Marco Hutter
Marco Hutter
That `Cesium MeasureText` folder had literally been lingering on my Desktop for nearly two years now ("Maybe, one day, I'll have another look at this... 🤔"). Now, it has a...
It's not perfectly clear how thoroughly things should be tested, and how deeply ~"unexpected behaviors" should be investigated. Most of the following is just a deep dive into the craziness...
> confirm these are differences of "a pixel or two" and not something that will add up to larger problems on multi-character labels. This does not only apply to "multi-character...
> Mostly out of fear of exchanging the current error for silently breaking layout on more legitimate labels that just happen to contain whitespace. It's hard to imagine _how_ it...
> The most northern box (with 100% alpha (FF) value so fully opaque) is drawn somewhat blended in CesiumJS: The odd thing is that it looks fully opaque in terms...
The level of obscurity here is ... nearly fascinating. There is a _lot_ of context, but from the description and images, I'd go out on a limb and make a...
Yeah, that's a strange one...  Likely not related to ~"low-level aspects of the rendering", but beyond that, I won't even _try_ to guess what could be happening there.
Given that the `texturesReferenceCounterById` that this PR is about was [introduced by me](https://github.com/CesiumGS/cesium/pull/12331), I'd like to mention that I'd usually prefer a `Map` over a `{}`. Usually `LinkedHashMap`, FWIW... I...
The performance is one thing. When you talk about a Proxy, then I wonder: If this is a real `Proxy`, wouldn't this distort the performance results considerably? It would be...
The _video_ from the first post here looks different from what I've seen. But the screenshot in the linked forum post looks more like what I reported in https://github.com/CesiumGS/cesium/issues/13074 ,...