Arewefastyet

Discussion of general topics about Mozilla Firefox
Post Reply
User avatar
therube
Posts: 21703
Joined: March 10th, 2004, 9:59 pm
Location: Maryland USA

Arewefastyet

Post by therube »

FWIW. (So take it with a ...)

https://mozilla.github.io/arewefastyet-speedometer/2.0/

i5-3570k, 16 GB RAM, x64 browsers, i didn't look at memory usage
started with clean Profile each time, except as noted
didn't particularly limit my actions, but didn't necessarily
also sit on the test window the full time, exception noted
i guess you do need to understand the test :about, in particular
if you go comparing against other browsers or browser versions

Code: Select all

Firefox:
45	25.90	45.0.2 "dirty"
45	35.39	45.0.2 clean

56	46.94	56.0.2 clean
56	22.73	56.0.2 + noscript 5.1.7, temp.allow.github
   Iteration 1	46.43 runs/min   background actions ongoing
   Iteration 2	45.86 runs/min   window did not have focus
   Iteration 3	10.14 runs/min   i opened a different browser
   Iteration 4	5.514 runs/min   [not FF] session & was
   Iteration 5	5.496 runs/min   posting a thread on a board
   Iteration 6	5.463 runs/min   don't know why the iterations
   Iteration 7	5.467 runs/min   go as they did ?
   Iteration 8	11.25 runs/min   40-40-10-5-5-5-5-10-40-40 ?
   Iteration 9	44.96 runs/min   i only went back to the test
   Iteration 10	46.75 runs/min   when it was appropriate for me,
		                 so not necessarily when it finished
56	45.28	56.0.2 + noscript 5.1.7, temp.allow.github,
		       2ND.RUN, in forground, no other actions ongoing,
		       note: same results as with "56 clean"
56	42.02	56.0.2 + noscript 5.1.7 + uBlock Origin 1.4.16

57	53.67	57.0 clean
57	52.35	57.0 + noscript 10.1.2, temp.allow.github
57	47.80	57.0 + noscript 5.1.7 + uBlock Origin 1.4.16


SeaMonkey:
249	25.44	2.49.2
256	36.25	2.56a1
So, arewfastyet?
Well, per the test, 57 > 56 > 45 ~= SeaMonkey 2.56a1

Now, what does that mean - in the real world.
A world where "fast" is not an end-all.
A world where extensions are used, & do matter.
A world where people do care about the look of "their"
browser & how they're able to change it & make it look &
react the way "they" want it.

So, arewfastyet?
Hardly.

Do we gain by giving up.
No.
Fire 750, bring back 250.
Mozilla/5.0 (Windows; U; Windows NT 6.1; en-US; rv:1.9.1.19) Gecko/20110420 SeaMonkey/2.0.14 Pinball CopyURL+ FetchTextURL FlashGot NoScript
User avatar
therube
Posts: 21703
Joined: March 10th, 2004, 9:59 pm
Location: Maryland USA

Re: Arewefastyet

Post by therube »

About Speedometer

Speedometer measures simulated user interactions in web applications.

The current benchmark uses TodoMVC to simulate user actions for adding, completing, and removing to-do items. Speedometer repeats the same actions using DOM APIs — a core set of web platform APIs used extensively in web applications — as well as six popular JavaScript frameworks: Ember.js, Backbone.js, jQuery, AngularJS, React, and Flight. Many of these frameworks are used on the most popular websites in the world, such as Facebook and Twitter. The performance of these types of operations depends on the speed of the DOM APIs, the JavaScript engine, CSS style resolution, layout, and other technologies.

Although user-driven actions like mouse movements and keyboard input cannot be accurately emulated in JavaScript, Speedometer does its best to faithfully replay a typical workload within the demo applications. To make the run time long enough to measure with the limited precision, we synchronously execute a large number of the operations, such as adding one hundred to-do items.

Some browser engines use an optimization strategy of doing some work asynchronously to reduce the run time of synchronous operations. While returning control back to JavaScript execution as soon as possible is worth pursuing, a holistic, accurate measurement of web application performance involves measuring when these related, asynchronous computations actually complete. Thus, Speedometer measures the time browser spends executing those asynchronous tasks in Speedometer, estimated as the time between when a zero-second delay timer is scheduled and when it is fired.

Note: Speedometer is not meant to compare the performance of different JavaScript frameworks. The mechanism we use to simulate user actions is different for each framework, and we’re forcing frameworks to do more work synchronously than needed in some cases to ensure run time can be measured.
Fire 750, bring back 250.
Mozilla/5.0 (Windows; U; Windows NT 6.1; en-US; rv:1.9.1.19) Gecko/20110420 SeaMonkey/2.0.14 Pinball CopyURL+ FetchTextURL FlashGot NoScript
User avatar
therube
Posts: 21703
Joined: March 10th, 2004, 9:59 pm
Location: Maryland USA

Re: Arewefastyet

Post by therube »

The Lastpass Extension Slows you down

Confirmed.
New Profile, install Lastpass, open & test:

Code: Select all

57	24.88	57.0 + lastpass 4.2.1.21
Fire 750, bring back 250.
Mozilla/5.0 (Windows; U; Windows NT 6.1; en-US; rv:1.9.1.19) Gecko/20110420 SeaMonkey/2.0.14 Pinball CopyURL+ FetchTextURL FlashGot NoScript
User avatar
Omega X
Posts: 8225
Joined: October 18th, 2007, 2:38 pm
Location: A Parallel Dimension...

Re: Arewefastyet

Post by Omega X »

Something similar was reported for uBO and ABP. You pretty much have to turn off addons when running those tests.
User avatar
therube
Posts: 21703
Joined: March 10th, 2004, 9:59 pm
Location: Maryland USA

Re: Arewefastyet

Post by therube »

Well, for what I've done, I did run with no extensions. And with.

So you have an artificial test & you're expected to run it in an artificial manner (only) ;-).
At least running it with & without, may give you some indication of (presumably) the impact an extension may or may not have when running your browser (err, the test).
Fire 750, bring back 250.
Mozilla/5.0 (Windows; U; Windows NT 6.1; en-US; rv:1.9.1.19) Gecko/20110420 SeaMonkey/2.0.14 Pinball CopyURL+ FetchTextURL FlashGot NoScript
User avatar
Frank Lion
Posts: 21173
Joined: April 23rd, 2004, 6:59 pm
Location: ... The Exorcist....United Kingdom
Contact:

Re: Arewefastyet

Post by Frank Lion »

Omega X wrote:Something similar was reported for uBO and ABP. You pretty much have to turn off addons when running those tests.
Yah, yah, whatever.

I am assured by Chris Beard, CEO of Mozilla, that Firefox 57 is twice as fast as something that is half as fast and that's good enough for me - https://twitter.com/cbeard/status/930450524624838658

Mind you, it was also KellyAnne Beard who claimed that a recent Mozilla rally was attended by more people than any other rally in human history. Period.
"The only thing necessary for the triumph of evil, is for good men to do nothing." - Edmund Burke (attrib.)
.
User avatar
mightyglydd
Posts: 9813
Joined: November 4th, 2006, 7:07 pm
Location: Hollywood Ca.

Re: Arewefastyet

Post by mightyglydd »

@
Firefox has great customization tools built right in?
https://twitter.com/asadotzler?ref_src= ... r%5Eauthor
#KeepFightingMichael and Alex.
User avatar
Frank Lion
Posts: 21173
Joined: April 23rd, 2004, 6:59 pm
Location: ... The Exorcist....United Kingdom
Contact:

Re: Arewefastyet

Post by Frank Lion »

Asa Dotzler‏ @asadotzler
Nov 25

Did you know that Firefox has great customization tools built right in? Go to the 13 year old Firefox 0.8 PR menu and click Customize... On that screen you can drag and drop things from where you don't want them to where you do want them. You can also change Themes and toggle toolbars. Btw in other breaking news, Queen Victoria passed away a couple of years back
"The only thing necessary for the triumph of evil, is for good men to do nothing." - Edmund Burke (attrib.)
.
User avatar
mightyglydd
Posts: 9813
Joined: November 4th, 2006, 7:07 pm
Location: Hollywood Ca.

Re: Arewefastyet

Post by mightyglydd »

Frank Lion wrote:I am assured by Chris Beard, CEO of Mozilla, that Firefox 57 is twice as fast ...
Oh ! it's over twice as fast !!!!!...
https://blog.mozilla.org/blog/2017/11/1 ... x-quantum/
#KeepFightingMichael and Alex.
frg
Posts: 1361
Joined: December 15th, 2015, 1:20 pm

Re: Arewefastyet

Post by frg »

therube I just did some additional measurements with SeaMonkey in a Virtualbox vm.

Score - Ver
15.97 - 2.49.2
27.72 - 2.53
29.54 - 2.55

They more or less match yours. I don't see much speed improvement with 2.55 (Gecko 58). All in a vm so lower scores because of the additional overhead. If I find some time I will try with enable-webrender but don't expect much improvement here with only one content process and e10s off.

I get around 20.17 in the real world with 2.49.2 but with 2 E5-2670. Looking at the i5-3570k it has a much higher base and max. frequency so the approx. 25% difference compared to your 2.49.2 measurement fits right in.

My interpretation of the stats would be that 56 is already quite fast compared to the old version and 57 is around 10 to 25% faster in synthetic benchmarks like this one only.

FRG
Post Reply