WEBVTT

00:00:00.080 --> 00:00:07.920
this reflex equipped monitor just has the tool built into it for measuring

00:00:05.040 --> 00:00:07.920
in-game latency

00:00:11.440 --> 00:00:18.640
remember this thing ba-bam a photo sensor

00:00:16.400 --> 00:00:23.600
i remember and i also remember asking NVIDIA oh so nicely to release it to the

00:00:21.359 --> 00:00:29.439
public and while that might not be quite what we've got here today what we do

00:00:25.840 --> 00:00:32.239
have is a display that has very similar

00:00:29.439 --> 00:00:35.280
functionality built right into it yes my friends

00:00:33.040 --> 00:00:39.360
this gaming monitor has a function that will tell you the exact delay between

00:00:37.680 --> 00:00:45.680
when you click the mouse and the photons actually hit your eyes

00:00:42.480 --> 00:00:47.920
so we thought wow that's pretty cool

00:00:45.680 --> 00:00:53.520
why don't we use it then to see which in-game settings have the biggest impact

00:00:50.879 --> 00:00:57.680
on our gaming experience and we're also going to see the impact of our sponsor

00:00:55.440 --> 00:01:01.120
storyblocks save time and money with storyblocks video they give you studio

00:00:59.520 --> 00:01:06.440
quality stock video clips for a fraction of the cost check them out today at the

00:01:02.640 --> 00:01:06.440
link in the video description

00:01:13.200 --> 00:01:21.600
an NVIDIA reflex display like our ASUS pg259 qnr does exactly what i said in

00:01:19.040 --> 00:01:26.400
the intro it measures the latency from the click to the frame actually reaching

00:01:24.479 --> 00:01:30.640
the display and the muzzle flash actually going

00:01:27.840 --> 00:01:35.439
critically what it doesn't do is measure the display's latency which is why we're

00:01:33.040 --> 00:01:40.000
still calling on NVIDIA to release the ldap as well

00:01:37.200 --> 00:01:44.560
but before we dive in let's go over how this works

00:01:41.600 --> 00:01:49.280
on the back of the display are usb ports now you might think oh well this is just

00:01:46.799 --> 00:01:52.399
a usb hub like on the non-reflex model of this exact monitor that you guys

00:01:50.880 --> 00:01:57.840
checked out a while ago and you would be right except that if

00:01:54.880 --> 00:02:03.119
you plug a supported mouse into it then plug the display's usb output into

00:02:00.479 --> 00:02:08.160
your pc like we've done right here the display will recognize precisely when

00:02:06.000 --> 00:02:13.040
the mouse button is clicked in theory before the pc even receives the signal

00:02:10.640 --> 00:02:17.599
it then measures the time between detecting the click and a change that it

00:02:16.000 --> 00:02:21.680
monitors on screen pretty straightforward so far but how do

00:02:19.760 --> 00:02:25.200
we tell the monitor where to look will it just monitor the entire image and use

00:02:24.400 --> 00:02:29.840
like ai to tell if something relevant changes

00:02:27.840 --> 00:02:34.160
i mean yeah it's NVIDIA so that's not even impossible to believe in the future

00:02:31.520 --> 00:02:37.280
but for now it's a little bit simpler what we do is go into the on-screen

00:02:36.239 --> 00:02:41.280
display and then we can tell it hey

00:02:39.920 --> 00:02:45.040
turn on the monitoring rectangle

00:02:43.200 --> 00:02:48.000
now this is already right but let's go ahead and have a look at some of the

00:02:46.400 --> 00:02:52.080
other options that are in here you can see you can try a couple of the presets

00:02:50.560 --> 00:02:59.360
centered right handed left-handed according to you know where you might expect your gun or especially a muzzle

00:02:55.440 --> 00:03:01.360
flash to be you can change the size

00:02:59.360 --> 00:03:05.040
make it bigger or smaller and you can also adjust the sensitivity once it's

00:03:03.280 --> 00:03:09.120
tuned every time we pull the trigger on our firearm or click the mouse we're

00:03:07.360 --> 00:03:12.879
gonna see a little millisecond readout in the bottom corner or if it's having

00:03:11.680 --> 00:03:17.760
trouble figuring out what it's looking at it'll just show a bunch of x's so

00:03:14.959 --> 00:03:23.280
let's dive in and see what our starting latency looks like here we are running

00:03:19.920 --> 00:03:24.640
at 640 by 480 i mean i guess

00:03:23.280 --> 00:03:29.599
did we want to start at a best-case scenario for latency or worst case well

00:03:27.040 --> 00:03:32.560
i guess this is pretty best case all right so what are we looking at here

00:03:31.120 --> 00:03:37.360
then there's there's a lot of fluctuation and that's normal

00:03:34.480 --> 00:03:41.680
21 milliseconds from click to seeing something happen that's insane 23. like

00:03:39.760 --> 00:03:45.200
that's from click to for like all the processing and everything that's less

00:03:42.879 --> 00:03:48.959
than a frame at 60 frames per second for something to happen on screen that's

00:03:47.360 --> 00:03:53.680
ridiculous like it doesn't sound like it's that fast but it is it's really

00:03:51.200 --> 00:03:56.879
fast if you're moving then uh the different shapes of the muzzle flash

00:03:55.040 --> 00:04:01.200
especially if you're like firing really rapidly yeah can cause it to

00:03:59.360 --> 00:04:04.879
misread i think all right what do you want to do next

00:04:02.480 --> 00:04:10.959
uh we could bump the resolution sure let's bump the rest 22 16.

00:04:08.480 --> 00:04:15.519
16. that's like

00:04:12.400 --> 00:04:17.120
a frame at 60 hertz but like

00:04:15.519 --> 00:04:20.639
the whole chain so games that have NVIDIA reflex

00:04:18.880 --> 00:04:26.080
technology actually have in engine optimizations to bring this time down

00:04:23.040 --> 00:04:28.320
but this reflex equipped monitor just

00:04:26.080 --> 00:04:31.120
has the tool built into it for measuring in game latency

00:04:33.840 --> 00:04:39.440
we also don't have ultra low latency enabled in the driver either really

00:04:37.919 --> 00:04:44.320
this is just stock settings in the driver yeah okay yeah let's do it that

00:04:41.919 --> 00:04:48.639
maximum performance i would not pay extra to have this

00:04:45.759 --> 00:04:52.720
feature in my monitor neither would i i might

00:04:49.919 --> 00:04:55.720
would you why you could just like uh especially

00:04:57.199 --> 00:05:02.160
yeah i don't have one of those liners right that's fair but they should just

00:05:00.639 --> 00:05:08.080
do this for 50 bucks and then you have something you can carry forward with you okay the problem though is that for this

00:05:04.880 --> 00:05:09.919
to work you actually do need to like

00:05:08.080 --> 00:05:14.320
wire up a thing to your mouse okay all right all right all right all right yeah but they could just sell you a mouse for

00:05:12.560 --> 00:05:22.960
that and then you dial it in and then you use your regular mouse 24 30 37 30

00:05:18.800 --> 00:05:25.440
26 24 31 19 21

00:05:22.960 --> 00:05:29.280
25 24. these kinds of fluctuations are actually really normal like you wouldn't

00:05:26.720 --> 00:05:32.960
normally see them or feel them right because like all of this kind of stuff

00:05:30.960 --> 00:05:37.039
is just like happening in the background and by the time it updates to the screen

00:05:35.440 --> 00:05:41.440
the screen is basically doing something else your eyes are living in the past

00:05:39.280 --> 00:05:46.400
more or less your eyes are living in the past Anthony 2020 next we need to figure

00:05:44.479 --> 00:05:50.320
out how to minimize brain lag who throws an incendiary grenade look at me i got

00:05:48.000 --> 00:05:54.880
fireproof feet i got sandals on bro okay you know what let's see how bad we can

00:05:52.400 --> 00:05:59.039
make it not helping matters right now or rather helping them is the fact that

00:05:56.320 --> 00:06:03.440
counter-strike is an esports title and has received significant optimization in

00:06:01.280 --> 00:06:07.199
this way now you can play games where you can really feel that there's a lot

00:06:05.680 --> 00:06:12.160
more engine lag even if you haven't changed anything else your graphics card

00:06:09.280 --> 00:06:15.680
mouse monitor etc i could turn off a and see what happens that's always something

00:06:13.600 --> 00:06:19.840
that has been a perceived difference for me is that turning on really high levels

00:06:17.680 --> 00:06:23.440
of anti-aliasing can make some games feel really mushy yeah hold on a second

00:06:22.639 --> 00:06:29.039
though does that look like a little bit less spikiness

00:06:26.800 --> 00:06:34.639
maybe yeah looks like it's hovering between the 20 to 30 range right now

00:06:32.240 --> 00:06:38.319
17 29 25. like i don't think i've seen anything

00:06:35.840 --> 00:06:40.880
significantly higher than 30 yet whereas before we were spiking as high

00:06:39.919 --> 00:06:46.639
as like 40-ish yeah

00:06:43.759 --> 00:06:51.919
interesting okay maybe anti-aliasing is still bad by like

00:06:49.520 --> 00:06:56.160
a few milliseconds hey look when you want every competitive edge i mean that

00:06:53.759 --> 00:07:01.280
is NVIDIA's marketing for this this is almost definitely a little bit better

00:06:59.280 --> 00:07:04.560
not you might feel only because i can look at the numbers Anthony i cannot

00:07:02.960 --> 00:07:08.319
feel the difference between those i was going to say like no no no no no when

00:07:06.639 --> 00:07:12.240
did you join the esports leagues yeah exactly right we could try a lower

00:07:10.479 --> 00:07:16.000
refresh rate um i don't know how

00:07:14.080 --> 00:07:19.759
much you care about that considering if you're buying a 360 hertz monitor are

00:07:17.759 --> 00:07:23.120
you going to lower the refresh rate no not unless you're an idiot

00:07:21.440 --> 00:07:30.240
so well with that said i don't think it supports ulmb at 360 hertz uh

00:07:26.880 --> 00:07:31.440
maybe not 240 only oh there we go it's a

00:07:30.240 --> 00:07:36.639
nice dim nice dim look for the display

00:07:34.720 --> 00:07:41.440
look i'm not saying i'm a ulmb hater david i'm just saying it's worse and you

00:07:38.960 --> 00:07:44.479
should feel that it's all about dyc bro if you're in a dim room a dim display

00:07:43.120 --> 00:07:48.240
doesn't matter that may be true but i try to not be my

00:07:46.240 --> 00:07:53.919
old gamer handle you know closet gamer i put my whole computer and

00:07:51.039 --> 00:07:58.199
monitor in my closet okay so let's have a let's have a looky-loo here

00:08:00.639 --> 00:08:07.680
oh yeah it's worse that that is worse it's definitely worse you are officially

00:08:05.199 --> 00:08:11.280
objectively worse david yeah

00:08:08.960 --> 00:08:15.360
get good scrub yeah not that i would want to play you but

00:08:13.039 --> 00:08:18.720
whatever 66 i mean we never saw anything even close to that lower hurts

00:08:16.879 --> 00:08:21.759
definitely more worse okay confirmed what's our next project that's that's

00:08:20.319 --> 00:08:26.560
actually kind of crazy given that it's 240 versus 360.

00:08:24.000 --> 00:08:30.400
like 360 only just became a thing i know right right

00:08:28.400 --> 00:08:34.000
big difference it does somehow

00:08:32.240 --> 00:08:37.279
oh i know you're getting one of these i'm going on the record right now if a

00:08:35.760 --> 00:08:41.599
480 hertz monitor doesn't come out in the next year david will own a 360 hertz

00:08:39.919 --> 00:08:47.920
monitor now let's play around with the title that actually has NVIDIA reflex

00:08:45.760 --> 00:08:54.000
so NVIDIA reflects low latency we're going to on plus boost i have no idea

00:08:51.200 --> 00:08:57.600
what that does okay let's just go on let's go vanilla on we're not going to

00:08:55.839 --> 00:09:00.480
use deep learning super sampling we're just going to

00:08:59.200 --> 00:09:05.279
run at low and then we're going to try cranking it to high

00:09:03.120 --> 00:09:09.680
fortnite oh gee i don't know Linus i have no idea

00:09:07.600 --> 00:09:14.800
what you're talking about fortnite how about that one

00:09:12.880 --> 00:09:18.880
disgraceful i could probably write a better search

00:09:16.480 --> 00:09:24.080
than this fun fact speaking of dos i actually set

00:09:20.959 --> 00:09:26.240
up an emulator for a pentium one 200

00:09:24.080 --> 00:09:29.519
megahertz with mmx the reason for it is i brought home the hot wheels pc and i

00:09:27.680 --> 00:09:33.120
was like oh this sucks doesn't it my kids were like these games are great

00:09:31.440 --> 00:09:36.160
oh okay yeah i guess content's still king by the way stay subscribed because

00:09:34.880 --> 00:09:40.480
maybe we'll do a video about that emulated old pc versus using dosbox

00:09:38.880 --> 00:09:45.839
versus just getting old hardware Anthony's wanted to do something like that for a long time dos gaming build

00:09:44.080 --> 00:09:49.120
constantly on the back burner still on my shelf this is a great chair because

00:09:47.839 --> 00:09:52.000
if i have to go to the bathroom i don't even have to move

00:09:51.120 --> 00:09:59.560
except i can fix that i didn't ask for that

00:09:56.200 --> 00:10:01.680
lttstore.com ltdstore.com

00:09:59.560 --> 00:10:05.920
13.8 milliseconds and look at the consistency

00:10:03.519 --> 00:10:09.839
that's the more impressive thing this isn't really changing at all like

00:10:08.240 --> 00:10:14.640
we're going from like 13 to 15 or 11 like that's within two or

00:10:13.040 --> 00:10:19.040
three milliseconds this this is nuts

00:10:17.360 --> 00:10:23.360
turn on uh either reflex plus boost or turn off low

00:10:21.920 --> 00:10:28.480
latency and see what happens all right let's try reflex plus boost

00:10:25.600 --> 00:10:34.160
okay this seems very similar yeah yeah it might be maybe a little bit

00:10:31.120 --> 00:10:38.360
more consistent it seems to more favor

00:10:34.160 --> 00:10:38.360
the 12 and a half 13.

00:10:46.399 --> 00:10:52.399
i don't know that may be placebo i think that's the

00:10:50.959 --> 00:10:56.320
kind of thing where if you can tell the difference you're an idiot

00:10:59.680 --> 00:11:05.120
yeah yeah it's pretty placebo i think at this

00:11:03.519 --> 00:11:08.880
high frame rate we're probably not going to notice a difference well why would cs

00:11:07.279 --> 00:11:12.079
go so much slower so you're saying fortnite was just already very optimized

00:11:10.640 --> 00:11:18.240
and all NVIDIA's done is add some branding to it uh it's probably that it actually just

00:11:15.600 --> 00:11:22.480
makes more sense at a lower frame rate right

00:11:19.839 --> 00:11:25.839
that's fair okay so let's turn rtx on that really does look a lot better and

00:11:24.720 --> 00:11:31.040
that's with all the other graphic settings on low yeah i know right

00:11:29.200 --> 00:11:35.040
you know you could make the argument i mean textures i would want to turn my

00:11:32.880 --> 00:11:39.440
texture quality up but other than that you could probably make the argument for

00:11:36.880 --> 00:11:43.279
just turning on ray tracing cranking almost everything else down

00:11:41.760 --> 00:11:50.000
and just enjoying that look at that soft lighting

00:11:45.440 --> 00:11:53.040
on his jacket that looks crazy good

00:11:50.000 --> 00:11:55.839
like that looks like the shade

00:11:53.040 --> 00:12:00.560
with a more cartoony game like this like the flatter look for textures and stuff

00:11:58.160 --> 00:12:03.519
actually looks fine yeah whoa

00:12:01.360 --> 00:12:09.120
that is way worse yes we're only at 200 FPS now

00:12:06.560 --> 00:12:12.720
so i guess that would be why but turning ray tracing on ain't gonna give

00:12:11.040 --> 00:12:17.440
you a competitive advantage that's for sure bud it's still consistent though

00:12:15.519 --> 00:12:21.279
and that's probably the most important thing i don't know i would take 10

00:12:19.120 --> 00:12:26.160
milliseconds over retracing would you take

00:12:23.200 --> 00:12:28.880
10 milliseconds over a more stable response time

00:12:29.120 --> 00:12:35.519
no but it was pretty stable at 10 millimeters it was stable before this

00:12:33.120 --> 00:12:40.079
this game seems ridiculously stable yeah uh try with rtx plus

00:12:37.680 --> 00:12:43.680
reflex see what that looks like it might make up the difference okay

00:12:42.000 --> 00:12:47.680
there we go oh wow it did actually make up for a lot

00:12:45.760 --> 00:12:51.279
of it yeah it's not quite where it was but it's like

00:12:49.120 --> 00:12:54.639
quite a bit better yeah that's pretty cool

00:12:52.880 --> 00:12:59.120
this experience with the product really doesn't make it any easier to recommend

00:12:56.800 --> 00:13:03.600
for or against it though on the one hand

00:13:00.399 --> 00:13:07.600
it's a 50 to 100 adder on your monitor

00:13:03.600 --> 00:13:08.639
that doesn't affect performance at all

00:13:07.600 --> 00:13:13.920
except that it can help you tune the

00:13:10.639 --> 00:13:15.279
performance of your system

00:13:13.920 --> 00:13:19.760
hold on guys you're not even going to talk about the coolest part come on the

00:13:17.680 --> 00:13:23.200
latest beta for GeForce experience gives you a whole bunch of performance metrics

00:13:21.360 --> 00:13:27.600
to gawk at check it out not only does it show you GPU performance stats and let

00:13:25.120 --> 00:13:33.279
you adjust power targets on the fly it also integrates directly with the NVIDIA

00:13:30.959 --> 00:13:36.160
reflex latency analyzer built into the display

00:13:34.399 --> 00:13:39.680
here's how it works up top we've got two stats that everyone should be able to

00:13:37.519 --> 00:13:44.160
get regardless of the display render present latency is how long it

00:13:42.079 --> 00:13:48.240
takes a frame to get from being rendered to being sent to the display while

00:13:46.320 --> 00:13:52.880
render latency is how long it takes to actually render the frame now if we turn

00:13:50.639 --> 00:13:57.279
on latency monitoring in our monitor's osd

00:13:54.160 --> 00:13:57.279
and wait a couple seconds

00:13:58.959 --> 00:14:04.800
bam check that out even more good stuff this is a list of all the latency

00:14:03.440 --> 00:14:08.320
information that the monitor is collecting we've got the exact location

00:14:06.639 --> 00:14:12.959
of the monitoring rectangle we've got the system plus display latency same as

00:14:10.959 --> 00:14:16.560
what the monitors osd shows us and if we've got a compatible mouse like this

00:14:14.399 --> 00:14:20.959
rotary shock RAM we've got mouse click and total system latency which is nearly

00:14:18.720 --> 00:14:24.560
as accurate if not just as accurate as ldap what does it mean for a mouse to be

00:14:22.800 --> 00:14:28.959
compatible with the reflex latency analyzer two things a validation

00:14:26.720 --> 00:14:32.720
database and firmware NVIDIA is compiling a list of average click

00:14:30.880 --> 00:14:36.320
latencies for existing mice in order to make as many of them compatible as

00:14:34.320 --> 00:14:40.639
possible but some mice are able to capture the exact moment a click is

00:14:38.160 --> 00:14:44.880
registered in firmware and transmit it to the display on each click NVIDIA is

00:14:42.880 --> 00:14:49.279
launching with support for the 30 most popular esports mice on prosettings.net

00:14:47.360 --> 00:14:53.760
with per click latency reporting supported on the rog chakra on the rog

00:14:51.519 --> 00:14:57.120
shock room core razer deathadder pro and steel series rival 3. so if you don't

00:14:55.839 --> 00:15:02.800
have one of the supported mice then you still get the GeForce experience performance overlay stats just without

00:15:00.959 --> 00:15:06.480
mouse latency added in to get accurate results at home you'll need to make sure

00:15:04.240 --> 00:15:10.079
that you get at least 20 samples of data like 20 clicks and try to find a dark

00:15:08.240 --> 00:15:12.959
background where a change in brightness will show up well muzzle flashes like

00:15:12.079 --> 00:15:19.839
that and anything else that's instant when you click should you nicely and that's

00:15:16.480 --> 00:15:21.920
it back to our regularly scheduled Linus

00:15:19.839 --> 00:15:25.279
i still prefer it as a standalone thing that you could carry forward with you to

00:15:23.360 --> 00:15:30.480
future monitors rather than built into it but i think the value of it from just

00:15:28.160 --> 00:15:35.600
a tuning and optimizing perspective is definitely there yeah i think it would

00:15:32.880 --> 00:15:38.959
actually make a lot more sense if you as the

00:15:36.399 --> 00:15:43.360
center of your ring of friends decide to count your monitor around and

00:15:41.519 --> 00:15:48.480
straight up optimize each one of their machines right for a little profit

00:15:46.560 --> 00:15:51.519
that that might actually or you could just be a good friend

00:15:50.560 --> 00:15:55.279
you know and help that center person buy it so

00:15:53.600 --> 00:15:58.480
they don't have to use it okay okay okay you've got an esports team you've got a

00:15:57.040 --> 00:16:01.519
bunch of these monitors but you only want to spend the extra money on one

00:16:00.240 --> 00:16:06.399
yeah that makes it so you go through and you set up all the different machines and they all have the same settings and

00:16:04.639 --> 00:16:10.720
you're good to go or one person buys this monitor post the results on reddit

00:16:09.199 --> 00:16:14.720
and then everyone just uses those same settings yeah hey there it is all right

00:16:12.959 --> 00:16:18.560
so one of you go ahead and buy it the rest of you just get the one without the

00:16:16.639 --> 00:16:23.440
extra module nonsense speaking of nonsense it's never nonsense

00:16:21.279 --> 00:16:27.600
when i transition to our sponsor storyblocks gives you studio quality

00:16:25.440 --> 00:16:30.639
video clips for a fraction of the cost you can download all the stock video

00:16:29.279 --> 00:16:35.279
your heart desires from their member library including hd and 4k footage

00:16:33.839 --> 00:16:39.519
after effects templates motion backgrounds and more we actually use it

00:16:37.440 --> 00:16:43.600
quite a bit here on techwiki and all the content is royalty free for both

00:16:41.440 --> 00:16:46.560
commercial and personal projects like youtube videos

00:16:44.880 --> 00:16:50.480
new clips are added regularly so there's always something fresh to download and

00:16:48.000 --> 00:16:54.639
you can check out our link below to get story blocks today if you enjoy this

00:16:52.959 --> 00:16:59.920
type of video where we're tweaking and optimizing things maybe check out our

00:16:56.160 --> 00:17:02.399
video on one usmas ctr utility lets you

00:16:59.920 --> 00:17:06.480
take your ryzen processor and just get more performance out of it for free