Unsupported browser

For a better experience please update your browser to its latest version.

Your browser appears to have cookies disabled. For the best experience of this website, please enable cookies in your browser

We'll assume we have your consent to use cookies, for example so you won't need to log in each time you visit our site.
Learn more

A day in the life of ADT3

  • Comment
The ADT3 modelling solution has a wealth of technical ability but does not have the intuitiveness to match

Eighteen months ago I reviewed Architectural Desktop 2 (ADT2) and I did not like it.

I pointed out the inconsistencies in Autodesk's approach to CAD modelling, the inability of its family of modelling applications to share geometry and, in particular, its method of addressing interoperability - the way the geometry created in one application is transferred to another for reuse. All proprietary-modelling applications suffer from data loss or more specifically loss of information from data during translation. However, data produced in ADT2 could not even be viewed in native AutoCAD applications of the same version without a specific object enabler (an extra mini-application, which is able to read the data from ADT2 and display it in a format which AutoCAD will understand). Furthermore, the much-loved and well-supported LT could not read the data at all as it did not have support for object enablers.

Thankfully, with the 2000i suite of products much has changed. LT now supports object enablers and, with release version three of Architectural Desktop (designed specifically to work with 2000i), many new functions have been incorporated. In an attempt to make ADT3 a more comprehensive building modelling solution, new functions have been added and include curtain walls, window assemblies, structural members, slabs and floor openings, ducts, area calculations, display management tools and live sections.

As it can be very difficult to get a grip on a new product and explore its functionality in depth, I decided to take a different approach this time. Instead of getting my paws on a copy of the software, sitting down with the manual and treading gingerly through the online tutorial, I decided to delegate, leaving both eyes free to watch and both hands free to record my observations on my trusty Palm Vx.

I arrived in Guildford at the UK headquarters of Autodesk to give ADT3 a good going over, or at least a good looking at. My strategy was simple; instead of testing the software I wanted to observe it being applied to solving a problem by people who already knew the software inside out. Like a mixture of the Crystal Maze, Challenge Anneka and Busman's Holiday (yes my years in architecture school were well spent), I arrived armed with plans and sections of a building which I wanted to see constructed in ADT3. In return Autodesk assembled a team and placed them at my disposal for one day to demonstrate ADT3's all-round modelling ability and section-cutting prowess.

I had the impression that I was being viewed with caution but when I produced the drawings of Le Corbusier's Villa Savoye the mood in the room lightened. I was not, they decided, out to trip them up. I like using the Villa for modelling as it balances solid with void and contrasts simple geometric forms with the asymmetric curves of the screen walls on the roof. It is also easy to subdivide the whole into separate component parts across more than one file and, unlike Rome, it can easily be built in a day. In short, it is the perfect subject for this simple test.

As with all new projects, we started with a blank canvas. ADT3 shares the 2000i interface and can start up with the 'classic' AutoCAD interface or with the newer-look web-friendly 'AutoCAD Today' page that has been designed to look like a web page. I do like the latter but it does not sit comfortably on my meagre 800 x 600-resolution laptop, so I am forced down the route of the traditional classic interface in order to see all of the available options. First we created a new grid file (in the interests of good practice and collaborative working we decided to split the model into logical partitions) and used the grid tools, which had a simple layout and were easy to use, to place the grid. In a flash it was complete and we were ready to move on.We saved the file and created a second file for the ground-floor columns.

Using the omnipotent X-ref tools we attached the grid file and populated the new file with structural columns. The library for selecting the columns was comprehensive and XML-based (written with text) making it easy to transport across slow Internet connections for data sharing. Whenever columns are placed they can also be 'associated' to the grid so that if the grid spacing should change the columns change too.

Unfortunately for us both the grid and the columns have to reside in the same file and as we were using many different files referenced together the associations did not work. This is a shame, as anyone who has used and understands X-refs knows of their value and uses them wherever possible.

We saved the columns and created another model file for the first-floor walls, referencing the grid and columns files. The walls went in easily but had to be moved into the correct vertical location, which seemed odd for a tool designed for 'real-world' models.

The floor and roof slabs were placed and again had to be moved vertically into place.

Surely with such an application it is possible to set a 'virtual' base for each piece of geometry but this was not demonstrated. I was beginning to think that, even for people who know the application well, it is not logical or intuitive to use.

Creating windows of varying sizes was easy. Using 'Window Styles' for different display representations (Display Reps) it was easy to create the first and then duplicate and modify it to create all other iterations. This had to be the best bit yet.We checked the file size and it weighed in at a hefty 428K. This seemed excessive so we checked the size of the template which registered a considerable 385K. This meant that the entire model we had created thus far contributed just over 40K to the size of the file.Now that is efficient!

In a world of collaborative working practices where data is shared across both intellectual and geographical borders, ADT3gets a huge thumbs up for its approach to layer management. At any time the layer names being used can be 'exchanged' for others without affecting the integrity of the data. In this way your own internal layer standard can be changed to look like any other standard at any time with just a few button clicks.Well done Autodesk.

Placing ramps and stairs proved to be quite an ordeal. At first we tried to place ramps as stairs with a zero riser height, but it did not work. Then we tried to place them as a slab and rotate it into place. Again, we had problems with the pivot points of the slab for rotating. At one stage we were about to revert to using ACIS solids for the geometry but chose not to, as the solid would not Boolean (cut back to or extend forward to meet another piece of geometry). All of this was trying, and failing involved enough keyboard strokes and right-click pop-up menus on the mouse to bamboozle most concert pianists.

Eventually the problem was solved and we moved on to the curved walls at the top of the building. Again, this operation was trouble free and impressed me but left me wondering how they could get some commands so right while seemingly forgetting about others entirely.

With time running out we cut short the modelling process and proceeded to extracting plans and sections from the model. Over the past three to six months I have looked at a number of applications of this genre - model-centric applications from which all construction documentation is extracted - and I can honestly say that when it comes to cutting sections from the model, ADT3 comes a poor last. I would not expect it to match the sheer ability of Revit, which makes section cutting and manipulation look easy. AllPlan from Nemetschek and TriForma from Bentley are also easier to use and deliver more capable section-cutting tools than ADT3.

I really wanted to like this version of ADT following my disappointment with the previous one. While some of my earlier concerns have now been addressed, it appears that the competition has managed to address the fundamental issues of the single building model better. That said, ADT3 has much strength and I know of people who use version two and achieve great results. Adopting the third way (3D that is) is hard enough to get your head around without having to learn complex new tools. To make the leap from 2D to 3D successfully you need some simple intuitive tools. ADT3 has an abundance of technical ability but unfortunately it is not intuitive.

Joe Croser is a director with Adrem-DCX, specialist in CAD consultancy and training.

To contact him e-mail Joec@adrem-dcx. com or tel 07973 263360

  • Comment

Have your say

You must sign in to make a comment

Please remember that the submission of any material is governed by our Terms and Conditions and by submitting material you confirm your agreement to these Terms and Conditions.

Links may be included in your comments but HTML is not permitted.