June 19, 2015 │Sylvia Knauer, Autodesk
As you start working in Revit 2016 you may experience noticeably greater speed, especially when working with large models. It’s not your imagination; scalability improvements in Revit 2016 are at work behind the scenes speeding your software use.
First it’s important to understand that performance is different from scalability. Enhancements that help software process objects more quickly can alleviate some of the performance issues that may occur as a model gets complex. But performance enhancements alone are not enough. You also require scalability, which is a matter of being smarter about how software processes data to relieve the software of work that slows it down. Performance of the software scales, meaning that the software performs operations in the same way whether a model is tiny or huge. While other software providers may focus on using multiple threads to process more data, Revit is enhancing both performance and scalability not only by using multiple processors, but more importantly by rethinking the entire system’s use of data.
Performance improvements, like multi-threading or removal of localized bottlenecks, only help so much. They are incremental changes that do not change the nature of the game. Scalability improvements, on the other hand, rethink the basic assumptions of the system to make it independent of model complexity. No matter the size of the model, the time to perform the same action in a scalable model is the same. Revit scalability is a matter of being smart about the data.
We took two approaches to improving scalability in Revit 2016. The first was to make some operations faster by:
- Sharing data: Ensuring that repetitive data is shared rather than copied is one way to make operations run faster. In Revit 2016, MEP contractors and detailers can create fabrication-ready models inside Revit using content from the Autodesk Fabrication products (Autodesk CADmep, Autodesk ESTmep, and Autodesk CAMduct) for a more coordinated model. MEP Parts saw significant speed improvements from shared data. Due to less processing of data, regeneration times improved significantly. Reinforcement is faster due to use of shared data and skipping processing of unused data. Both regeneration times and draw times improved.
- Being smarter about data: Improvements that calculate less data using quick rejection, simplified data, or on-demand access. A few examples in Revit 2016 include:
- Not drawing things that are outside the view.
- Remote snaps include only the objects in the visible portion of the view, rather than the view extents, improving snapping behavior while working with large system models.
- Revit models with links display faster, because geometric display calculation works only on links that appear within the crop region.
The second way we improved scalability was to make some operations run in the background while you continue to work by:
- Enabling continuation of work while the model is updating: In Revit 2016, you can navigate models more smoothly and continuously because you don’t have to wait for the software to finish drawing elements at each step when you pan, zoom, and orbit around a view. The new Navigation during Redraw option speeds navigation by interrupting the display of model elements as you move through the model. Users see more of the model as they’re navigating and can navigate much faster without need to wait for the view to update fully.
- Making calculations only when requested: New settings in the Calculations drop down in Revit 2016 improve model performance in large projects. The “Performance” setting suppresses the processing of system-level calculations with most duct and pipe systems. The “Volume Only” setting improves Revit performance while working with large Fire Protection, Vent, and Other classification system types by turning off calculation of the Volume parameter.
Parallel Processing & GPU:
Revit becomes faster using parallel processing – Revit performs better through use of multiple CPU cores or GPU to compute data concurrently. Examples include multi-threading and graphics hardware acceleration. In Revit 2016, the draw speed of loadable families is improved through better use of the GPU (graphics card). As a result, navigating a model with multiple identical instances of families is a lot faster.
In fact, the following tools in Revit take advantage of multiple processors and multiple core processors for calculations, increasing the performance of the tool in Revit:
- New to Revit 2016, progressive display (Navigation During Redraw) is used by all models all the time
- 2D Vector export such as DWG and DWF
- Mental Ray Rendering
- Wall Join representation and update in plans and sections
- Element Loading. Loading elements into memory is multi-threaded, reducing view open times when elements are displayed for the first time in the session.
- Parallel computation of silhouette edges such as outlines of curved surfaces, used to accelerate navigation of perspective views.
- Translation of high level graphical representation of model elements and annotations into display lists optimized for graphics cards, engaged when opening views or changing view properties.
- File open and save
- Point Cloud Data Overlay
In addition, Revit 2016 features a number of user productivity features and enhancements, which coupled with improvements to scalability drive overall improved project performance. Users will definitely sit up and take notice.
Thanks for reading!
Please enjoy a limited number of articles over the next 30 days.
For total access log in to your The BIM Hub account. Or register now, it's free.