Elite for Emacs | Hacker News Reader

Elite for Emacs

sami.salkosuo.net

257 pointsImmortalin46 comments

I've always wondered if there is scope for a MMO version of Elite that is less ambitious (especially graphically) in some ways than the current Elite: Dangerous but still has the essence of Elite; I'm thinking 3D graphics with a retro vibe, maybe primarily text based menus for in-station activities (and certainly no desire for on-planet or first person shooter components). Even keep the generative universe of the original with its nutty planet descriptions.

Slightly unrelated: if you're into "3D graphics with a retro vibe", check out "Return of the Obra Dinn" by Lucas Pope (of "Papers, please" fame):

https://dukope.itch.io/return-of-the-obra-dinn

It has nothing to do with Elite or space adventures or anything like that, though.

His development blog is awesome too, I specially liked when he was working out the best way to create effective looking dithering for 3D objects.

And then a mathematician who happened to read the blog started to think about it and created a new dithering algorithm just for him[0], which was pretty awesome.

[0] https://forums.tigsource.com/index.php?topic=40832.msg121280...

Not MMO, OOlite can be "retro" with the right options on.

Extensions (aka "OXP" or "OXZ" in Oolite lingo) can give you in-station text based activities (the extension is named "Life in the Frontier" IIRC) and planet descriptions, among other things (there are about 400 extensions available from the integrated package manager, not counting older ones that haven't been ported).

Extensions are written mostly in Javascript and the community is very friendly and helpful. Last but not least, it's free and open source.

I second Oolite. And in terms of adding networking, a game following the original Elite model should be fairly "MMO friendly" in that each system is basically closed - players can jump in and out, but there's no continuous flight between systems (unless they've added that to Oolite since I last checked it out), so e.g. scaling it to multiple servers can be done on the star-system level up until there's quite a lot of players.

I haven't looked at the extensions mechanism - it'd be interesting if players could "claim" systems and dynamically control gameplay capabilities in their system. Would create challenges with e.g. the economy/trading etc. (preventing users from setting up a server that gives them infinite credits to bring back to other systems) - perhaps that'd actually be a suitable use for a blockchain...

(Not that I have time to spend on something like this, but I'll back it in spirit if someone hacks Oolite or similar into a MMO... I might back it with money too if someone where to want to crowdfund something like it)

but there's no continuous flight between systems

I'm working on a game in the spirit of Elite, but with continuous flight between systems, in the spirit of Star Control 2.

https://www.reddit.com/r/gogamedev

I actually don't really like the continuous flight between systems. I prefer the simplicity of Elite. The massive increase in complexity is one of the reasons I never liked the sequels. But don't pay attention to me - I'm unlikely to be representative when it comes to a gamer audience.

I actually don't really like the continuous flight between systems.

Are you familiar with Star Control 2's "hyperspace?" It's really not all that more involved than flying into stargates or using the map in Elite, especially for short hops. I wanted to get away from point to point stargates in order to avoid the creation of choke points. I wanted to eliminate "gate camping."

That might be interesting, though I also think that choke points around gates open up interesting game play in itself.

But note that Elite doesn't really need to have choke points around gates, as it doesn't have gates. It has system to system jumps, that leave you "somewhere" far out from the station.

In Elite it doesn't matter whether or not that "somewhere" is in theory the same spot or a random position near the planet, as the system, including any ships, is re-generated from scratch each time you enter.

So in a multiplayer version you could avoid gate camping simply by making it a random position, but as I mentioned I think choke points might be interesting.

Elite has a simple mechanism that is easily extended to prevent such choke points getting out of hand too: Reputation with local law enforcement.

Generally this means you want to do your dirty deeds far away from the stations, but it'd be easy enough to extend it towards the jump points, and make the station security pay attention if there are repeated reports of ambushes.

I'm currently working on an MMO that's in the same spirit but much less graphically ambitious. (2D 1980's "Arcade Space" -- especially vector graphics.) However, it's very ambitious non-graphically. Not only is there a procedurally generated universe of 2^100 locations, there will also be crafting involving a procedurally generated Tech Tree that's just as vast as the world. I'm also planning on incorporating user-scripting of most entities in-game and compiling those scripts to a format which can be subjected to Genetic Algorithms. The scripts and crafted items will be "productized" to be sold or licensed to other players with in-game currency. I currently have a server running on AWS, capable of supporting 70 players comfortably in the same instance, all at the same location. (You get O(n^2) when all ships are in the same location.)

The project is called "Emergence Vector."

https://www.reddit.com/r/gogamedev

I've always wanted to play a space mmo with some nice low poly graphics like this for example: http://kennethfejer.deviantart.com/art/lowpoly-spaceship-450...

looks cute

I know what you mean. Independence War 2: Edge of Chaos [1] is very representative of that. Unfortunately, it's single player (there is a multiplayer arena but it's not great).

A few years ago a friend and I were working on a spiritual successor to EoC. When the Star Citizen kickstarter was announced, we gave up - no chance to compete with that, and we found they could apply our vision with a lot more money and a much more serious team.

[1] GOG: https://www.gog.com/game/independence_war_2 - Highly recommended if you're into space sims.

Privateer would be fun transformed into a MMO, but I don't think graphics are much of an issue these days as you can pretty much scale for need.

The issue I see with Elite/Privateer/etc is that its hard to encourage group play which tends to be a near requirement for most MMOs to succeed.

If ships are the primary draw how do you group people up? Surely many would not want to just play a specific role or share a ship and even if they did there would be only so many cool positions to go around.

> The issue I see with Elite/Privateer/etc is that its hard to encourage group play which tends to be a near requirement for most MMOs to succeed.

I don't think it would have to be "massive", multiplayer with a dozen of players or so could be just fine if we talk about coop (aka PvE) only: you could form convoys with escorts, attack/defend bigger objectives than in solo play, etc.

Competitive multiplayer (PvP) brings a lot of issues, compromises and downsides; one of them being the ability to customize your game. I casually play Oolite, and it's key feature is precisely the ability to modify almost anything in the game. Those mods for instance let you escort an AI ship or be escorted by AI wingmen, which compensates a bit for the lack of multiplayer.

> If ships are the primary draw how do you group people up

When you look at it, this genre is not very different from an ARPG: you upgrade your ship just like to equip your character; your ship can be a freighter or a light fighter and it is similar to character classes. If you need more variety you can for instance introduce distinct damage types (energy damage with lasers and kinetic damage for missile) with specific defences (shields vs lasers and armour vs missiles). Or you can reserve certain capabilities like interstellar jumps to certain ships.

I'd love a modern remake of Space Rogue

Space Rogue was such an awesome game: it had action, it had (for Commodore64) revolutionary filled 3D vector graphics, but most importantly, it had an involved, mysterious, and compelling story. And it was a challenging game which made one play hard, but it rewarded that with lots of generated Endorphine. Anyone remember getting the neurostabilizer boosters, without the personal shield?

That is a game truly worthy of a remake, as is Battle Tech: the Crescent Hawks' inception. I'd also finally like to find my father with his Phoenix Hawk mech in a sequel to that game...

Doesn't EVE Online scratch that itch for you?

I'm thinking something that still has a meaningful dogfighting experience (like the original Elite - whereas EVE Online when I last played was more like an RTS?) and generally far more accessible.

But a 'casual' and accessible mix of Elite and EVE with a MMO persistent universe and retroish graphics is what I have in mind.

Whereas the new Elite: Dangerous as impressive as it looks (and I'd love to play it... just don't have a beefy PC or Mac at the moment) misses something I think by not having everyone flying in the same universe (I know the simulation is shared and that there is also a solo play mode available). I've wondered why they didn't adopt a monthly subscription approach to support the servers to make that possible?

I'm thinking something that still has a meaningful dogfighting experience (like the original Elite - whereas EVE Online when I last played was more like an RTS?) and generally far more accessible.

I am working on an MMO that has a meaningful dogfighting experience. It's not twitch and rapid perfect aiming like an FPS. Instead, it's about relative positioning. Also, it's 2D, using an Asteroids flight mechanic.

https://www.youtube.com/watch?v=nSlpbxv4zZc&list=PLDIbNZGhS5...

I think that was their fatal mistake. The peer-to-peer networking model has made the game feel lifeless and that none of your actions have a persistent effect.

Really wish David B had focused on making a fun solo version of the old game; or a decent online one, not this ... muddle.

yes, a mud version of elite could be quite interesting and without the expense of aaa level graphics.

I played that version in 1985.

You did? There was a MMO space trading/combat game in 1985? What was it called?

There was actually one from '88 at least. Sort of. But text based. Some would consider it a MUD: Federation [1] from IBGames (and its currently still running sequel Federation II.

It started on Compunet, moved to GEnie, then AOL, then the web. The guy behind it - Alan Lenton - is a really interesting character.

[1] http://ibgames.com/archives/fed/index.html#classic

Elite originally came out in 1984

Elite isn't a MMO game.

They also just released "Getting work done" for vim

That one can write elite in elisp is why one can write gnus, eww, CEDET, slack-mode, magit and thousands of other packages yet unwritten in elisp.

True, I think half the point of emacs is _not_ doing work but hacking on emacs instead :-)

See also:

M-x butterfly

M-x doctor

M-x tetris

M-x hanoi

etc. etc.

also M-x dunnet

is a text based adventure game in most versions of emacs.

This is amazing.

Anyone know of any other text-based trading style games? I think there could be a market for this, as even Eve Online is nicknamed "Spreadsheets in Space", and I wonder if there are other games that have embraced that and left out the quasi-dogfighting mechanics.

I'm working on a spaceflight game that features trading and can be operated from a text interface, though it features graphics by default. My vision is to be a bit more story-based than the procedurally-generated or MMO-type games would be, but it might be what you're looking for: https://technomancy.itch.io/bussard

It features programming as one of its core mechanics on top of the piloting features. I made a trailer for it here: https://p.hagelb.org/bussard-1.3.webm

My goal for the next beta is to make it so you can purchase an upgrade for your ship's computer in-game that would allow you to connect to it from an external program like Emacs over the nREPL protocol: https://gitlab.com/technomancy/jeejah You start the game without an autopilot routine, but eventually you find/write one, and at that point you can play from Emacs or Vim or the CLI or whatever. Right now the external connection only provides you with a REPL, but the protocol is extensible, and I plan to add features that would allow you to run more of a HUD-style REPL in your client of choice.

Moving away from clojure? I can't get to the itch.io site from work for some reason, but saw an older version github. What were some of the factors in choosing lua? Has it been weird, difficult, or liberating?

Not as much "moving away from Clojure" as "haven't needed to write any server-side code in a long time". I would still use Clojure for that, unless it was a socket-level cluster thing where BEAM would work better.

Lua is a great fit for this for several reasons. Mainly it's because I want to make it a game where you can learn to program, and teaching beginners Clojure is incredibly difficult. Lua's relentless simplicity really shines here. Teaching tables is a snap, and trying to teach lists, vectors, hashes, and structs in Racket makes my kids' heads swim. Secondly because the of in-game programming, you need sandboxing, and Lua has excellent facilities for this. And Lua's speed and access to the LÖVE game framework make it the best choice for games that I've found.

Lua has its problems, but in the context of games 90% of them can be summed up with "you had a nil where you didn't expect it" which is something Clojure is quite bad at too. I would say in some senses it's been liberating to work on a codebase that doesn't need to interact with the madness of the web, and being able to count my dependencies on one hand (and read through their sources exhaustively) is wonderful.

What were you trying to do on itch.io that didn't work?

I think it was just a corporate firewall issue. That manifests in odd ways that don't always point clearly to the firewall.

Oh, this is fantastic!

I've toyed with the idea of creating a browser-based version of the original Elite. Went as far as the galaxy generation, based on the C version. Then had trouble, because the algorithm is pretty much obfuscated (unintentionally) behind layers of bit manipulation. And you need to do it the exact same way, otherwise you won't get the iconic galaxy, starting on Lave and with Riedquat as the nearby anarchy.

From a quick glance, the elisp version is slightly more readable in some places. I guess I should try again.

Also, will dig up my Portal Ending recreated in Emacs, as I didn't know there were other crazy brains that found this sort of thing interesting. I wonder if I need to get Valve's permission first.

Elite in LISP, showcasing both serious coding skill and the power of LISP. If LISP can do this, what else would be possible if it could talk to one or more 3D accelerators, and send synthesised, mixed sound to one or more audio devices?

This article finds me in a middle of a dilemma, pondering whether to develop my web application with AWK through cgi-bin (functionally of course), or use ANSI common LISP...

Common Lisp can do all of those things, as it has bindings for SDL 2.0 (https://github.com/lispgames/cl-sdl2) and OpenGL (https://github.com/3b/cl-opengl). As a newly converted Lisp programmer, I think that, after you get past the unusual syntax, you'll be much more productive and have much more fun with Lisp than any other language. You should try it!

I worked on this for a few years with some others. A fun experience to make a cl library for games that would work on the three major platforms and several cl implementations. We got bored and abandoned it though. https://github.com/lispbuilder/lispbuilder/blob/master/READM...

Can you give an example from the code of how this shows the power of lisp? From reading the article it sounds like it's a port from the C version.

The second sentence leads me to believe we are being trolled here.

I mean that he was able to do this at all it shows what LISP can do, not literally. If you want to see the power of LISP though, http://www.gigamonkeys.com/book/practical-an-mp3-database.ht..., and look how trivial it is to add functionality without changing the structure or semantics of the program.

It would be very hard to make a programming language that was not capable of this.

That's pretty magical.

How elitist.

Reply on news.ycombinator.com