GPT-5.6 против Fable 5 - кто лучше делает игры?

GPT-5.6 Sol, Claude Fable 5, GPT-5.5, and Claude Opus 4.8 are given three identical tasks and create 12 games. I'm blind-testing the results, trying to guess the authors, and checking whether the new OpenAI model is better than Fable 5 at game development. Timestamps: 00:00 GPT-5.6 vs. Claude Fable 5 00:19 Intro 00:50 Games from the OpenAI press release 01:34 Preparing three prompts 02:35 Launching four neural networks 03:47 Blind test and rules 04:52 Test 1: Sea of ​​Thieves clone 05:12 Pirate game 1 07:48 Pirate game 2 09:30 Pirate game 3 10:57 Pirate game 4 13:16 Who made pirate games 14:19 Test 2: Spore clone 14:54 Evolution 1 16:53 Evolution 2 20:06 Evolution 3 23:25 Evolution 4 26:35 Who made the Spore clones 27:23 Test 3 Cutscene and Action 28:10 Lighthouse 1 30:06 Lighthouse 2 32:04 Lighthouse 3 33:27 Lighthouse 4 34:39 Who Made the Lighthouse Games 35:42 Key Takeaways 36:30 Summary Like and tell us which of the 12 games turned out better and which one won your personal ranking. 📲 Telegram channel: https://t.me/studioigor ❤️ Support the channel on Boosty: https://boosty.to/studioigor ⚡️ Donate: https://boosty.to/studioigor/donate #NeuralNetworks #GameDevelopment #VibeCoding