Is this the end of the STL file?

An explanation, a manifesto, call it what you will... Death to STL. A narrative on the history and the future of STL in our words.

headshot of Duann Scott
Duann Scott
October 8, 2019 • 5 min read

The STL file format has served as the defacto file format for 3D printing for over 30 years, but that time is quickly coming to an end. At nTopology we believe that the depth and complexity of information required to drive existing advanced manufacturing and emerging technologies is accelerating the demise of the STL file format as the 3D printing standard. STL simply cannot capture, nor communicate enough information to realize design intent.


How are we helping to eliminate STL?

To enable this shift we are working closely with manufacturing hardware companies and their internal software divisions to find better ways to communicate our shared customers’ geometry. See this post and this post. We are making direct to manufacturing possible with no intermediary file formats and no 3rd party software. Most importantly, we are working with our customers to understand their requirements so that we can deliver those requirements as quickly and efficiently as possible.


But first… history

Let’s explore how we ended up here. The STL which has served the 3D printing industry valiantly was introduced by 3DSystems in 1987 as an export file transfer from CAD to their first SLA rapid prototyping machines. In this context, the STL was a triangle mesh that was a fairly suitable representation for prototyping geometries which were designed, originally, for traditional manufacturing. These prototypes required reduced complexity and tolerance for basic form and fit to achieve their value. Over the years, the industry has become more sophisticated and additive manufacturing is not simply for prototyping anymore. We are making end use parts. Consequently, the STL format lacks the ability to communicate not only the nuance and complexity of the geometry but also has no way to communicate the manufacturing information, tolerances, materials or really… anything.


It’s nothing but pain and misery…

Anyone who has gone through the process of taking a design from concept to manufacture has, no doubt, spent countless hours in laborious and painful mesh repair, closing holes, flipping normals and probably weeping into their keyboard in frustration (possibly causing an electrical short due to the sheer amount of shed tears). This is not just because of the nature of the mesh format but also because the CAD systems that author the geometry are likely built on a BRep kernel (also developed in the 1980s along with big hair and glam bands) where errors in the surface patch model are hidden in their own tools, and only become exposed and amplified when discretized and exported as a mesh format. Just like big hair and glams bands have become a relic of the past so too should STL files. Death to STL!


Hope springs eternal

There’ve been attempts to extend and revitalize the STL with STL 2.0/AMF that did not really see wide industry adoption. There is, however, the 3MF format which offers some hope even though it still retains an underlying mesh component. The extensions have been both ratified and extensible which means that manufacturing data can be potentially communicated. nTopology is a founding member of the 3MF consortium and will continue to explore workflows where our customers can use the 3MF to realize their design intent, however, we will undoubtedly continue to explore and create deep and direct communication with our software and that of hardware partners. 

So why now and how does nTopology factor into this?

As mentioned, most existing CAE and CAD systems have an underlying BRep or mesh based geometry kernel which by nature are limited in the complexity of information they can both generate and communicate. That’s not to say these tools aren’t important and aren’t powerful. They are, indeed, what got us to where we are today and have enabled us as a society to design and manufacture almost EVERYTHING around us today. But as the industry continues to become more sophisticated, design freedom for complexity must be realized. Enter nTop Platform which can capture, control and communicate complex geometry. 

The combination of design freedom possible with our unique geometry kernel, along with the ability to incorporate any data, be it simulation, empirical testing, generative algorithms or brute force numbers from a painfully maintained excel spreadsheet means a standard or generic file type will not suffice. nTopology is partnering with hardware and software companies, researchers and national labs, and most importantly our customers to ensure they can incorporate nTopology into their engineering workflow to enable them to achieve frictionless communication of data to manufacture. Communication that has speed and computational efficiency which reduces time and costly errors associated with the discretization and loss of data with intermediary file formats such as the STL. We are in effect helping make the STL obsolete and that’s why I say ‘death to STL!’ 

Death to the STL! 

If you are interested in directly connecting from nTop platform to your manufacturing workflow please request a demo below or contact partnerships@ntopology.com


Request demo

headshot of Duann Scott
Written by
With a background in Industrial Design, Duann is passionate about changing the way we approach design to unlock the potential of emerging manufacturing techniques. Quitting his Ph.D. in Australia to move to the United States to live out his thesis, Duann was soon recognized by the New York Times as a leading evangelist for the new business models these tools open up. Duann is VP of Partnerships at nTopology.