WEBVTT

00:00:00.160 --> 00:00:07.120
it's a fact of life that wherever you go someone is always trying to sell you

00:00:04.720 --> 00:00:13.440
something whether it's a brand hyping up their latest 1.21 gigahertz CPU or a

00:00:10.880 --> 00:00:17.840
so-called influencer rubbing your face in a brand new gaming setup that you

00:00:15.360 --> 00:00:22.320
would have to save for years to afford for the third time this week

00:00:20.080 --> 00:00:27.920
but i'ma let you in on a little secret come on come over here

00:00:24.800 --> 00:00:29.359
you don't have to buy anything in fact i

00:00:27.920 --> 00:00:34.480
bet the old crappy quad-core pc that you already

00:00:32.320 --> 00:00:39.680
have is good enough for anything you'd need it to do and if it's not you could

00:00:37.200 --> 00:00:45.840
get one of those for much cheaper than the fancy 8 or 16 core cpus that all the

00:00:43.680 --> 00:00:49.200
big players want you to buy let me show you

00:00:47.680 --> 00:00:53.600
after i show you today's sponsor glasswire are you lagging out while

00:00:51.280 --> 00:00:57.680
gaming use glasswire to see what apps are wasting your bandwidth and causing

00:00:55.199 --> 00:01:03.600
your games to lag get 25 off using offer code Linus at the link below

00:01:00.000 --> 00:01:03.600
and yes i appreciate the irony

00:01:10.560 --> 00:01:15.520
there have been many popular 4-core cpus over the

00:01:15.600 --> 00:01:23.439
there have been many popular 4-core cpus over the last 10 years and we can't test

00:01:20.720 --> 00:01:29.040
all of them but we do have a handful dating from today all the way back to

00:01:25.520 --> 00:01:30.960
early 2013. we've used an rtx 2080

00:01:29.040 --> 00:01:36.000
founders edition for all of our tests and we ran our gaming tests at 1080p so

00:01:33.680 --> 00:01:40.240
more of the burden will be on the CPU rather than the GPU i don't want to

00:01:37.920 --> 00:01:42.880
waste your time so let's jump right into the results where

00:01:43.200 --> 00:01:49.280
wow look at that i mean it sure is easy

00:01:46.320 --> 00:01:56.079
to show you what a piece of crap your old 4770k is in it the 3300x which

00:01:53.439 --> 00:02:01.360
represents a modern quad-core CPU manages to beat it handily and stepping

00:01:58.719 --> 00:02:05.960
up to six or eight cores especially with the 5800x which features AMD's latest

00:02:03.600 --> 00:02:09.679
zen 3 architecture is going to deliver a completely different computing

00:02:07.840 --> 00:02:14.480
experience that is as long as you actually use any of these

00:02:12.239 --> 00:02:19.440
applications the truth of the matter is that the benchmarks that we and other

00:02:17.120 --> 00:02:24.480
members of the tech press use to compare processors are not chosen for how

00:02:22.560 --> 00:02:28.959
representative they are of the typical user's daily workload rather they're

00:02:26.640 --> 00:02:34.720
often chosen because they allow us to demonstrate how their performance

00:02:30.959 --> 00:02:37.120
differs for heavy users or in future

00:02:34.720 --> 00:02:40.640
workloads that might not be mainstream yet but we expect them to be at some

00:02:39.040 --> 00:02:46.480
point down the road like there's a pretty good chance that you personally

00:02:42.720 --> 00:02:48.480
have never fired up v-ray or cinema 4d

00:02:46.480 --> 00:02:54.400
there's a good chance you don't edit or encode videos so if we flashed these

00:02:52.080 --> 00:02:57.360
charts in front of you you could easily end up spending a lot of money for

00:02:56.160 --> 00:03:03.680
something that you might never need even file decompression which is

00:03:01.440 --> 00:03:07.920
something that we all do often when we unpack a download from the internet and

00:03:05.760 --> 00:03:12.080
is a poster child for the benefits of multi-core cpus

00:03:09.760 --> 00:03:17.200
well it's a relatively light workload these days and most decompression tasks

00:03:14.560 --> 00:03:22.319
will complete in seconds on any of the cpus on this chart so

00:03:19.040 --> 00:03:24.959
does it really matter the two or three

00:03:22.319 --> 00:03:29.840
times a week that it comes up if it's 25 seconds or 45 seconds it's time for the

00:03:28.080 --> 00:03:35.599
other side of this coin while our cinebench r23 single thread web browsing

00:03:33.040 --> 00:03:39.360
grand theft auto 5 and cs go results demonstrate that there are certainly

00:03:37.200 --> 00:03:43.680
benefits to modern processors the addition of more cores

00:03:42.080 --> 00:03:47.519
isn't the biggest one and check this out if we covered the

00:03:45.760 --> 00:03:53.360
legend on these assassins creed valhalla results you'd be hard-pressed to guess

00:03:49.599 --> 00:03:55.599
which cpus have four six or eight cores

00:03:53.360 --> 00:04:00.560
go ahead and try but hold on a second Linus everyone knows that FPS and action

00:03:58.720 --> 00:04:05.040
games are still dependent on just a handful of threads you need to test

00:04:02.480 --> 00:04:11.599
strategy games that can use a multi-core CPU to calculate all the possible ai

00:04:08.080 --> 00:04:13.840
moves or whatever and yeah someday that

00:04:11.599 --> 00:04:17.919
will probably be a valid point but it is not this day

00:04:15.360 --> 00:04:23.040
and we did a quick civilization 6 run on our zen 2 cpus to demonstrate that point

00:04:20.880 --> 00:04:28.400
now with the latest gen consoles from the big players featuring eight core 16

00:04:25.840 --> 00:04:32.800
thread cpus now we are going to see developers really push to take advantage

00:04:30.880 --> 00:04:38.000
of everything they can squeeze out of them but if history has anything to go on

00:04:36.160 --> 00:04:42.000
it'll be a while before we see a game that was really built from the ground up

00:04:40.240 --> 00:04:46.160
for this generation of hardware and it'll be years before we see these many

00:04:44.320 --> 00:04:51.680
core designs used to their fullest potential which doesn't mean our video

00:04:49.440 --> 00:04:55.840
is over not even close i haven't even mentioned lttstore.com

00:04:53.759 --> 00:05:02.880
with its great designs and high quality shirts yet the thing is game consoles do

00:04:59.840 --> 00:05:05.280
pretty much one thing they game but your

00:05:02.880 --> 00:05:08.080
pc could be doing any number of things capturing video from your webcam

00:05:06.800 --> 00:05:13.759
removing ambient noise from your microphone feed encoding a stream to share with your friends or viewers what

00:05:11.520 --> 00:05:18.880
happens to our quad-core performance when we throw common background tasks

00:05:16.160 --> 00:05:22.320
into the mix so let's fire up premiere we're not working on it right now it's

00:05:20.160 --> 00:05:26.960
just sitting there uh we're gonna need discord a little bit of obs

00:05:24.479 --> 00:05:30.800
stream that gameplay right and love me some background youtube music playlist

00:05:28.880 --> 00:05:35.280
action all right let's do this again despite the fact

00:05:32.960 --> 00:05:39.360
that we had all of this extra software open in the background none of our cpus

00:05:37.840 --> 00:05:44.960
performed in a way that i would describe as being unplayable our older four core

00:05:42.479 --> 00:05:50.080
chips did struggle significantly more in cs go with all these programs running

00:05:47.520 --> 00:05:56.080
with the 4770k actually losing nearly half of its performance but that still

00:05:53.199 --> 00:06:00.080
left it at over 120 frames per second and that would be more than good enough

00:05:58.000 --> 00:06:04.479
for most people unless they're especially competitive in which case

00:06:02.240 --> 00:06:10.400
guys turn off your background tasks all right and moving to our more modern quad

00:06:06.800 --> 00:06:12.160
core 3300 x it lost only a marginally

00:06:10.400 --> 00:06:17.840
higher percentage of its performance when compared to the eight core 5800x

00:06:15.280 --> 00:06:23.520
in gta 5 the performance hit was even more negligible what this tells us then

00:06:20.160 --> 00:06:25.680
is that sure if you have more overhead

00:06:23.520 --> 00:06:30.720
overall there will be less to worry about when you open extra apps if you

00:06:28.160 --> 00:06:34.400
have a mini core CPU but as long as you're not going to be running 30 pieces

00:06:32.800 --> 00:06:38.560
of software while you're running your game and expecting to get the maximum

00:06:36.560 --> 00:06:42.400
FPS you're still going to get a completely playable experience

00:06:41.440 --> 00:06:48.960
but there are other advantages to overbuilding a gaming rig particularly

00:06:46.639 --> 00:06:52.960
with modern AMD processors if you allocate four of your course

00:06:51.199 --> 00:06:56.000
exclusively to your game and this is something you can do manually or there

00:06:54.479 --> 00:07:01.680
are tools to help you do it theoretically it could help smooth out

00:06:58.319 --> 00:07:03.919
the experience both by removing extra

00:07:01.680 --> 00:07:07.360
clutter from background tasks on those cores

00:07:04.880 --> 00:07:12.560
and by having all of the game tasks running on the same physical group of

00:07:09.840 --> 00:07:17.680
course for faster cache access at least in theory in practice we didn't observe

00:07:15.680 --> 00:07:22.000
nearly as much difference as we expected but some users on reddit have reported

00:07:20.160 --> 00:07:28.000
measurable performance improvements particularly in cs go with first gen

00:07:24.960 --> 00:07:29.919
ryzen chips so your mileage may vary the

00:07:28.000 --> 00:07:33.199
truth is though that if you don't have a fortune to spend on a gaming pc

00:07:32.000 --> 00:07:38.880
that's okay a chip from 2013 may come with unpatched

00:07:36.720 --> 00:07:44.400
security flaws that have scary names like spectre and meltdown and microsoft

00:07:41.919 --> 00:07:50.479
warns that older platforms experience 52 more kernel mode crashes in Windows

00:07:47.599 --> 00:07:56.080
11. but real talk spectre and meltdown are only of concern to enterprise and

00:07:53.120 --> 00:08:02.319
business customers for the most part and 52 more than point two percent is point

00:08:00.080 --> 00:08:07.520
three percent that sounds like a risk worth taking when a 4770k can be had for

00:08:05.039 --> 00:08:12.720
about a hundred bucks on ebay there is still the question of whether it's wise

00:08:09.599 --> 00:08:15.039
to buy into aging technology like ddr3

00:08:12.720 --> 00:08:19.120
when older ryzen 3000 chips can be had for not much more but that one is just

00:08:17.520 --> 00:08:24.160
gonna come down to your individual budget which we have no way of knowing

00:08:21.440 --> 00:08:28.560
what we do know is that four cores is still enough unless you actually need

00:08:26.800 --> 00:08:34.320
extra multi-core grunt for content creation or for non-gaming workloads

00:08:32.800 --> 00:08:38.560
just like you need to check out this sponsor message thanks ting mobile for

00:08:36.719 --> 00:08:41.919
sponsoring this video ting mobile has new rates that make it easier than ever

00:08:40.560 --> 00:08:45.279
to see how much you can save by switching they've got unlimited talk and

00:08:43.839 --> 00:08:49.600
text for just 10 data plans starting at 15

00:08:47.440 --> 00:08:54.880
with their set 12 plan which includes 12 gigs of data for 35 a month and

00:08:51.920 --> 00:08:58.240
unlimited data for 45 a month and if you liked their previous pay for only what

00:08:56.480 --> 00:09:01.440
you use plans they're still around they're called ting mobile's flex plans

00:08:59.760 --> 00:09:05.120
now and they charge just five dollars a gig data can even be shared if you have

00:09:03.440 --> 00:09:09.440
a family plan so you can connect more phones to save more anyone who signs up

00:09:07.760 --> 00:09:14.080
for a ting plan in september gets entered to win one of four ting branded

00:09:11.519 --> 00:09:18.880
yeti prize packs and everyone who signs up for an unlimited plan automatically

00:09:16.240 --> 00:09:23.760
receives a free yeti tumblr as well so check out the details at Linus.ting.com

00:09:21.279 --> 00:09:27.120
oh and you'll also get 25 in credit towards your account if you guys enjoyed

00:09:25.600 --> 00:09:32.480
this video maybe check out our ryzen 5000 review which has a bit more detail

00:09:29.279 --> 00:09:36.399
about how tasks pinned to certain CPU

00:09:32.480 --> 00:09:36.399
cores can have a performance benefit