Apply AI and GPU Simulation for More Efficient CAM Programming

Advances in computer-aided manufacturing (CAM) programming capabilities are happening faster these days. In fact, Mastercam has pivoted to a new semiannual release cycle.

The company says this shift from annual to semiannual updates reflects its commitment to delivering continuous innovation and responding to the evolving needs of modern manufacturing. Rather than waiting a full year between major updates, customers will now receive significant new capabilities twice per year.

For example, two notable advances in its most recent Mastercam releases include artificial intelligence (AI) functionality to enable users to adjust feeds and speeds using voice or text commands, and graphics processing unit (GPU) simulation to achieve fast, high-resolution verification.

AI-Assisted Programming Advantages

Mastercam 2026.R2 marked the beginning of how programmers can interact with their CAM software. With new AI-powered capabilities, expanded voice control and deeper integration with learning resources, Mastercam Copilot has evolved into a powerful assistant to help programmers work faster, smarter and with greater confidence. Since the initial release, these features have continued to advance. This release delivers enhancements across four major areas.

Feed and speed editing. Making quick global or operation-specific adjustments no longer requires digging through parameters. With Mastercam 2026.R2, you can adjust critical cutting values using simple Mastercam Copilot commands.

For example, update spindle speed or feed rate for one operation, multiple operations or all operations; apply changes in Mill, Lathe and Router (Wire support is coming in a future update); view a confirmation dialog before the change is applied; and revert changes instantly if needed.

Machine group and operation creation. Setup time can be significantly reduced with Mastercam Copilot’s expanded command support. You can now generate machine groups and populate them with operations using a single natural language instruction.

Here’s an example command:

“Create a mill machine group using the default machine with a contour operation and a 2D high-speed operation.”

With this, Copilot can create a machine group with a custom name, a machine sourced from the /CNC_MACHINES directory or your default machine, or one or several operations — more than 200 toolpath types will be supported as ambiguous naming is resolved. This can dramatically reduce repetitive setup work and overall programming time.

Active listening mode. Copilot enables hands-free interaction with Mastercam. First, enable “allow active listening” in the new Copilot settings panel and then optionally choose to keep active listening on by default. Next, use the wake-phrase, “Copilot…”, to trigger a command. (For example: “Copilot, create a rectangle.”)

Copilot listens, pauses briefly and executes. Then, you can direct Copilot to stop listening using the UI toggle or by simply saying, “Copilot, stop listening.” This feature is especially helpful during multitasking, training or when you want to maintain workflow momentum without switching tools.

Video scanning. All myMastercam hosted videos including tutorials, walkthroughs and training content are now part of Copilot’s reference library. That means you can ask: “Hey Copilot, what feed rate was used in the Mill 2D tutorial video?” Copilot will then scan the video, extract the information you asked for, provide a timestamp and include a direct link to jump to that exact moment.

This transforms video content from “something you have to watch” into actionable, searchable knowledge eliminating the need to scrub through long tutorials to find a single detail.

Fast and Accurate GPU Simulation

As a programmer, have you experienced this? You’ve programmed a complex five-axis part with millions of toolpath points, and now you are staring at a simulation progress bar that barely moves. Your workstation is tied up, deadlines are approaching and you are left wondering whether the lower resolution simulation you had to run will catch every gouge.

This bottleneck is common when using traditional central processing unit (CPU) simulation which often forces programmers to compromise between lower accuracy for speed or higher accuracy with longer wait times.

That said, Mastercam 2026.R2 offers an alternative with GPU-accelerated simulation and verification, offering much faster processing for supported milling workflows. With the proper GPU, programmers can realize 10-times performance increases compared to CPU-based simulation.

Traditional CPU simulations create a variety of constraints. For example, there’s an accuracy tradeoff. Simulating millions of toolpath points demands significant CPU resources. To get results sooner, programmers often reduce simulation resolution. But doing so increases the risk of missing subtle collisions or inefficiencies. Also, large three- and five-axis milling tool paths can tie up a workstation for hours, delaying programming and stalling iterations. Plus, when verification is slow, programmers might avoid experimenting with optimal strategies. Potential cycle time reductions go unexplored because simulation simply takes too long.

By contrast, Mastercam’s GPU-accelerated simulation changes what is possible for supported milling workflows. (GPU Simulation is native to Mastercam 2026.R2 and works within both Machine Simulation — external post — and Mastercam Simulator.)

By utilizing the parallel processing power of modern graphics cards, Mastercam can process material removal and collision detection much more efficiently. This is accomplished in part thanks to parallel computation optimized through Vulkan 1.3, discrete texel modeling for precise material removal, and real-time performance gains when tool-versus-stock collision detection is enabled. With a compatible GPU, CPU simulations that previously took 3-4 hours at practical resolution can often now be completed in 15-20 minutes at maximum resolution via GPU.

But beyond time savings, there are other advantages. Programmers can try multiple strategies, test different tool motions, and experiment with optimizations without waiting all day for results; verify with greater confidence because full-resolution simulations help catch tool/stock issues that might slip through lower-detail runs; and improve quoting responsiveness because faster verification makes it easier to present accurate machining estimates during customer discussions.

For more information about these and other new capabilities offered in the latest Mastercam release, visit mastercam.com.