So my phone starts blowing up around 10am with texts from my brother-in-law, all caps, something like "THEY DID IT WE HAVE UNLIMITED ENERGY NOW." I had the DOE livestream on in the background while I was supposed to be fixing a broken cron job, and yeah, Jennifer Granholm and the Lawrence Livermore folks were up there announcing that the National Ignition Facility had actually pulled off fusion ignition. Net energy gain. First time in human history a controlled fusion reaction produced more energy than what went into it. That part is real and it's genuinely a big deal, I don't want to undersell it.
But the number everyone's running with is not the number that matters, and almost nobody in the coverage I've read today bothered to explain the difference.
Here's what actually happened on December 5th (they held the reaction, then sat on the announcement for over a week, presumably to build the hype machine, which, fine, I get it, government funding depends on headlines). NIF fired 192 laser beams at a tiny pellet of hydrogen fuel, delivering 2.05 megajoules of energy to the target. The fusion reaction that resulted kicked out about 3.15 megajoules. That's a gain of roughly 1.5x - more energy out of the fuel capsule than the lasers put into it. Scientific breakeven, target gain over 1. Real milestone, took them since the facility opened in 2009 to get there, and I'm not being sarcastic when I say it's a legitimately cool achievement for the physicists who've been grinding on this for over a decade.
Except that 2.05 megajoules of laser energy on target didn't come from nowhere. Charging those lasers pulled something like 300 megajoules from the grid, because NIF's laser system is unbelievably inefficient at converting electricity into laser light hitting a target the width of a peppercorn. So the actual wall-plug-to-output ratio is still deeply, deeply negative. Like, you'd need the output to be roughly 100x higher just to break even on the whole facility's power draw, not counting the fact that this thing can only do one shot every few hours because everything has to cool down and reset. A working power plant needs to do this multiple times a second.
None of that made it into the "HOLY GRAIL OF CLEAN ENERGY ACHIEVED" headlines I saw today, and it's kind of driving me nuts. I spent twenty minutes typing out a reply to my brother-in-law trying to explain the target-gain-vs-wall-plug-gain distinction and I don't think it landed, because by the time I finished he'd already sent me a different article that used the phrase "limitless energy" unironically.
I want to be clear I'm not trying to be the guy who poops on good science news. This is a real result and Kim Budil, the LLNL director, said something on the call that stuck with me, along the lines of most significant scientific challenges taking a series of breakthroughs over a decade or more to become reality, and fusion ignition being one of those "several" breakthroughs, not the last one. That's an honest way to frame it and I wish more of the coverage had used her framing instead of running straight to "we solved climate change today."
Realistically we are talking decades before anything resembling a fusion power plant sends electricity to your house, if it happens at all. NIF itself isn't even designed to be a prototype power plant, it's a research facility built mostly to study weapons physics without needing live nuclear tests, the civilian energy angle is almost a side effect of the design. There are other approaches too, the tokamak folks (ITER, Commonwealth Fusion) are doing magnetic confinement instead of laser inertial confinement, and there's a real argument for which approach actually scales to commercial power first. I don't think anyone credible thinks it's this one.
Anyway. I closed about four tabs of breathless coverage today, fixed exactly zero lines of that cron job, and now I'm annoyed at tech journalism in a very specific and boring way. Go read the actual DOE press release if you want the real numbers instead of the pull quotes. It's less exciting than "we cracked the sun open in a lab," but it's also just correct, which counts for something.