Rom Grk

But for objects, referential shallow comparison is the correct behavior, while for `NaN` it's incorrect. And I feel like disallowing `NaN` pushes a lot of responsability onto users, as `number`...

> The Preact 10.x line supports IE11, Object.is is non-viable for that reason. You could use a shim as a fallback, `Object.is || function is(...) { ... }`. > Passing...

I like preact and I'm ok with making efforts to support it, but I'm worried we're just hiding an issue that could pop up at runtime again. I see value...

You can add here, but I would recommend patching preact as it's a more robust way of fixing the issue, not only for the datagrid but also for any other...

Didn't notice. With the points data, adding the mask back would be quite simple: `maskRect.width = points[current].x - points[0].x`. I also see in argos that disjointed paths will need some...

I have added a commit to compute the points lazily as the animation runs, and the performance is terrible. It's at least 2ms for each point, very ugly, doesn't run...

The width is explicitly set to `110px` so it could be tied to the state & column width bugs that we've explored in other issues.

That function is only called at a single site, seems like it could be an excessive number of `updateDimensions` being triggered during mount. I'd take it but I'm on linux,...

If you copy the [measureScrollbarSize](https://github.com/mui/mui-x/blob/c4303217a3e25bb6aaefe5c79ff971e4a8348895/packages/x-data-grid/src/hooks/features/dimensions/useGridDimensions.ts#L385) function and set the prop `scrollbarSize` it could maybe improve performance easily on your side: ```javascript // calculate once const scrollbarSize = measureScrollbarSize(document.body, -1, undefined)...

No, we don't have enough information here. We'd need to at least be able to see the code to better assess what's causing it. It would also be nice to...