WEBVTT

00:00:00.160 --> 00:00:07.520
logitech has been claiming for years now and Corsair recently joined them that

00:00:04.960 --> 00:00:10.960
their wireless gaming mice are indistinguishable

00:00:09.040 --> 00:00:16.960
from their wired ones because they've managed to get the processing latency so

00:00:13.840 --> 00:00:19.680
low that either one is less than one

00:00:16.960 --> 00:00:23.519
millisecond the fastest pulling or sample rate that's available over the

00:00:21.680 --> 00:00:30.560
usb connection and like that's pretty cool no cord dragging

00:00:27.920 --> 00:00:34.079
on your mouse hand has obvious benefits for gamers

00:00:32.000 --> 00:00:38.559
but the thing is is that we've mostly taken logitech's word for it up until

00:00:36.239 --> 00:00:42.160
now because we've never had any of the equipment that we would need to validate

00:00:40.559 --> 00:00:45.120
their claims well

00:00:43.760 --> 00:00:51.039
no more so we recently rented a phantom 4k

00:00:48.399 --> 00:00:55.760
high-speed camera from keslo camera for our recent collaboration with gav from

00:00:53.360 --> 00:00:59.359
slo-mo guys and we figured what the hey

00:00:56.960 --> 00:01:03.680
we've already got it we might as well use it oh that's hot

00:01:01.760 --> 00:01:07.520
speaking of hot this segway to our sponsor for today's video story blocks

00:01:05.360 --> 00:01:10.880
video storyblocks offers you studio quality stock video clips for a fraction

00:01:09.280 --> 00:01:22.000
of the cost check it out today at the link in the video description

00:01:22.000 --> 00:01:29.360
admittedly we haven't devised a way to intercept the signal from the mouse

00:01:27.360 --> 00:01:35.200
before it ever gets to the operating system video card driver game engine and

00:01:32.400 --> 00:01:39.759
finally your display so we aren't going to be able to address claims from brands

00:01:37.360 --> 00:01:44.079
like logitech and Corsair about their mice achieving sub 1 millisecond

00:01:42.400 --> 00:01:48.159
transmission speeds what we can do however is determine

00:01:46.560 --> 00:01:54.720
whether any of the mice that we're looking at today including both wired

00:01:51.040 --> 00:01:56.560
and wireless models are faster or slower

00:01:54.720 --> 00:02:00.560
in a way that could possibly be meaningful to gamers

00:01:59.040 --> 00:02:06.240
because here's the thing our perception of mouse lag or mouse

00:02:03.280 --> 00:02:11.599
latency is dictated by the total delay that we feel between moving our hand and

00:02:09.280 --> 00:02:15.840
seeing the corresponding action on the screen

00:02:12.720 --> 00:02:17.920
so as long as we can ensure consistency

00:02:15.840 --> 00:02:22.319
in all of our other variables any difference that we observe from one

00:02:20.000 --> 00:02:27.200
mouse to the next comes down to the signal transmission time whether it's

00:02:24.239 --> 00:02:30.319
along a wire or whether it's wirelessly to

00:02:28.080 --> 00:02:35.519
a dongle let's have a look at our test rig here we're running competitive

00:02:32.160 --> 00:02:38.480
shooter cs go on this Corsair vengeance

00:02:35.519 --> 00:02:43.680
gaming rig with a core i7 8700 and GeForce rtx 2080 we've paired this with

00:02:41.920 --> 00:02:49.599
one of the fastest gaming monitors around this 1080p 240hz one from acer to

00:02:48.160 --> 00:02:55.280
record each run and measure our total mouse to screen

00:02:52.400 --> 00:03:01.040
latency we're using a phantom flex 4k from vision research at 4k

00:02:58.000 --> 00:03:05.120
1000 frames per second so that means

00:03:01.040 --> 00:03:05.920
every frame here is just one millisecond

00:03:05.120 --> 00:03:11.680
now you might think that to get a precise

00:03:08.480 --> 00:03:14.080
measurement for each mouse we could just

00:03:11.680 --> 00:03:19.519
move the mouse around with our arms and then count the frame delay

00:03:16.319 --> 00:03:22.959
but not so at a thousand frames per

00:03:19.519 --> 00:03:26.159
second any organic body is gonna have

00:03:22.959 --> 00:03:28.640
too much give to get a clear start point

00:03:26.159 --> 00:03:33.519
so we had to get a little bit uh more creative

00:03:30.000 --> 00:03:34.560
meet norm so named for its two not or

00:03:33.519 --> 00:03:40.239
gates with alex's creation here we can swing a

00:03:37.360 --> 00:03:46.080
hammer at the mouse and with these conductive surfaces on the backs of our

00:03:42.640 --> 00:03:48.319
mice and on the tip of our hammer we can

00:03:46.080 --> 00:03:53.760
record the exact moment of contact with the indicator led

00:03:51.599 --> 00:03:57.360
that's on this breadboard here that removes the reliability of our

00:03:55.360 --> 00:04:00.959
measurements dramatically and it's pretty fun to do

00:03:59.439 --> 00:04:03.840
for our first round of testing we're going to keep things simple we're going

00:04:02.480 --> 00:04:08.720
to take one of logitech's high performance wired mice so this is their

00:04:06.000 --> 00:04:12.159
new generation mx518 with their hero sensor and we're going to put it head to

00:04:10.720 --> 00:04:17.280
head against one of their high performance wireless mice in this case

00:04:14.480 --> 00:04:21.519
the g703 so we're gonna do the wired one first because the conventional wisdom

00:04:19.359 --> 00:04:26.440
would go that that's your that's your baseline right first run mx518

00:04:29.759 --> 00:04:35.759
fantastic wow that's a really short delay

00:04:33.600 --> 00:04:40.400
our first run with the mx-518 taught us a couple of things number one is that we

00:04:38.160 --> 00:04:44.800
could expect with a wired mouse anyway about 15 milliseconds of latency

00:04:42.800 --> 00:04:48.479
according to our measurements and that we could expect a high degree of

00:04:46.320 --> 00:04:52.240
consistency one other key takeaway was that the creation of the apparatus to

00:04:50.560 --> 00:04:56.639
light up the led when the hammer hits the back of the mouse was entirely

00:04:54.320 --> 00:05:01.360
unnecessary as it turns out using different mode in premiere to examine

00:04:58.720 --> 00:05:06.560
two frames we could more easily identify the true starting point for our mouse

00:05:03.120 --> 00:05:09.120
movement i am fascinated by this test we

00:05:06.560 --> 00:05:15.600
really have to get the actual milliseconds this is really interesting

00:05:12.960 --> 00:05:19.680
that seems like a really big distance before it actually registers

00:05:17.759 --> 00:05:24.000
that doesn't seem right we're not making the rules here it is what it is now one

00:05:22.320 --> 00:05:28.560
special thing that we're making sure to do here is we're putting the dongle right here

00:05:26.720 --> 00:05:34.000
above the table to ensure that it's not subject to any unfair interference

00:05:31.840 --> 00:05:38.400
first wireless mouse test here we go with the g703 wireless we weren't quite

00:05:36.080 --> 00:05:43.120
sure what to expect logitech claims the performance should be basically the same

00:05:40.880 --> 00:05:47.600
but our common sense tells us that it couldn't possibly be especially with

00:05:45.280 --> 00:05:51.919
respect to performance consistency because wireless is far more susceptible

00:05:49.600 --> 00:05:56.160
to interference than a wired connection well as it turns out within the margin

00:05:54.000 --> 00:06:04.759
of error of our experiment they were the same wow that actually looked

00:05:59.039 --> 00:06:04.759
a bit faster but i might just be crazy

00:06:06.639 --> 00:06:12.639
so there you have it guys not only did we learn that our wireless mouse

00:06:10.240 --> 00:06:16.319
performed the same as our wired mouse within the margin of error of our

00:06:13.919 --> 00:06:21.520
experiment we also saw that they both performed incredibly consistently from

00:06:19.520 --> 00:06:25.600
one run to the next but since we've got the rig set up anyway we figured why not

00:06:23.840 --> 00:06:31.520
investigate some other manufacturer's mice so we're gonna start with Corsair's

00:06:28.400 --> 00:06:34.319
dark core RGB se

00:06:31.520 --> 00:06:39.360
we chose this one because it's Corsair's sort of equivalent sub one millisecond

00:06:36.800 --> 00:06:44.160
latency mouse and we wanted to know how their claims stack up

00:06:41.919 --> 00:06:48.160
Corsair's dark core RGB sc was an interesting one for us it was still

00:06:46.800 --> 00:06:52.720
within the margin of error of our experiment just as fast as both of the

00:06:50.479 --> 00:06:57.120
logitech mice but when we averaged our results it did measure a little bit

00:06:54.960 --> 00:07:02.240
slower the next thing we wanted to find out on our voyage of discovery today was

00:06:59.440 --> 00:07:06.880
whether it actually matters if a mouse can be set to a thousand hertz pulling

00:07:04.400 --> 00:07:12.800
rate so the final mouse ultra light sunset has a bit of a bad reputation for

00:07:09.680 --> 00:07:13.919
being capped out at 500 hertz does it

00:07:12.800 --> 00:07:19.599
matter only one way to know it actually tied for the lowest single

00:07:17.680 --> 00:07:23.919
response time but compared to the logitech mice it was less consistent

00:07:22.240 --> 00:07:27.840
could this be due to the 500 hertz pulling rate until we have a thousand

00:07:26.000 --> 00:07:32.400
hertz monitor it'll be hard for us to tell we're not done yet though for our

00:07:29.840 --> 00:07:37.759
last test we pulled out a logitech mx anywhere 2 mouse this is by no means a

00:07:35.520 --> 00:07:41.440
piece of gaming equipment but it gives us a great point of comparison for all

00:07:39.680 --> 00:07:45.440
the high performance gaming mice that we've looked at so far well

00:07:43.759 --> 00:07:49.440
this was a great lesson for gamers everywhere if you want to have the best

00:07:47.440 --> 00:07:55.360
most responsive experience you're going to want to choose a gaming mouse it was

00:07:52.240 --> 00:07:57.680
over double the average response time of

00:07:55.360 --> 00:08:02.560
all of our gaming grade mice and making matters worse that's with logitech's 2.4

00:08:00.639 --> 00:08:06.800
gigahertz wireless connection rather than bluetooth which introduces even

00:08:04.479 --> 00:08:10.479
further latency and more interference the final nail in its coffin for gaming

00:08:08.720 --> 00:08:15.199
is that the sudden acceleration from our hammer kick would cause the sensor in

00:08:12.560 --> 00:08:19.599
some cases to lose tracking and behave erratically so the conclusion here is

00:08:17.440 --> 00:08:23.360
pretty straightforward while i want to again acknowledge that we weren't able

00:08:21.680 --> 00:08:27.599
to test all the way down to one millisecond what we can take away from

00:08:26.000 --> 00:08:32.560
our experiment is that at least for logitech we have validated their

00:08:29.919 --> 00:08:36.880
position that anyone who claims to be able to tell the difference between

00:08:34.000 --> 00:08:41.599
their wired and wireless mice is either imagining it or full of

00:08:39.599 --> 00:08:45.200
but okay tell you what i will i will guess this one is faster i'll say this

00:08:42.880 --> 00:08:50.000
one's wired oh that's the wired mouse uh oh that's bad

00:08:47.120 --> 00:08:53.680
okay in fairness to me those mice had slightly different cables on them and

00:08:51.760 --> 00:08:59.760
that was probably what i was picking up on now a secondary observation here is that

00:08:57.279 --> 00:09:04.720
Corsair's wireless appears at least within the margin of error of our test

00:09:01.680 --> 00:09:07.200
to be on par with logitechs though it is

00:09:04.720 --> 00:09:10.800
notable that our average results were slightly worse

00:09:08.880 --> 00:09:15.200
our next conclusion is that anyone claiming to be able to tell the

00:09:12.080 --> 00:09:17.120
difference between 500 hertz and 1000hz

00:09:15.200 --> 00:09:22.640
pulling rates on a high-performance mouse is also imagining it even if you

00:09:20.880 --> 00:09:26.720
were capable of feeling a two millisecond difference in responsiveness

00:09:24.800 --> 00:09:31.760
your monitor because you don't have more than a 240hz display would not be able

00:09:30.080 --> 00:09:37.279
to display it finally do gaming mice have a value

00:09:36.000 --> 00:09:42.240
absolutely not only did our mx anywhere 2 which to

00:09:39.839 --> 00:09:46.800
be clear isn't even a terrible mouse and we weren't even using with bluetooth

00:09:44.240 --> 00:09:53.760
have measurably higher latency thanks at least partially to its 125 hertz pulling

00:09:49.920 --> 00:09:56.000
rate it also didn't track as well or as

00:09:53.760 --> 00:10:01.360
consistently perhaps thanks to its less sophisticated sensor or its power saving

00:09:59.279 --> 00:10:04.800
features speaking of saving

00:10:02.800 --> 00:10:10.480
i've been saving this segway for a long time this video is brought to you by

00:10:08.000 --> 00:10:14.320
storyblocks you can get studio quality stock video clips for a fraction of the

00:10:12.320 --> 00:10:17.760
cost with storyblocks video you can download all the stock video your heart

00:10:16.000 --> 00:10:22.560
desires from their member library including hd and 4k footage after

00:10:20.560 --> 00:10:26.800
effects templates motion backgrounds and more and all the content is royalty free

00:10:25.040 --> 00:10:32.560
so you can use it for commercial projects like youtube videos if your us

00:10:29.920 --> 00:10:35.440
or personal projects like youtube videos if you're us

00:10:33.600 --> 00:10:38.959
new clips are added regularly so there's always something fresh to download so

00:10:37.120 --> 00:10:42.000
check them out today at the link in the video description

00:10:40.640 --> 00:10:45.440
so thanks for watching guys if you disliked this video you can hit that

00:10:43.760 --> 00:10:50.240
button but if you liked it because you love science hit like get subscribed or

00:10:48.800 --> 00:10:54.079
maybe consider checking out where to buy some of the stuff we featured at the

00:10:51.839 --> 00:10:57.360
link in the video description also down there is our merch store which has cool

00:10:55.680 --> 00:11:01.279
shirts like the one i'm wearing and our community forum which is totally worth a

00:10:59.200 --> 00:11:05.040
join by the way i'm sorry the audio was so bad in these videos guys we're just

00:11:03.360 --> 00:11:10.760
we're in the science warehouse and it doesn't have

00:11:06.560 --> 00:11:10.760
it's not coated so