YHK - yet another hacker news clone

Brood War Bench

57 points benswerd 6 hours ago 33 comments

benswerd 6 hours ago | parent

+ Playable Agent driven Starcraft

conorcleary 36 minutes ago | parent

Dibs on the fly brain

bee_rider 1 hour ago | parent

A ton of conversations about the game must be in the training set. I wonder, is there any way just from watching how they play, of telling if they tend to pick strategies that people complain or meme about online?

stackghost 58 minutes ago | parent

I wonder if there is a library to decipher brood war replay files. Perhaps an agent could learn by watching.

frutiger 56 minutes ago | parent

I haven’t checked for SC:BW but Blizzard has official parsers/replayers for SC2 on GitHub.

benswerd 41 minutes ago | parent

Not hard to build. I was shocked at how fast/easy this was to pull together.

malfist 1 hour ago | parent

This is a great idea for a benchmark. Something all the benchmarks seem to be missing is strategy, tactical solutions in most of the benchmarks are all thats required but here requires actual long term thinking and tactical thinking, balancing and orchestration.

shard972 47 minutes ago | parent

That’s why I tried making https://wrathbench.shard.page

Game_Ender 46 minutes ago | parent

Any details about the harness the agents were given? I am curious what representation of the screen and world state was provided to the agents and what tools they had available.

benswerd 42 minutes ago | parent

Oh sorry I should be more clear on that. Will add to report.

For agent harness I did Claude Code, Codex, Grok Build. This was primarily a cost driven decision — I have a lot of free tokens and I didn't want to pay API prices for this.

For game harness I used minimal BW-API issue command and get observation apis as tools. I felt this was the most fair way to do it on my small scale.

In the future I would like to integrate code mode and multiple games/I think if it was a best of 5 where each agent could learn from its past games and build its own automations over time that would be much more interesting.

pelagicAustral 46 minutes ago | parent

Unrelated to the benchmark...

I love StarCraft. I started playing it right from the beginning, most of my friends right now are from that era. I literally met people that have spread to almost every continent when I was in my early teens. We played at internet cafes and did not have access to the internet, that was priced differently...

I miss those days so much.

Everybody was from a different background back then, and nobody was anything other than a guy that plays StaCraft at the cybercafe... And now, we are in our 40's and I know Math teachers, history teachers, oil rig operators, software programmers, professional gamers, lawyers and more... hahah So crazy to think about it... and I know them, we talk, what a world.

sidewndr46 40 minutes ago | parent

Later we even listened to Eminem and played violent video games. Most of have never even been charged with a crime, much less abused someone.

benswerd 30 minutes ago | parent

I agree.

My first time playing StarCraft was at summer camp around a decade after it came out.

All the smartest people played it so I wanted to too. Great decision, I have been continually impressed with the people who StarCraft introduced me to.

nemo1618 30 minutes ago | parent

Even at Burning Man, in the middle of the desert, there is a camp that hosts a StarCraft tournament every year (on the dustiest setups you've ever seen!) :)

benswerd 30 minutes ago | parent

So dope

PorciiVorbesc 12 minutes ago | parent

So ... Zerging Man?

gary17the 18 minutes ago | parent

My best greetings to all Starcraft Elders clanners of yesteryear :).

snicky 3 minutes ago | parent

Oh boy, I've spent more hours playing it than I dare to admit. I won over 10000 battle.net games ... on just one of my several accounts :) When the SC2 beta came out, I played about 50-100 games and never bought the full game, because I knew it would be like heroin to me, and I was already an adult that had to take care of himself.

GodelNumbering 40 minutes ago | parent

A friend of mine created GoBench[1][2] that evaluates LLMs on 9×9 Go using KataGo opponents as Elo anchors, you see real capability differences there, like Astra Max substantially leading all other models. I think strategy is a generally interesting area to evaluate LLMs on

[1] https://rolandgao.com/blog/gobench/

[2] https://rolandgao.com/gobench.pdf

winwang 40 minutes ago | parent

Would be interesting if you could team a fast and slow agent together -- slow model can either act directly or maybe just communicate to the fast model.

benswerd 33 minutes ago | parent

I might open this up to a tournament if enough people want. Any interest?

dschuessler 39 minutes ago | parent

Somewhat related: In 2018, Google DeepMind had already created AIs that were capable of beating professional gamers in StarCraft 2 (the sequel to Brood War): https://www.youtube.com/watch?v=cUTMhmVh1qs

benswerd 34 minutes ago | parent

I predict LLMs will reach superhuman level and beat even that model in the next 12 months

orbital-decay 12 minutes ago | parent

Starcraft is APM-dependent. Unless the latency will improve greatly in frontier reasoning LLMs (which is unlikely), it will remain a bit like excavator knitting.

benswerd 11 minutes ago | parent

I predict latency will improve greatly in the next 12 months to more than 4x speed on current frontier tasks

loeg 10 minutes ago | parent

Yeah but Starcraft needs, like, 10-20x the APM these agents are doing.

benswerd 8 minutes ago | parent

I’m not convinced a lot of it can’t be solved with code mode.

Marine staggering for example seems like an ideal code mode task.

gadtfly 38 minutes ago | parent

Did it play by looking at screenshots and sending clicks, or was there other mediation/symbolization?

It sounds like it might have been actually played in real time, which would be very important to distinguish.

I have recently seen other harnesses letting agents play real-time games in what seems like discrete time slices, turning eg Portal into something turn-based https://www.youtube.com/watch?v=ruuGXFAmiOE

loeg 11 minutes ago | parent

There's a bot data stream already; it's probably hooked up to that rather than screencap. Yes, I believe these were playing in real time.

faeyanpiraat 35 minutes ago | parent

There is currently a bot beating everyone on the ladder. Just watched it today on Artosiscasts yt channel.

tweakimp 14 minutes ago | parent

If you want to see human written bots in action or compete in the bot ladder yourself, try https://aiarena.net/

minimal_action 5 minutes ago | parent

I think we're on the early days of games you connect with your agent to. Human + AI units one versus the other. Like knights with their horses. Not sure which is the horse..

AntiRush 4 minutes ago | parent

Back in 2010, during the early days of bwapi, there was a Brood Wars AI tournament held by the Expressive Intelligence Studio at UC Santa Cruz. It's interesting to see how different the approaches were back then, vs this or Deepmind's SC2 work.

https://web.archive.org/web/20091124210529/http://eis.ucsc.e...

There's a great contemporary Ars Technica piece by a competitor:

https://arstechnica.com/gaming/2011/01/skynet-meets-the-swar...

As an undergrad I did a project using genetic programming. It was not very successful, but it was a lot of fun.

https://tomisin.space/archive/starcraft-genetic-programming/