Everyone knows that make proper mesh for complex geometry on the first try is impossible.
Robot makes perfect mesh, but, unfortunatelly uses only one core of CPU, so process is quite long on powerful computers.
Please, make multuthreading support for meshing and we can save a lot of precious hours :-)!
In some codes wind load is divided in two parts - dynamic and static.
For example, in Russian Codes, the dynamic part of wind load depends on: static load, masses and natural frequency.
Unfortunatelly, Robot hase no capabilities to calculate dynamic wind load automatically.
Quite long slabs on grade has irregular foundation pressure, thus they have to have irregularal elastic coefficient to calculate forces in slabs correctly.
Unfortunately, right now Robot has no capabilities to set non-constant elastic coefficient for the slabs.
For some calculations sometimes we need to take into account different Young modulus depending of the nature of load cases. This is expected for a single model, to avoid duplication of model data. We see such requirements in Eurocode and also AISC codes.
Working on my last project (it's an concrete structure, about 150 000 nodes and 170 000 FE-s) I was extremely annoyed by Robot Str 2013 general performance. For example, changing an load value is operation with time consumption not less that half of minute on quite modern hardware (core i7 etc.). Changing values in tables is not much quicker...
I understand that you can recommend to use diaphagms instead shell etc, but in general - performance of Robot is not the best one, I can say the worst one. Now it is 2013 year and yes, sometimes we have to analyse and calculate models with more than 3 bars...
Here I do not speak of calculating time, meshing time - it's only problem of general performance, for example, changing load values in large scale models. So, my Idea is to optimize code (I'm fully understand that it can take a lot of time) to make Robot works much quicker.
I have been looking at Robot for several years now. I tried the demo in 2011 and 2012. I am a long-time user of Autodesk products (Civil 3D, Architecture, ADR) and I am attracted by Robot because it is an Autodesk product now.
Each time I look at Robot I get the impression that not much is happening in development and that Robot is neglected by Autodesk. There are no updated manuals or tutorials or support or training seminars and the GUI has no relation to other Autodesk products and does not have the quality feel of Architecture.
Autodesk has a huge opportunity to take the lead among the likes of RISA, Ram, S-Frame, and SAP2000 with a comprehensive user friendly structural design/analysis package that integrates seamlessly with Revit and Autocad. For my part I am looking for some leadership and innovation and committment from Autodesk to this venture.
It would be very useful to implement RC design of typical pile caps and circular spread footings (including footings with variable thickness).