spider-mario

Those are still numerical scores and, as far as I know, essentially based on “does it solve the task correctly or not”. By “qualitative”, I meant for example some analysis of how it succeeds (e.g. quality of prose or code) or its failure modes (lack of persistence, etc.).

I’m tempted to say that ideally, we’d have more in-depth, qualitative reviews of AI models so that it’s possible to form an impression of how suitable each model would be to one’s particular use case (instead of just “this model gets you 34 intelligence per dollar; this one, 38” as if they could be reduced to such numbers), but that would be a lot of effort given the sheer number and frequency of model releases.

Dioxide?

> If you think about it, by the time you reached for the mouse and started aiming for the button you already pressed it with your finger. The touchpad reduces the travel of the hand but not the time to aim. Hitting something with a finger is always faster.

Depends on how accurate you need to be. How fast/confident you can be at hitting with a finger can easily drop off a cliff.

You can get it with a VESA mount instead of the base.

> CRT refresh rates is super high, usually starting at 120hz...

That’s not my memory of it at all. They were often set to 60Hz by default, and increasing the setting to 75Hz would make it noticeably less flickery.

> and now we are trying to do the same? you want to feel headache again?

The headaches were from when it was too low.

I recently tried MiMo 2.6 Pro on a simple app and was positively impressed (modular code with sensible tests). I then tried it on a more difficult problem (which Opus 5.5 solved in minutes) and it looped for 12 hours without making any apparent progress before I finally pulled the plug.

And a refreshingly straightforward UI compared to modern behemoths.

Doesn’t that understate it? Isn’t a sales guy still important, to figure out what to build and to actually get it in the hands of users? Or are we to believe that he just founded NeXT with zero idea of what he wanted the company to make, and the engineers just so happened to make it?

(« n’est-ce pas »)

Or for an analogy: “IKEA wants you to pay $50 to assemble furniture you bought”

Well… yeah. Or you can assemble it yourself for free if you prefer.