Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Firefox 94

mozilla.org

Post reply on HN
Print view View original

281–290 of 340 posts

Re: Firefox 94

#281

Post by rdslw »

_odey wrote: post #273

Quick question (and a bit of an aside): let's say I'm interested in a specific feature being implemented (profile switcher like in chrome, built in feature as in [1], not an add-on as in [2]). How do I go about getting more attention to potentially have this prioritized? Making an account in their bug tracker and spamming the threads sounds annoying, I'd personally get upset if someone did that to my repos or issue t…

TL:DR: don't waste your time. There are literally security features not implemented since 2017.

E.G: no proper support for webauthn (CTAP2) in firefox on macos (abysymal situation shown here: https://fidoalliance.org/fido2/fido2-web-authentication-weba...) linux not much better. Safari, Chrome, even Edge have it, while firefox is stagnating since 2017 (bug report: https://bugzilla.mozilla.org/show_bug.cgi?id=1530373 ) :-(

When I hear about features like Colorways being prioritized over things like proper webauthn support on important platform (4 years now) it makes me sad.

Would have helped personally, but moving parts of firefox to rust not even decimated the contributors margin, but probably reduced it 100x. and here we are with rs_webauthn lib lacking necessary fixes in security important area :(

p.s. I knot how my post sounds, but I'm frustrated with (some) mozilla decisions.

Re: Firefox 94

#283

Post by heresaPizza »

dt2m wrote: post #164

My daily driver is a 2014 MacBook Pro and I just recently switched to Firefox from Safari. It's so much faster it's not even a fair comparison – I feel like I've gotten a whole new computer. Even compared to Chrome it loads pages so much faster on old hardware. Props to Mozilla!!!

I have a 2015 Pro and even if Safari uses less battery, I feel the same thing about Firefox. It is fast and smooth and has less bugs than Safari. However the thing that made my laptop feel cool again is the overwhelming amount of extensions available. There are so many of them and they really improved the ux. Safari has way less and most of them are commercial projects by big techs that collect datas or paid extensions by indie devs that I still can’t 100% trust because they’re closed source. Most of the extensions you find with the search bar on firefox is open source.

Re: Firefox 94

#284

Post by rndgermandude »

throw10920 wrote: post #231
rndgermandude wrote: post #209

Earlier quoted context omitted.

>Excellent performance and reliability. Quantum delivered a lot of performance wins, and work continues at a rapid rate on reliability. Unfortunately, to be competitive on performance and reliability given Mozilla's funding required dropping XUL :( Not really true. For starters, they did not drop XUL with Quantum. In fact, they did not drop XUL to this very day[1]. The thing they did drop was XUL and low level access…

> Firefox performs better for me in general This is genuinely surprising for me. Would you mind showing me what your JetStream[1] numbers are? [1] https://browserbench.org/JetStream/

I don't care about synthetic benchmarks. And this one essentially only tracks js (and webasm) performance, not the overall performance, not even just web performance as this would entail DOM manipulation, layouting and rendering, which this thing doesn't really do. It may have some meaning when all you do every day is use websites that are mostly compute instead of mostly idle. Only very few users will do that, while most users will use mostly idle websites, me included, and then it doesn't really matter if Firefox or Chrome can execute a piece of code 5ms faster than the other. It mattered back in the sunspider days, when js perf was really bad, and Firefox 3 outperformed the IE of the time by a factor of something like 20. But we're at a state right now of highly optimized js engines, so a benchmark difference saying V8 is marginally faster than Spidermonkey isn't really meaningful over all to me and how I am using the browser.

I care a lot more about "lived" performance, and that's where Firefox performs better for me. YMMV.

E.g. I regularly have noticeable stalls in Chrome e.g. when doing page navigation or just switching already loaded tabs. Nothing massive, but I do not have the same thing in Firefox. I didn't mean to imply there is a real major difference. Firefox is just a little better.

But anyway, since you asked: I just very unscientificially ran the benchmark in Firefox and Chrome just now, wasn't motivated enough to make it entirely fair by shutting down other processes which may "steal" CPU time, etc, and the grand result was... Chrome's score was 1.103x the score of Firefox[0]. Chrome is faster in raw js perf, but at the same time, this is still misleading, I would say, considering the context I provided.

[0] I know that on more high-end systems than this shitty laptop I am writing on right now, e.g. systems like my actual desktop, the difference might be larger[1], but even then it still holds true for me that js perf differences are mostly noise in my daily use, and Firefox edges out Chrome in other areas that I actually notice.

[1] e.g. mozilla's own arewefastyet have Chrome at about 1.4x as fast in JetStream2, and they aren't running this on a shitty laptop like mine. V8 is a terrific js engine, I am not denying that. But js perf is only a small part of the entire picture.

https://arewefastyet.com/win10/benchmarks/raptor-desktop-jet...

Re: Firefox 94

#285

Post by filereaper »

wartijn_ wrote: post #202
filereaper wrote: post #161

It wasn't mentioned in the release notes but for Linux users EGL is enabled by default now. I've been manually turning this feature on so this is a welcome improvement. https://news.ycombinator.com/item?id=29047339

They've mentioned it without saying directly that that's what they've done. > And on Linux, we’ve improved WebGL performance and reduced power consumption for many users. It links to a page with some more info[0]. Exiting stuff. [0] https://mozillagfx.wordpress.com/2021/10/30/switching-the-li...

Thanks for this!

I found another article by Martin Stransky who worked on this feature and gives more technical details: https://mastransky.wordpress.com/2021/10/30/firefox-94-comes...

Re: Firefox 94

#286

Post by yoavm »

arianvanp wrote: post #52
mawise wrote: post #2

I'm still waiting for Touch ID support on Macs https://bugzilla.mozilla.org/show_bug.cgi?id=1536482

The lack of any work whatsoever on Webauthn is infuriating. Nobody seems to be actively working on implementing the spec at all.

WebAuthn with a YubiKey, using Firefox on Linux, works perfectly for me.

Re: Firefox 94

#287

Post by kd913 »

BMorearty wrote: post #96

I recently switched from Chrome back to Firefox after the news about Google forcing people to stay logged-in to Chrome for tracking purposes. I’ve switched in the past but always ended up switching back for one reason or another. This time I’ve actually been very happy with it and don’t think I’ll switch back soon. Better in Firefox: * Ctrl+Tab cycles through tabs in recently-used order. Just like Cmd+Tab. This featu…

No mention of the Multiaccounts Containers addon?

https://support.mozilla.org/en-US/kb/containers

How about the fact that ublock origin is more effective on Firefox?

https://news.ycombinator.com/item?id=24212176

Then there are the other features like TRR, first party isolation, ECH, actual video hw decoding support on Linux...

Re: Firefox 94

#288

Post by nacs »

utrack wrote: post #207
BMorearty wrote: post #189

Earlier quoted context omitted.

Yeah, search the preferences for the word “Sponsored” and uncheck anything you don’t want to see.

It's opt-in AFAIK.

It's opt-out for many -- for me it was enabled by default after an upgrade.

Re: Firefox 94

#289

Post by arianvanp »

yoavm wrote: post #286
arianvanp wrote: post #52

Earlier quoted context omitted.

The lack of any work whatsoever on Webauthn is infuriating. Nobody seems to be actively working on implementing the spec at all.

WebAuthn with a YubiKey, using Firefox on Linux, works perfectly for me.

They only implement ctap1/u2f .

CTAP2/FIDO2 part is completely unimplemented. And that's why e.g. TouchID doesn't work. Or newer devices that don't support the old u2f standard.

CTAP2 gives fancy features like enforcing a pin (which Firefox doesn't implement) or fingerprint (which Firefox doesn't implement).

Re: Firefox 94

#290

Post by BMorearty »

kd913 wrote: post #287
BMorearty wrote: post #96

I recently switched from Chrome back to Firefox after the news about Google forcing people to stay logged-in to Chrome for tracking purposes. I’ve switched in the past but always ended up switching back for one reason or another. This time I’ve actually been very happy with it and don’t think I’ll switch back soon. Better in Firefox: * Ctrl+Tab cycles through tabs in recently-used order. Just like Cmd+Tab. This featu…

No mention of the Multiaccounts Containers addon? https://support.mozilla.org/en-US/kb/containers How about the fact that ublock origin is more effective on Firefox? https://news.ycombinator.com/item?id=24212176 Then there are the other features like TRR, first party isolation, ECH, actual video hw decoding support on Linux...

My comparison list was based on things I care about / have noticed so far, not based on a full feature comparison. Multiaccounts Containers addon isn’t something I care much about. I do kind of like the Facebook container addon but would be fine without it.

Funny you mention uBlock Origin being more effective on Firefox. As I mentioned, it seems to block things I don’t want it to block when I’m using Firefox.

Post reply on HN