The machines we send are only as capable as the brains we give them
In the long human effort to extend our reach beyond the cradle of Earth, the limiting factor has often been not courage or rocket fuel, but the fragility of the thinking machines we send ahead of us. NASA and Microchip Technology have announced a partnership to forge a new generation of space processors — chips capable of 100 to 500 times the computational power of those currently flying — while remaining hardened against the radiation that makes deep space so hostile to ordinary electronics. The work, driven by NASA's Jet Propulsion Laboratory, reflects a quiet but profound conviction: that the minds we give our machines will determine the scope of what humanity can discover.
- Every rover and lander beyond Earth's magnetic shield operates on processors that are decades behind in performance — radiation hardening has always come at the cost of raw computing power, and that trade-off is now a ceiling on what science can achieve.
- With communication delays stretching into minutes between Earth and Mars, a spacecraft that cannot think for itself is a spacecraft that hesitates — and in hostile terrain, hesitation is failure.
- NASA's Jet Propulsion Laboratory has spent decades writing clever workarounds for underpowered space chips; this partnership with Microchip Technology is an attempt to stop compensating and start building hardware that matches the ambition of the missions.
- The target is dramatic: processors 100 to 500 times more powerful than current space-grade chips, capable of real-time terrain analysis, autonomous navigation, and simultaneous management of multiple scientific instruments.
- No flight timeline or detailed specifications have been released, but the announcement itself signals that engineers believe the core technical barrier — hardening a modern, high-performance chip against cosmic radiation — is no longer insurmountable.
NASA and Microchip Technology have announced a partnership to build a new generation of processors designed for the unforgiving conditions of deep space — chips intended to power rovers, landers, and eventually crewed missions to the Moon and Mars. The computational ambition is striking: these processors will deliver between 100 and 500 times more power than the chips currently operating on active missions, a leap that is less an upgrade than a transformation.
The engineering challenge at the heart of this work is radiation hardening. Beyond Earth's protective magnetic field, cosmic rays and solar radiation relentlessly bombard spacecraft electronics, flipping transistor states unpredictably and corrupting data. Space-grade chips have historically been built to survive this assault — but the architectural choices that make them resilient have also kept them slow. NASA and Microchip are attempting to break that trade-off for the first time.
JPL, which manages NASA's robotic exploration programs, has long understood the ceiling that underpowered processors impose. For decades, its engineers have designed software and mission profiles to work around hardware that cannot keep pace with the complexity of modern science. A processor 100 to 500 times more capable would change the calculus entirely — enabling rovers to analyze terrain in real time, make autonomous driving decisions, and handle data from multiple instruments at once, all without waiting for instructions from Earth.
The implications extend to crewed missions as well. Future astronauts on Mars will face communication delays of many minutes with mission control, making real-time autonomous systems for habitat management, life support, and emergency response not a luxury but a necessity. Today's space processors cannot support those systems. These new chips, if successfully developed, could.
No specific flight timeline or technical specifications have been announced — the partnership is still in its early declaration of intent. But the confidence embedded in that declaration matters. The machines humanity sends to other worlds are only as capable as the minds we give them, and this effort suggests those minds are about to grow considerably.
NASA and Microchip Technology have announced a partnership to develop a new generation of processors built specifically for the demands of deep space exploration. The chips are designed to power rovers, landers, and eventually crewed missions to the Moon and Mars—environments where radiation exposure, extreme temperatures, and the inability to simply swap out a failed component make ordinary computer hardware useless.
The computational leap is substantial. These new processors will deliver between 100 and 500 times more processing power than the chips currently flying on active space missions. That's not merely an incremental upgrade. It's the difference between a rover that can process images slowly and one that can analyze terrain in real time, make autonomous decisions about where to drive, and handle the data streams from multiple scientific instruments simultaneously. For a machine operating millions of miles from Earth, where communication delays stretch into minutes, that autonomy matters enormously.
What makes this partnership significant is the engineering constraint that has always limited space computing: radiation hardening. Beyond Earth's protective magnetic field, cosmic rays and solar radiation bombard spacecraft and their electronics. Standard processors degrade rapidly under this assault. Their transistors flip states unpredictably, corrupting data or causing crashes. Space-grade chips are built with redundancy, shielding, and architectural choices that make them far more resilient—but those same choices have historically meant accepting lower performance. NASA and Microchip are attempting to break that trade-off, delivering both the radiation tolerance that deep space demands and the raw processing muscle that future missions require.
The Jet Propulsion Laboratory, which manages NASA's robotic exploration programs, has been the driving force behind this development. JPL engineers understand intimately what current processors can and cannot do. They've worked around the limitations of existing space-grade chips for decades, designing clever software and mission profiles to compensate for hardware that simply cannot keep pace with the complexity of modern science. A processor 100 to 500 times more powerful would fundamentally change what's possible.
The implications ripple outward. More processing power means rovers can operate more independently, reducing the need for constant commands from Earth. It means instruments can collect richer data and process it onboard rather than transmitting raw streams back home. It means future crewed missions—which will require real-time decision-making in environments where radio contact with mission control is delayed by minutes—become more feasible. The chips could enable autonomous systems for habitat management, life support monitoring, and emergency response that simply cannot exist with today's space processors.
Neither NASA nor Microchip has announced a specific timeline for when these processors will fly on actual missions, nor have they detailed the exact specifications or architecture. The announcement itself is largely about the partnership and the capability target. But the fact that this work is underway signals confidence that the technical challenges are solvable—that you can harden a modern, powerful processor against radiation without sacrificing the performance gains that make it worth building in the first place.
For the next decade of space exploration, this partnership may prove as consequential as the rockets themselves. The machines we send to other worlds are only as capable as the brains we give them.
A Conversa do Hearth Outra perspectiva sobre a história
Why does a processor need to be different for space? Can't NASA just use faster commercial chips and shield them better?
Radiation doesn't just damage the chip's exterior—it flips individual bits inside the transistors themselves. A commercial processor might crash or silently corrupt data within hours. Space-grade chips are built with redundancy and error correction, but that architecture has always meant accepting slower speeds. This partnership is trying to do both at once.
And that's never been possible before?
Not at this scale. You can harden a chip, or you can make it fast, but the engineering trade-offs have been brutal. Every redundancy you add slows things down. Every optimization for speed makes you more vulnerable to radiation. This is about breaking that wall.
What changes if a rover suddenly has 100 times more processing power?
It stops being a remote-controlled toy. Right now, rovers are semi-autonomous because we can't send commands fast enough—there's a 20-minute round-trip delay to Mars. A faster processor means the rover can see a rock, analyze whether it's worth studying, and decide to drive toward it, all on its own. That's the difference between exploring a few hundred meters and exploring miles.
Is this mainly about rovers, or does it matter for crewed missions too?
Crewed missions are where this really becomes critical. Humans need real-time systems for life support, navigation, emergency response. You can't have a 20-minute delay when something goes wrong. A processor this powerful could handle the autonomy and redundancy that keeps people alive.
When will we actually see these chips on a mission?
That hasn't been announced. This is the partnership announcement—the commitment to build them. Getting them tested, qualified, and integrated into a spacecraft takes years. But the fact that NASA and Microchip are doing this now suggests they believe it's solvable, and that the next generation of missions will need what these chips can do.