Let a billion videogames bloom

Everything game development: news, lessons, discussion

Weekly Links #310

08 March 2020 — No Time To Play

Hello, everyone! I started this week by taking that closer look at SFML promised last time. As the test is a success, I'm now seriously considering using it for a much-needed game port. First, however, to deal with a little side project that imposed itself on me. (Creativity works in strange ways.) Which I only started on Friday, after spending most of the week migrating old articles to join the new one above in the engine section of the website.

In unrelated news, this week I also wrote a mini-rant on Twine and community, while Emily Short shared some thoughts about the GDC cancellation, as part of her end-of-February link assortment. In the mean time, many more events of all kinds were canceled worldwide, prompting worries about the long-term effects on various industries. Gee, you mean outsourcing so much to just one country was a bad idea? Or for that matter making so much depend on a few huge annual events set up in rich countries, such that it takes ridiculous amount of money and planning to get there? And then you have all the private companies suddenly discovering the value of letting people work from home. It only took them 35 years to figure it out. Worse, it was fear that prompted the decision, after all the rational arguments were ignored.

In more cheerful news, the 7DRL Challenge took place this week. Details under the cut, along with comments on two long-form articles. Which I'm afraid makes for a very short editorial, but sometimes it can't be helped. Thank you for reading.


After starting the week with some write-ups of my own, it's time to look at what others are saying. This time a four-years-old retrospective of Final Fantasy VII, courtesy of Gamasutra's good habit of resurfacing old treasures now and then. At first I just wanted to round out the new RPG section, but there's so much to unpack in there, it's amazing. About the cartoonish graphics enabling a whimsy now lost, doubly so as newer games take themselves much too seriously. About the easy-to-follow plot and simple gameplay enabling characters to shine as they tell their own story. About the same characters being allowed to be flawed human beings, who don't know everything, and misremember what they do know. About limiting scope so you can focus on one part of the game world and give it justice.

About letting the game be a game instead of putting all the effort into cutscenes only to end up with a glorified movie.

It's the same advice I always give aspiring writers: make sure you have something to say. Know why you're telling that particular story. Why you care, because otherwise why should the rest of us? And build upon a foundation. Don't rush to erect a scaffolding before you know what shape it needs to take. And for that you need to look back as well as forward. If nothing else, to see how far you've come.


As the week ends, The Digital Antiquarian revisits the demise of Commodore, and there are so many business lessons contained in there it's hard to make a good summary. Let me try anyway:

Don't be a jerk. Money can't buy respect and trust.

Seriously, that's it. No matter what, business is a human activity, that takes place between people and is subject to the same rules as any other social interaction. You can only ignore that for so long before everyone starts turning their back on you.

Incidentally, the same story also illustrates why it's a terrible idea to write software for a specialized computer architecture that works like nothing else on the market. Computers are general-purpose by nature; take advantage of that. I said this before and I will again: the Spectrum had a well-rounded CPU and a bitmap display, both very slow. It gave birth to games like Total Eclipse, Knight Lore and Lords of Midnight, while other machines were flooded with the same-y platformers and side-scrollers that could run well on their oh-so-clever sprite engines. Read: custom chips bound to become obsolete a few years down the road no matter what, simply due to changes in manufacturing.

Swappable components are a lot less important, seeing how laptops overtook desktops, then tablets overtook laptops. It matters much more that the new GPU you get can still be programmed with the same API as the one before. And even OpenGL has been slippery for a decade now.


Last but not least: as I was saying, the 7DRL Challenge took place this week. In fact it's still ongoing as of this writing, and I'm yet to see even half of its 90+ entries. Still, couldn't help but notice how many of them are experimental. Like Fable on your Table, a game meant to be played with papercraft miniatures, using the computer as an assistant. Quite a few hacking games, too. Also SerpentRL, a turn-based roguelike based on Snake, of all things. Speaking of which, there are more trends to take note of:

  • many games use ASCII and/or are turn-based, with relatively few in real time;
  • conversely, most if not all have simplified control schemes: the alphabet soup appears to be a thing of the past, and good riddance!
  • many games are made from scratch, or at least with lighter frameworks and libraries, meaning I can play them for a change;
  • many entrants are at their first game on Itch, but there are also many veterans of the 7DRL.

The only thing missing from the picture is the RogueTemple community. I expected furious discussion of the 7DRL on their forums, but instead, a couple of timid announcements that just showed up are (so far) all the activity surrounding the only big yearly event dedicated to the genre. Guess most of the discussion has been taking place right there on Itch, but really?

Anyway, I can't seem to find any more games of note at the last moment, so enjoy the Sunday and see you next time!

P.S. In the end, the 7DRL topped out at 210 entries. My favorite remains Raccoon Librarian, by Chris L. Hall.

Tags: meta, rpg, classics, business, hardware, roguelike, community

Comments? Tweet