Grasshopper 2 — First Impressions

Grasshopper 2 does not feel like an update. After an afternoon of building a small statistical definition, the honest impression is that of an entirely new piece of software that happens to share a family name with the one I have been using for years.

01The small things you notice first

The scroll wheel now works in the opposite direction. It is a detail, but it is the kind of detail that interrupts a decade of muscle memory every few seconds. Component alignment tools, which had become part of how I keep a definition readable, are not there yet — so canvases drift out of order faster than one would like.

02Panels are for displaying, and nothing else

The panel has been reduced to a pure display device. It shows values and, usefully, their metadata: a single number appears alongside its data structure, so a value of twelve arrives labelled as a three-by-four set. It was also the only component able to render the curve coming out of the new Pareto Distribution component in a legible way.

What the panel no longer does is act as an input. Typing a list of numbers into it and wiring it forward — the fastest improvisation in the old Grasshopper — is gone. Inputs now require the proper component: a Number parameter from Params, switched to accept multiple values through 'Set Many Values'. It is cleaner, more explicit, and considerably slower to reach for.

03A generous statistics library

The compensation is a genuinely rich set of statistical components. Distributions, including Pareto, now sit natively on the canvas, which turns a category of economic and analytical modelling into something you can assemble in minutes rather than script. Familiar operations have also been renamed and relocated — list item retrieval is now Item From Index — and right-click menus carry new sections for state, metadata mixing and baking.

“It is a beautiful piece of work — and a reminder that fluency is not transferable.”

Overall the interface is elegant and the engineering effort is evident. But a considerable amount of relearning is required before basic workflows feel automatic again. That is not a criticism of the software so much as an accurate description of the road ahead.