WEBVTT

00:00:00.399 --> 00:00:06.680
man Google Chrome kind of sucks now right I remember when everyone was

00:00:04.400 --> 00:00:13.160
screaming stop using Internet Explorer Chrome is so much faster was that all a

00:00:09.639 --> 00:00:15.519
lie no actually Chrome was a revolution

00:00:13.160 --> 00:00:19.920
it was lightweight fast as I'll get out and it brought great features like tab

00:00:17.800 --> 00:00:24.400
sandboxing so that if one of your tabs froze your entire browser didn't have to

00:00:21.840 --> 00:00:30.519
crash like the old days so what changed then to turn Chrome into this bloated

00:00:26.480 --> 00:00:33.680
RAM devouring Behemoth maybe it was

00:00:30.519 --> 00:00:35.399
you me I mean I don't have the best

00:00:33.680 --> 00:00:39.320
browsing habits I leave way too many tabs open but Chrome has a way of

00:00:37.559 --> 00:00:43.760
dealing with that using Dynamic caching for the tabs I'm not using sure but is

00:00:42.000 --> 00:00:47.239
Google the one who installed all those extensions I mean I don't have that many

00:00:45.719 --> 00:00:52.480
I just have zoto for tracking my research language

00:00:49.160 --> 00:00:54.600
tool tele party and an ad locker could

00:00:52.480 --> 00:00:59.519
it be that his insatable thirst for extra functionality is bogging down his

00:00:56.520 --> 00:01:03.559
browser as it turns out yes I need to make some Chang

00:01:00.840 --> 00:01:08.119
but not every extension is made equal while some had no effect on browser

00:01:05.560 --> 00:01:13.680
performance others like last pass reduced performance by as much as 12% an

00:01:11.320 --> 00:01:18.680
effect that Stacks across your library of resource sucking add-ons with 2third

00:01:17.119 --> 00:01:23.400
of people using Chrome as a default browser it's clear that we need to see

00:01:20.960 --> 00:01:28.439
just how bad this situation is explore what can be done to fix it and talk

00:01:25.520 --> 00:01:33.159
about Google's solution including why it might be a bad one so quickly now to the

00:01:33.640 --> 00:01:39.799
lab wait this isn't the lab this is the

00:01:36.640 --> 00:01:41.479
segue to our sponsor ODU if you own a

00:01:39.799 --> 00:01:45.479
business you'll know what I mean when I say you probably have too many apps

00:01:43.320 --> 00:01:49.360
running at the same time with odu's all-in-one business software you can

00:01:47.240 --> 00:01:53.759
combine all of those into a single unified platform so click the link in

00:01:51.439 --> 00:01:57.200
the description to learn more our Labs team has a fair bit of experience under

00:01:55.600 --> 00:02:02.560
their belt testing video games and productivity apps but this is one of our

00:01:59.240 --> 00:02:04.280
first for into browser performance and

00:02:02.560 --> 00:02:09.840
how exactly you measure browser performance is a big question primarily

00:02:06.920 --> 00:02:15.160
because browsers are big programs chromium the open source framework that

00:02:11.760 --> 00:02:17.200
Chrome is based on is 31 million lines

00:02:15.160 --> 00:02:21.480
of code that is almost as much as the Linux kernel the core of Linux based

00:02:19.440 --> 00:02:26.360
operating systems and it's close to the estimated count for older complete

00:02:23.519 --> 00:02:30.640
operating systems like Windows XP which Probably sounds ludicrous right until

00:02:28.800 --> 00:02:35.760
you consider that I mean outside of games how many programs do you run that

00:02:33.080 --> 00:02:39.840
aren't in your browser and of those ones how many of them are just web apps that

00:02:37.400 --> 00:02:43.280
could run in your browser so the thing is modern browsers need to be able to

00:02:41.640 --> 00:02:47.920
support pretty much everything from 28-year-old geoc cities pages to

00:02:45.400 --> 00:02:50.480
full-blown Photoshop clones so coming back to the question of how to measure

00:02:49.640 --> 00:02:56.599
their performance it's challenging because browsers do a lot of different things

00:02:54.720 --> 00:03:00.760
but we're going to do our best starting with a machine that is based on the most

00:02:58.400 --> 00:03:05.879
common specifications in a recent steam Hardware survey to look at memory use we

00:03:03.799 --> 00:03:11.519
opened up Chrome to a set of common web pages and on a clean system measured our

00:03:08.519 --> 00:03:13.799
idle RAM usage at around 1.1 GB which

00:03:11.519 --> 00:03:18.920
increased to 1.25 when we were streaming 4K video even on a system with just 8

00:03:16.720 --> 00:03:22.879
gigs of RAM that's not a problematic amount of usage and it's important to

00:03:20.799 --> 00:03:28.159
note that even if it was higher that isn't necessarily A Bad Thing programs

00:03:25.879 --> 00:03:32.879
that need to access information quickly will benefit from using up more R and as

00:03:31.040 --> 00:03:38.120
long as they give it up in the event that your Os or another program needs it

00:03:35.400 --> 00:03:42.640
it's not really a problem but what happens when we add extensions well as

00:03:40.760 --> 00:03:47.000
it turns out it depends on the type our password manager last pass and our

00:03:44.560 --> 00:03:52.040
thirdparty spell checker grammarly both used considerably more RAM than say

00:03:49.560 --> 00:03:56.879
coupon code collector honey one extension insert usual disclaimer about

00:03:54.560 --> 00:04:01.120
the impact on creators actually reduced our overall RAM usage and that was

00:03:58.840 --> 00:04:05.680
ublock origin which is clearly effectively blocking ads and trackers

00:04:02.959 --> 00:04:10.040
from sapping our system resources though funnily enough the more popular ad block

00:04:08.000 --> 00:04:13.280
extension didn't have the same effect which kind of makes us question its

00:04:11.519 --> 00:04:17.799
Effectiveness especially when they advertise that their extension saves

00:04:15.199 --> 00:04:23.400
memory overall with 10 of the most popular extensions we observed a nearly

00:04:19.840 --> 00:04:25.240
50% increase in RAM usage with half a

00:04:23.400 --> 00:04:30.759
dozen tabs open that's not that big of a deal but as your tab addiction grows so

00:04:28.360 --> 00:04:36.880
unfortunately does your Problem by the time we reached 48 tabs we were looking

00:04:32.960 --> 00:04:38.720
at a whole 2 GB of extra RAM usage for

00:04:36.880 --> 00:04:43.160
what is functionally the same user experience I mean you're not looking at

00:04:40.680 --> 00:04:46.080
all those Windows at once are you I mean of course if you have ample system

00:04:44.720 --> 00:04:51.680
memory you're probably not going to notice the difference anyway will you

00:04:48.680 --> 00:04:53.800
actually you might to gauge performance

00:04:51.680 --> 00:04:58.560
of our browsers we ran several benchmarks that aim to test the many

00:04:55.759 --> 00:05:02.360
features of the web in motion Mark which measures Graphics per performance in our

00:05:00.160 --> 00:05:06.440
browser most of our extensions caused virtually no change in performance

00:05:04.240 --> 00:05:10.960
except this is kind of a funny one our ad blockers you see ad blockers need to

00:05:09.160 --> 00:05:14.520
interact with the rendering pipeline in order to hide ads without turning your

00:05:12.759 --> 00:05:19.199
web pages into a completely fragmented mess and that appears to have resulted

00:05:16.520 --> 00:05:23.280
in a reduction in in browser rendering performance now in the real world this

00:05:21.560 --> 00:05:27.400
slowdown is going to be masked by other speed benefits but we thought it was an

00:05:25.199 --> 00:05:30.960
interesting observation nonetheless moving on to speedometer a benchmark

00:05:29.280 --> 00:05:35.280
that's f focused on simulating user experiences like working with a to-do

00:05:33.199 --> 00:05:40.039
list editing Rich Text rendering charts and reading news sites we found a

00:05:37.280 --> 00:05:44.600
totally different story here it's last pass causing a big slowdown and our best

00:05:42.520 --> 00:05:49.680
guess is that this is due to last pass trying to scan every text field for

00:05:46.919 --> 00:05:52.919
potential autofill opportunities in jet stream 2 a benchmark that's focused on

00:05:51.840 --> 00:05:57.479
smooth operation we found that ad block and

00:05:55.080 --> 00:06:03.080
grammarly did a lot of damage but then last pass was relatively benign and then

00:05:59.840 --> 00:06:05.080
in web expert 4.0 we measured just 3%

00:06:03.080 --> 00:06:09.240
reduced performance with all of our extensions enabled at once so sometimes

00:06:07.680 --> 00:06:15.120
your extensions matter sometimes they don't and sometimes different ones will

00:06:11.919 --> 00:06:17.479
or won't the results just don't really

00:06:15.120 --> 00:06:21.560
seem to add up to a clear picture but hey maybe the problem is we're looking

00:06:18.720 --> 00:06:26.479
too closely if we zoom out and look at the overall a clear Trend develops

00:06:24.599 --> 00:06:31.000
especially when we configure our browser like a regular person would and we have

00:06:28.520 --> 00:06:34.160
multiple extensions running concurrently with everything enabled we lose

00:06:32.639 --> 00:06:39.800
something like a quarter of our performance in the Chrome web browser if

00:06:36.280 --> 00:06:41.280
you're a major Surfer 25% that's the

00:06:39.800 --> 00:06:45.319
kind of difference that you could really notice in day-to-day use I mean think of

00:06:43.440 --> 00:06:49.919
all the extra LTT behind the scenes that you could watch on lm. g/ flat plane

00:06:47.479 --> 00:06:55.199
with all that extra time so what do we do about this well some extensions are

00:06:53.039 --> 00:06:58.720
always consuming some level of system resources and you can track this

00:06:56.720 --> 00:07:02.520
Yourself by pressing shift escape to open the Chrome tab task manager at the

00:07:01.160 --> 00:07:06.680
bottom of your list will be your extensions and writing this right now

00:07:04.560 --> 00:07:11.479
I'm noticing some of my extensions being needlessly active page downloader single

00:07:09.520 --> 00:07:15.599
file why is that open I'm not downloading anything but then the PDF

00:07:14.000 --> 00:07:18.960
editing extension that I'm using seems to respect the fact that I don't always

00:07:17.080 --> 00:07:23.160
need it running and it's not consuming anything now's a perfect time then to go

00:07:21.520 --> 00:07:27.080
through and call some of your more inconsiderate extensions especially if

00:07:25.400 --> 00:07:30.919
you aren't using them that much anymore all it'll cost you is a few seconds to

00:07:28.879 --> 00:07:34.560
click that little puzzle block and here's a good rule of thumb if you don't

00:07:32.639 --> 00:07:39.240
remember the last time you used it or what it's for just get rid of it you can

00:07:36.759 --> 00:07:43.479
always reinstall it later if you have lots of RAM and good tab management

00:07:41.319 --> 00:07:47.599
turning off the memory saving feature in Chrome will speed things up as well as

00:07:45.560 --> 00:07:51.879
Chrome will spend less time reloading Pages preloading Pages can also speed

00:07:50.159 --> 00:07:56.759
things up but if you're someone who uses an ad blocker or a privacy extension

00:07:54.039 --> 00:07:59.879
pre-loading isn't really an option speaking of privacy this is unrelated to

00:07:58.479 --> 00:08:03.520
your extensions but what you're in the menu why not stop by the add privacy

00:08:01.919 --> 00:08:07.159
section and turn all of that tracking off we can't really give you an after

00:08:05.560 --> 00:08:10.560
result for the impact of all the changes you just made I mean your mileage may

00:08:08.680 --> 00:08:14.360
vary depending on what you removed or didn't and your overall web browsing

00:08:12.400 --> 00:08:17.680
habits but from what we've seen anecdotally it is very possible that

00:08:16.440 --> 00:08:21.639
you'll notice a difference just not in startup times

00:08:20.400 --> 00:08:27.560
this was a really interesting part of our test that we're tacking on the end here we found that with a reasonable

00:08:25.400 --> 00:08:31.039
extension load out the application takes about the same amount of time to Launch

00:08:29.479 --> 00:08:35.440
regardless of what you do or don't have active then again I guess who cares

00:08:34.120 --> 00:08:40.279
anyway because when's the last time you closed Chrome I see that update button

00:08:38.680 --> 00:08:45.560
the last thing that you can do is to uninstall chrome and I'm only kind of

00:08:43.599 --> 00:08:49.480
joking right now the roll out of Google's manifest V3 is set to

00:08:47.720 --> 00:08:54.240
dramatically change the capabilities of chromium extensions which by the way

00:08:52.040 --> 00:08:58.040
also affect Microsoft Edge now especially hindering the effectiveness

00:08:55.760 --> 00:09:02.040
of ad blockers and privacy enhancing extensions we have haven't tested

00:09:00.079 --> 00:09:07.200
firefox's performance with you know extensions or without cuz realistically

00:09:04.760 --> 00:09:10.600
most of you aren't using it but hey maybe next time we do an update on this

00:09:08.920 --> 00:09:15.040
the situation will have changed because you'll have listened to Our advice why

00:09:12.680 --> 00:09:19.040
not give it a try and while you're at it why not try out our sponsor Odo if

00:09:17.480 --> 00:09:22.720
you're running a business keeping track of a hundred different apps all running

00:09:20.839 --> 00:09:27.079
at once can be a pain I mean just imagine all those Chrome tabs with odu's

00:09:25.279 --> 00:09:31.360
all-in-one business management software you can combine them into a single

00:09:28.839 --> 00:09:35.399
Unified operating system check out their CRM and sales apps which make lead

00:09:33.360 --> 00:09:38.880
generation and closing deals easier than ever their time sheets app helps you

00:09:37.399 --> 00:09:43.959
keep close track of your employee schedules and their accounting app lets

00:09:41.160 --> 00:09:49.200
you autocreate and send invoices and it integrates with 28,000 Banks from all

00:09:46.920 --> 00:09:54.000
around the world and if you only need one single app to shore up some holes

00:09:51.519 --> 00:09:58.640
hey you can get access to that for free so start saving time and resources by

00:09:56.040 --> 00:10:03.000
unifying your processes with ODU you can check them out in the description for a

00:10:00.279 --> 00:10:06.440
free 14-day trial or to set up a demo with their expert team thanks for

00:10:05.120 --> 00:10:11.640
watching if you guys liked this video maybe you'll enjoy another Deep dive into minor differences in computer

00:10:09.760 --> 00:10:15.920
performance check out our video where we tested 12 of the same CPU so that we

00:10:14.519 --> 00:10:20.760
could find a set of them that were nearly identical in performance why did

00:10:18.240 --> 00:10:25.720
we need those we'll never tell well we will in that video but not

00:10:23.200 --> 00:10:25.720
here