkedIn post about his organ family.  I didn’t realise that he had set it out with “sacred geometry” in mind: nicely done Ryan.  According to Wikipedia, the organ is very ancient in principle, but in its “present form” dates back to the 14th century.  Certainly this design looks classical to me.  I wonder what preceded it at Notre Dame de Paris.  (get your phones out)  Well it seems to be the third organ, part of Viollet-le-Duc’s restoration in the nineteenth century.  The first was built in 1403, some 240 years after construction of the cathedral began, and almost 60 years after its completion.




A standard feature of any conference is eating too much.  At the height of my “recovery from diabetes” I managed to be very disciplined for a couple of conferences in a row, and cut out alcohol completely.  I should have been more disciplined at the grab & go lunches in the Expo, represented here by salad and beer. But when you are invited for dinner with a group of friends, (half of whom you last saw in Tuscanny 18 months ago and half who you never met before) … discipline is not the order of the day.  The Volterra Reality Capture workshop was an experience of a lifetime for me and it was great to hang out with these guys on Wednesday night.




On the final day I attended a fascinating session on Topologic.  I don’t think I can begin to do it justice, but it reminded me of the “wooden block” model that I had made of Notre dame and hinted at how this could evolve in a rule-based way if the blocks were aware of their relationships to each other.  I had bumped into the Fulmax guys at the Unity stand on the previous day and we succeeded in getting the Notre Dame model up and running in their VR cave.  At some point I joined up with Paul and some of the other Volterra guys to do short video interviews about our BIM/heritage work.  





Then the day concluded with Alfredo’s Lab on gothic vaults.  As expected he was smooth as silk.  Full class, great response, finished on time.  Lab assistants’ selfie at the end.  Privilege to be part of that group. Paul’s xmas card (British translation of Holiday) which features Notre Dame and came home to Dubai with me, via Florida.





I opted to fly out that night, but had a couple of hours to hang around.  Ryan had posted on linked in about his chandeliers, giving insight into his approach to the geometry which was a bit deeper than I had realised.  I just love the variety of approaches in our group. Subtle differences that enrich the outcomes.   Finally I bumped into Dan Stine, an old friend of many conferences.  Again I am stagerred by how much my life has been enriched by making the effort to be part of the global BIM community.  


Many old friends that I didn’t bump into, although they were there.  Pretty much inevitable in a meeting of 12,000 souls.  I guess there was a time when gatherings of this size would be either religious festivals or large scale warfare.  





Thursday, December 19, 2019

MY "POINT WORLD" PERIOD


Fresh from my first adventure in Pumpkinland, I was full of confidence about what I could do with the Conceptual Massing tools (aka Point World)  I looked back over Zach’s own work in this field, and stumbled on a helix family.  What could I do with this?  This was before the component version of the stair tool had been released. I decided to make a parametric spiral stair.  It uses the divided surface feature plus a bit of jiggery pokery to keep the nested tread family horizontal.  I never would have enough self-belief to see this through if I hadn’t received that seal of approval from Zach. It’s a family that has had a little practical use, but not a lot.  Mostly it was another step along my journey of exploration.


https://grevity.blogspot.com/2012/01/adaptive-spiral-stairs.html






But wait… maybe it would be useful for the Notre Dame spiral stairs.  There is some value in a free-standing family that is not wholly defined by two reference levels.  Something that you can copy up or down and adjust with a couple of parameters.  Worth a try.

So I wanted to apply the lofted profiles approach to Sanitary Fittings.  I have come at this problem from many different angles now.  Lofting isn’t going to be a panacea for every situation, but it can work well for some kinds of shape.  In this case I came up with quite an effective half-round basin with soft edges


https://grevity.blogspot.com/2012/01/basin-street-blues.html






Back to Ronchamp and I decided to start again from scratch, using a more controlled approach to the vertical walls at the back.  Then I created a topo-surface that allowed me to explore the relationship of the chapel on top of the hill to the village of Ronchamp itself, down in the valley.


https://grevity.blogspot.com/2012/01/our-lady-up-hill.html






Another exploration of stairs.  This time using the system tool.  I did a couple of exercises from my own imagination, then I thought of some historic cases that I might want to understand better.  Michelangelo’s entrance lobby to the Laurentian Library.  Sadly this was closed when I was in Florence a year or so ago.  It’s interesting for the shaping of the treads, and the way it splits into three separate runs below the half-landing level.  While I was at it I roughed out the walls of this fascinating space.

The final example is the theatre of Marcellus, an array of wedge-shaped stairs giving a quick insight to another historic building.  I ought to take a second look at some of those historical studies, maybe open them up for others to pick up where I left off.




https://grevity.blogspot.com/2012/01/upstairs-downstairs.html





So what about ramps?  What about curved and ramped bridges?  These are situations that have since been tackled with tools like Dynamo and Grasshopper.  My motivation here was simply to explore and extend the capabilities of the existing Revit toolset.  


https://grevity.blogspot.com/2012/01/up-garden-path.html



Around this time I bought a video camera and started recording myself playing music.  It's a bit scary how little use that piece of technology got before it was overtaken by the convenience of the smart phone.  But at the time it seemed like a big stride forward and I managed to post some three songs, complete with some editing trickery.  Sadly it was that extra editing chore which probably spelled the death of the experiment.  Keep it simple stupid.



https://grevity.blogspot.com/2012/01/other-side-of-me.html








https://grevity.blogspot.com/2012/01/golf-in-gulf.html







Coincidence again.  From time to time my boss comes up with an interesting challenge "should I choose to accept it".  Often it’s a personal friend looking for help.  He has an idea in his mind to give to his friend but it needs to be tested and converted into a sketch proposal.  That’s my mission, and I it usually proves worthwhile for me as well as for the project.  By chance this one was perfect for the spiral stair family I had recently developed.  We tried a couple of configurations and thanks to Revit I was able to set up a nice sheet with plan, section and 3d view


https://grevity.blogspot.com/2012/03/helix-take-2.html








Manufacturer content once more.  David Light had been commenting on the Boon Edam revolving door families, which are really quite detailed and complex.  Recent experience of trying to use these has been a bit frustrating for me.  I think I will develop a simpler generic version for future use.  In particular, I would prefer to use nested components that can be fitted either into a curtain panel family or into a wall hosted door family.  We always use the Rough Width and Height to schedule structural openings.  Curtain Panels are instance based.  Makes scheduling them with the other doors very difficult. 



Anyway the main topic of this post was rubber flooring.  Nora have an extensive library of Revit materials.  It’s a nice idea, when will you do a render that picks up that fine level of detail?  Anyway I talked through these an other issues just in case anybody was listening.  Too often manufacturers are persuaded into providing Revit content without participating in BIM processes themselves.  It’s the wrong way around in my view.  We collaborate with them.  Therefor it is useful to use BIM (of some kind) in that collaboration.  If they are using a digital tool, they will see the benefit and invest in incremental improvements.  Isn’t that a better strategy than just pestering them for freebies.


https://grevity.blogspot.com/2012/03/rubber-content.html







Around this time we had a design partner who was developing some very interesting proposals, often with complex geometries.  This was a competition entry developed using Sketchup.  I was asked if we could use Revit to generate the floor plates.  The challenge was how to convert surfaces into solids.  I tried to model this in the massing environment, but it was a struggle.  Learnt a lot though and generated some interesting images.


https://grevity.blogspot.com/2012/03/drawing-curtain.html






Someone alerted me to an additional feature (add edge) that I had overlooked.  (could have been Matt Jezyk)  I hoped this would be the answer to creating these spiky shapes that Sketchup handles so easily.  Didn’t quite do it.


https://grevity.blogspot.com/2012/03/curtain-call.html






Sanitary ware again and downloads from various sources.  The NZ ones are well made using native Revit tools.  The US content was much less consistent, even within a single manufacturer.  Towards the end of the post I showed off a simple generic family with size parameters.  The geometry is based on the kind of thing we used to draw as architects back in the old days.  Simple abstract shapes, pleasing to the eye, not meant to show every nook and cranny, just a placeholder that looks good on a nice clean sheet set and can be cross-referenced to the specification.  That’s my starting point now.  You can always substitute a proprietary item later on when the choices are firmly made.


https://grevity.blogspot.com/2012/03/family-favourites.html







An "all text" post about the imminent release of 2013.  Interesting to note that we had already begun skipping releases by then.  Mostly this is a function of the length of time an architectural project continues for.  I start to address the moaning that had begun even then about how Revit development seemed to have slowed down.  Don't see the point myself.  It's the best tool I have available to me for this kind of job so get on with it.  Ultimately the main constraint is my ability to come up with interesting ways to use it.



https://grevity.blogspot.com/2012/03/10987654.html





Yet another attempt at the spiky form.  Once again, it’s a learning experience and some interesting images to share, but ultimately not really achieving the goal.  I'm guessing that the best way to tackle this kind of challenge would be to learn Dynamo, but it's way beyond my abilities.  Not sure I will ever become a "coding guy" to be honest.


https://grevity.blogspot.com/2012/03/more-bermuda-triangles.html







Quick Access Toolbar revisited.  All the different kinds of line!!!  I do wish that Revit could handle these with just two icons, one for 3d space and one for 2d, then the software implements the appropriate command based on whether you are in the project environment of family editor.  It doesn't seem efficient to have four icons on the QAT all the time, two of which will be greyed out, depending on the environment.


https://grevity.blogspot.com/2012/04/qat-qed.html