News for hackers. Stuff that matters.
Re:GPT-5.3-Codex by gallerdude on Thursday February 05, 2026 @03:15PM (#46904570)
Both Opus 4.6 and GPT-5.3 one shot a Gameboy emulator for me. Guess I need a better benchmark.
Re:GPT-5.3-Codex by paxys on Thursday February 05, 2026 @03:25PM (#46904717)
As coding agents get "good enough" the next differentiator will be which one can complete a task in fewer tokens.
Re:GPT-5.3-Codex by tgtweak on Thursday February 05, 2026 @03:36PM (#46904862)
Or quicker, or more comprehensively for the same price.
Re:GPT-5.3-Codex by nlh on Thursday February 05, 2026 @03:55PM (#46905167)
Or the same number of tokens in less time. Kinda feels like the CPU / modem wars of the 90s all over again - I remember those differences you felt going from a 386 -> 486 or from a 2400 -> 9600 baud modem.

We're in the 2400 baud era for coding agents and I for one look forward to the 56k era around the corner ;)

Re:GPT-5.3-Codex by well_ackshually on Thursday February 05, 2026 @05:01PM (#46906004)
There's hundreds of gameboy emulators available on Github they've been trained on. It's quite literally the simplest piece of emulation you could do. The fact that they couldn't do it before is an indictment of how shit they were, but a gameboy emulator should be a weekend project for anyone even ever so slightly qualified. Your benchmark was awful to begin with.
Re:GPT-5.3-Codex by plantain on Thursday February 05, 2026 @08:03PM (#46907686)
Your expectations are wild. Most software engineers could not write a game boy emulator - and now you need zero programming skills whatsoever to write one.
Re:GPT-5.3-Codex by randyrand on Monday February 09, 2026 @03:42AM (#46943066)
Ctrl-C + Ctrl-V. There. Done!
Re:GPT-5.3-Codex by nasreddin on Thursday February 05, 2026 @09:44PM (#46908370)
"a gameboy emulator should be a weekend project for anyone even ever so slightly qualified" do you really believe something so ridiculous?
Re:GPT-5.3-Codex by gf000 on Thursday February 05, 2026 @05:10PM (#46906100)
Is such an emulator not part of their training data sets?