Last week, Apple announced a new line of Mac computers that will be the first to run on its M1 processor chips. The company promises the new machines will be faster and more powerful than their predecessors, but for Mac-using chemists the shift conjures visions of software crashes as older software struggles to run on the new chips. Computational chemists and computer scientists have been building fixes they hope will ensure software runs smoothly, but one expert says he’s still going to wait before upgrading his machine.
Chips like the M1 execute the instructions written into computer programs, whether for word processing, web processing, or molecular dynamics simulation. Apple has used Intel chips in Macs since 2006. Computational chemist François-Xavier Coudert compares the transition to changing languages: the new chips simply won’t understand the old instructions.
Coudert, a researcher at the French National Centre for Scientific Research and a professor at PSL University, is part of a small group that has been volunteering to update quantum chemistry-related programs built on older programming languages like Fortran so they’ll work on new Macs. Many are free, open-source software that don’t have paid programmers assigned to keep them running smoothly, although Apple is providing some support to open-source developers. Coudert thinks most of these programs will eventually run smoothly on the new chips, but he advises users to wait 3–6 months before getting a new Mac unless they’re willing to deal with bugs that will inevitably crop up.
The situation is a bit different for PerkinElmer, the company that makes ChemDraw. David Gosalvez, executive director of science and technology at PerkinElmer Informatics, says the company’s software will rely on a compatibility mode Apple has added to new computers that will let the machines run old software. Coudert says that mode may not work smoothly or quickly enough for some computational chemistry programs. Gosalvez expects ChemDraw to “just work” until the company can fully update the software for the new chips over the next couple years.
David de Sancho, a computational biophysicist at the University of the Basque Country and the Donostia International Physics Center, is one user who’s content to be patient. Like others, he has a dedicated non-Mac computer and access to a supercomputer for more demanding computations. But he uses his Mac laptop for visualizations, data analysis, molecular dynamics simulations, and writing code. He says he doesn’t need the new chip’s performance on his laptop right now, and he doesn’t want to risk bugs and crashes that could interrupt his work. “I’ll just wait,” de Sancho says.