People love to bash developers who put up "Chrome recommended" banners, dismissing them as lazy, but when making games and such there are real issues that are impossible to circumvent. This bug is a prime example of that.
The point is that if shadowBlur performed poorly in Chrome, people would either find a way to circumvent it, or wouldn't use it. Chrome gets a free pass for its failings, while Firefox attracts "Chrome Preferred" banners, and this is based on market share rather than software quality.
For many aspects of web development you're absolutely right, but don't forget that this is a demo. It's a showcase of talent rather than an attempt to write something accessible in every browser.
People have shown demos at parties that require a specific graphics card and a specific build of a driver in the past; something that requires a specific class of browser engine is probably OK.
I'm dealing with a similar issue in Firefox (https://bugzilla.mozilla.org/show_bug.cgi?id=925025) and there's simply no way to do what I want without using a blur filter. So yes, if it didn't work in Chrome my game wouldn't exist but not sure how that is of any comfort to anyone.
That sucks! From my knowledge of browser-based games, though, I think a lot of people are using webgl, where the solution would be to write a blur shader. If your game was written in a world where Firefox had 90% market share, you'd be more likely to go straight for webgl.
I sympathize with developers dealing with browser bugs, but a majority of "Chrome recommended" banners are not. For many cases they are used as a cheap excuse not to care about or even deter non-Chrome browsers at all, even when they do work mostly or completely fine. Demos, in comparison, tend to have much narrower requirements due to various constraints (many D3D demos only worked with specific graphic cards at least in the initial release, for example).
I agree but the last time the pitchforks went out on HN regarding this was for a game, so not very different from demos.
The issue I'm personally dealing with is for a game revolving around a feature that's only working properly in Chrome (see sibling comment for link to bugzilla). I've tried solutions for mitigating it but since I'm just one guy I've basically given up for now and have resorted to a warning for Firefox users.
In all honesty I don't see how it's different from only releasing a game on Xbox and not PlayStation.
I guess that was shapez.io [1]? The situation around WebRender is pretty unfortunate, as AFAIK it is blocked by the compatibility issue in macOS. I have played it in Windows Firefox just fine though.
Anyway, I think it's more about the wording: it's very fine that the developer hasn't tested much in other browsers (not every developer has resources to do so, sure), but saying a certain browser is required or recommended implies that the issue is something to do with that browser (which might or might not be the case). Shapez.io also doesn't support macOS in its standalone build, but the statement on that was more amicable and honest; I wish the developer did the same for non-Chrome browsers as well.
> In all honesty I don't see how it's different from only releasing a game on Xbox and not PlayStation.
Consoles generally have much different requirements (hardware, vendor integration etc.) than PC games. For me it's more like Steam vs. Epic Store which I found pretty frustrating as a gamer.