I've gone down that rabbit hole and written countless versions of space invaders. What's really cool to me is that - to get it right - the secret is moving only one invader in the pack every 1/60th of a second, this gives the pack movement the same feeling as the original. The genius of the original coder in creating this illusion never escapes me.
This distinctive movement is a compromise with the limitations of a 2mhz 8080 with no video support hardware to speak of.
Looking at a commented disassembly (https://www.computerarcheology.com/Arcade/SpaceInvaders/), there's so little time available that each frame redraws nothing more than the player, their shot, one of the 55 aliens, and one of the three alien shots. Four things. When every moving object has to be manually erased and redrawn within the space of one frame, you kinda have to stagger the movement.
After 25 years of writing random space invaders clones - I finally achieved perfection after reading the Computer Archeology site, then I made my last clone and IMHO I'm finally happy to put this obsession down.
This is brilliant, your algorithm produces really great results - and your write-up is super! Would be great to have it as a simple function that we could use in our games ;-)
I hit the front page with ClassicVideoPoker.com (https://news.ycombinator.com/item?id=37763098) and got around 52k hits that day. That was over 1.5 years ago, and to this day, I still get a steady stream of users. A small percentage of them return daily to play, with some racking up hundreds of hours.
Unfortunately, I wasn't running any decent stats at the time, but I see now that I'm still getting referrals from "hackernewsletter," although I'm not sure of the source.
It was great to see the traction, but at the same time, it terrified me. I had all kinds of plans: logins, contests, a community, and new retro-styled games. Unfortunately, the "HN effect" paralyzed me. I was too afraid to push updates, fearing I might break something and lose the users I had. That fear persists even today, with a much smaller but extremely loyal user base.
BBC basic on this platform was amazing, inline assembly, SWI & SYS calls made calling firmware & inline assembly programming a breeze. Mapping memory and registers to basic variables was trivial. Such a nice programming experience.
I recall a few years ago writing a BBC basic program under Riscos that used the Raspberry Pi's BCM2835 undocumented random number generator - no problems. You can see how simple the source code is here:
Back in the iPhone 4 era, I had a simple app on the App Store called "3D Coin Toss" that I wrote in a day. With zero promotion, it brought in a predictable ~$700/quarter from ads and an IAP to disable them.
Interestingly, all my discoverability came from Google. My app was on the first page of search results, which drove users directly to the App Store.
Then, Google decided to compete. Searching for "coin toss" started returning Google's own top-of-page inline coin-tossing app as the very first result. Users could now toss a coin without leaving search results. Unsurprisingly, my user acquisition tanked.
It was my first experience with this, and I remember thinking, "Is this fair? Why is Google competing with me?"
Google "Coin Toss" or "Toss a coin", before the search results you'll find a google-written in browser-app which allows you to toss a coin and get a result. Google literally entered the coin-tossing software solution market! Lol. But not lol.
Back then I was furious about it, and yet, did nothing. Forget legal action, I didn't even gripe about it online. This is the first time I've posted or told anyone about it outside of a few friends that didn't understand. I felt powerless to do anything about it.
Further - look at AI generated search results being introduced now... same pattern. Why click through when you're got what you need before search results...
I doubt it, but it does highlight how weird the notion of "competition" is when juxtaposed with the notion of "a search engine controls its own results page."
(Personally, I resolve the paradox with "The goal is for neither Google nor app developers to 'win', the goal is to make it as easy to flip a coin as possible." Is my keyboard manufacturer competing with both if they put a button in the corner of the keyboard that either lights an LED or doesn't when pressed? Does the coin in my pocket compete with all three?)
I think it's as simple as "if you have a platform - be it OS, marketplace or search engine" - it should be illegal to compete with your platform participants.
For example, this law would have played out with Microsoft not being able to create Word (look at the history of what they did to the Windows version of WordPerfect). Amazon would not be able to introduce their own products and compete with their platform sellers. Apple wouldn't be able to take great independent app ideas and assimilate them into their OS. Google wouldn't be able to make a coin tossing app when its core business was successfully creating discovery for mine.
Perhaps a law like this would have prevented the formation of the mega-tech corporates that we see now? It's so easy to compete if you own the platform.
Companies frequently aren't trying to build a platform. They're just trying to build products people will buy.
I think trying to carve up the world of possible creations into marketplaces like that is sacrificing progress on the altar of capitalism.
If Microsoft added a coin flip to the start menu, are they also competing with your app? If somebody makes a keyboard that has a button on it and when you push it It lights one of two LEDs, are they also competing and should the law stop them? Am I competing if I'm carrying a quarter in my pocket? At some point, there's no compelling societal interest to protect your app from more convenient solutions to the end user.
In general, protection against monopolies in the United States hinges on harm to the consumer. It's real hard to argue that things are worse for the consumer when Google makes the process of digitally flipping a coin easier than installing an app.
I see your point. From the user's immediate perspective, getting a coin flip without an extra click is undeniably easier. But that's zooming in so close that you miss the entire picture.
The real question is what happens when that logic is applied to everything. First, it's a coin toss. Then it's the weather. Then a calculator. Then flight prices. Then hotel bookings. Then product reviews.
Step by step, the platform that was built to be a portal to a rich and diverse ecosystem of creators becomes a wall that primarily shows you its own products. The "progress" you're describing is the progress of a single entity consuming the ecosystem that once fed it.
The ultimate harm to the consumer isn't a slightly less convenient coin toss; it's the eventual death of that vibrant, competitive ecosystem. My tiny app was simply the first course in the platform's long meal of consuming its own creators.
Yes, this is entirely possible. It's why American law generally centers anti-trust on consumer harm as the litmus test; if those hotel bookings all have to go through one place, and as a result the hotels are too expensive, that's an issue. This is why Amazon gets to exist (but has now been sued in 2023 because the FTC is seeing behavior that is probably rent extraction).
If Google started charging a quarter a coin-flip while leveraging its control over search to suppress the fact you'd made a coin-flip app that was free or flat-rate to purchase, there might be a case there under US law.
I appreciate your comment, it drove me down a rabbit hole :-)
You were right that under the old interpretation, my app had no case. But that rabbit hole led me to the news that the interpretation itself has just been successfully challenged.
Judge Mehta found Google liable. The court has officially validated that the 'free vs. free' self-preferencing that killed my app is, in fact, illegal monopoly maintenance.
Fascinating to see a legal system's 'unhandled exception' get patched in real time.
Scott Alexander wrote an article [0] on the mild antipsychotic effects of nicotine as the reason why such a high proportion of people battling psychosis smoke cigarettes.
There is the dopamine hypothesis for schizophrenia and new research suggest serotonin also plays a role. Nicotine increases the dopamine in the brain so it sounds like to me smoking would be a form of self medicating.
