I've written simulators as a career for more than a decade, and I'm stunned at what a good job you've done. The simulation engine is excellent and the visualization is prettier (and more intuitive!) than any I've ever seen.
Growing up in Cincinnati, the Omnimax at the museum center was a huge influence. The light tunnel intro (one of many adapted from the Graphic Films Corporation logo [1]) absolutely blew my mind and gave me a lifelong obsession with computer graphics.
You probably already know this, but the light tunnel intro wasn't computer graphic imagery. It was created using optical effects generally referred to as slit scan. The same technique was used to create the star gate sequence at the end of 2001: A Space Odyssey.
Apple | Senior simulation software engineer | C++ | Hybrid (3 days per week in office) | San Diego CA or Austin TX | Full-time
We're seeking an experienced software engineer to build simulations for the Satellite Communications Group. The job requires excellent geometric intuition and fluency in C++, but prior experience in satcomm is not necessary. We've had successful applicants with backgrounds in physics, graphics, FEM, control systems, and hardware. We get to work on really cool stuff and build systems that provide essential off-grid connectivity to users around the world!
I apologize profusely for this: made this account to contact you. The Apple website has locked me out of my old account and isn't letting me make a new one. Can I please pass on my resume to you directly? Thank you.
My junior high friends that I've been having parties with for 30 years live in Minneapolis (where I grew up). They fly out for New Year's Eve each year.
But, in fact, some friends who regularly attended LAN parties in the Bay Area moved to Austin around the same time we did. And some others are also willing to travel for New Year's.
Plenty of people in tech moved from Silicon Valley to Austin to get a better tax / quality of life deal, even in my social circle.
Remote working becoming widely available really made a difference.
I'm in a completely different part of the world, but for similar reasons I ended up with a few friends in tech who moved to the same part of the world - and I've also met similar profiles to ours, attracted by the same reasons.
Where did you move to if you don't mind me asking? The chasm between SV tech comp and various "completely different part of the world" is massive. Were you able to meet your employer in the middle?
This appears to be by Kenta Cho, a.k.a. ABA games, who has a long history of writing wonderful small games in D. His work is well worth checking out. https://www.asahi-net.or.jp/~cs8k-cyu/
I feel like HN is on a kick right now with this concept of “small games”.
But folks are tripping over the definition of small. There’s the small as defined in the link you have, small as in low mechanic and low content like “Gone Home”, small as in asset flips, small as in just not AAA, small as in really intricate but not large, small as in procedurally huge but limited visuals and simple mechanics, small as in just a dating sim or porn, and then small as in old school “small games” like Doom Guy’s Commander Keen which is cute and good at the time but laughably unpolished for todays standards.
Since the beginning of the industrial era, people have been looking for side artisan-gigs to distract from how droll their day job is. This isn’t particularly different than woodworking, gardening, knitting and the legion of other hobbies used for this purpose.
I suppose it’s more of a Victorian thing but it’s also primarily the domain of the white collar class.
Someone who does carpentry as their day job is probably not going to do woodshop in the garage at home even today, but they just pick other things, like grilling.
It'll depend on who you ask and what lens you look in. "small" for a dev is very different from a consumer
And when you split into disciplines it gets even more murky. VN's are "small" for a programmer (you may not even need to program for many of them) but can be hundreds of art assets for an artist to work on. Likewise, something like Dwarf Fortress didn't even have a GUI at first but is filled to the brim with simulation state for a programmer to manage.
And if course that definition is relative. Final Fantasy VII was top of the line in 1997 and took 5+years, but it'd be at best a beefy 2-3 man project over a few years today.
I haven't visited this site in years though. I wonder if mobile games are the reason why. I stopped playing a lot of web/flash/html5 games a long time ago when gaming (for me) moved to the app store.
But I recognized the name and remember playing a lot of these games back in the day.
I always thought they were delightful. Building off a single game mechanic or two and then moving on to the next concept. The dev has a good system in place for shipping :)
Does anyone know a way to subscribe to find when he adds a new game? I checked out his blog and Twitter and they don’t seem to be a great way to follow just the games. Can’t find an RSS feed either.
Power conversion losses account for about 60% of the energy that goes into electricity generation in the US [1]. Superconducting generators and transformers could take that figure down to a fraction of that.
These algorithms work best on smooth functions with mid-to-low 100's of variables. They are very sample efficient, often obtaining superlinear convergence using just a linear number of evaluations of the objective function.
I'm very glad to see these algorithms get the maintenance they deserve. I pulled them apart and used lots of their pieces and theory for my PhD; MJD Powell's original code was subtle and difficult to understand. Hats off to those doing to the hard work of keeping them alive. These algorithms are the absolute best in their domain, and should be more widely known.
You raise a very good point in that the "formulate the problem the way the solver wants" step is legitimately difficult and full of pitfalls. Simply figuring out the translation can be hard, and even then there are many ways to formulate a problem which are mathematically equivalent but have drastically different performance when fed to the solver.
It really feels like a tools or language problem. Heck, we used to have to manually work out derivatives for continuous optimization problems, but nowadays programming languages with performant built-in autodiff often make this trivial. Removing the manual derivation hassle let loose a flood of cool ideas and applications, even though there was no technical hurdle preventing them in the first place.
Alternate problem specifications is a well-explored area (what is Prolog if not a way of describing problems for a constraint satisfier?), but I wonder how many other neat things are dammed up behind usability problems.
reply