WEBVTT

00:00:00.080 --> 00:00:05.680
do you cringe because every time you fire up a heavy game you're greeted with

00:00:03.439 --> 00:00:09.920
the line of your blower fan spinning up to max blowiematron on your graphics

00:00:08.000 --> 00:00:15.200
card disgusting well one solution is water

00:00:12.719 --> 00:00:18.560
cooling but that's expensive and time-consuming and not everybody is

00:00:16.880 --> 00:00:23.359
comfortable filling their electronics up with water let's face it we can't all be

00:00:21.039 --> 00:00:30.560
complete mad lads oh who's the idiot now fortunately

00:00:26.560 --> 00:00:35.520
ragin tech has the solution for you this

00:00:30.560 --> 00:00:38.239
is the morpheus 8057 a huge GPU air

00:00:35.520 --> 00:00:44.160
cooler that apparently fits both NVIDIA and AMD graphics cards and we are going

00:00:41.360 --> 00:00:48.640
to give this puppy a shot to see if it's a sizable upgrade

00:00:45.920 --> 00:00:53.600
over your stock cooler and also if it could be a simpler more cost effective

00:00:50.800 --> 00:00:58.079
solution than water cooling you know what else is cost effective

00:00:56.079 --> 00:01:01.600
telling you about our sponsor glassware with glasswire you can instantly see

00:00:59.760 --> 00:01:06.400
your current and past network activity detect malware and block badly behaving

00:01:04.000 --> 00:01:10.159
apps on your pc or Android device use offer code Linus and get 25 off

00:01:08.400 --> 00:01:13.159
glassware at the link in the video description

00:01:19.680 --> 00:01:26.960
the basic concept here of strapping a giant cooler to your graphics card is

00:01:23.680 --> 00:01:29.759
nothing new last year we put a tower CPU

00:01:26.960 --> 00:01:32.960
cooler onto an rtx 2080 ti with great results

00:01:30.799 --> 00:01:36.720
that is as long as you don't want to install any expansion cards and

00:01:35.200 --> 00:01:42.720
purpose-built products for this have existed for over a decade

00:01:38.960 --> 00:01:45.040
so what's so special about this then

00:01:42.720 --> 00:01:48.720
when it comes to performance honestly speaking

00:01:46.079 --> 00:01:53.520
probably not much in fact region tech appears to have reused photographs of

00:01:51.119 --> 00:01:58.240
one of their previous GPU coolers on the 8057 product page but that's not

00:01:55.680 --> 00:02:03.040
necessarily a bad thing compared to the relatively i don't know

00:02:00.560 --> 00:02:06.719
whatever the relatively small cooling elements that are included on many

00:02:04.880 --> 00:02:11.200
graphics cards i'm expecting a reduction in noise and

00:02:09.280 --> 00:02:13.200
an improvement in performance out of this thing

00:02:12.400 --> 00:02:17.680
so really then the new model number comes

00:02:15.440 --> 00:02:22.800
down to compatibility rather than being stuck fabricating your

00:02:20.000 --> 00:02:27.920
own mounting hardware the morpheus 8057 includes everything that you will need

00:02:24.640 --> 00:02:31.360
to mount to either a reference pcb card

00:02:27.920 --> 00:02:33.040
from NVIDIA or from AMD

00:02:31.360 --> 00:02:37.120
certainly looks pretty nice highlights of the product include the 12 heat pipes

00:02:35.200 --> 00:02:42.959
so you can see these six over on this side wrap around right

00:02:40.720 --> 00:02:46.640
there and they terminate right here and then these six over here carry over this

00:02:45.360 --> 00:02:51.599
way so really each half of this is kind of cooled by

00:02:50.720 --> 00:02:57.519
half of these heatsink fins it's got a nice black finish on it

00:02:55.440 --> 00:03:01.519
definitely a fan of that and it's rated for a total of

00:02:58.680 --> 00:03:06.000
360 watts of heat dissipation of course it can have wonderful specs and be very

00:03:04.000 --> 00:03:10.480
beautiful but none of that matters if the performance sucks so let's go ahead

00:03:08.400 --> 00:03:14.480
and investigate that step one is to install the appropriate bracket the

00:03:12.400 --> 00:03:18.480
arrows need to point towards these curved heat pipes in the back step

00:03:16.319 --> 00:03:22.800
number two after you've determined where the holes line up you're gonna go ahead

00:03:20.319 --> 00:03:26.000
and install these standoffs into the bottom of the cooler one of my little

00:03:24.560 --> 00:03:30.720
tricks is that i actually like to do a dry wipe of chips before i do a wet wipe

00:03:28.640 --> 00:03:35.280
with alcohol it can help pull off the majority of the thermal goop before you

00:03:32.560 --> 00:03:42.000
start just spreading it around holy bananas this is a lot of thermal

00:03:38.560 --> 00:03:44.799
interface junk to okay

00:03:42.000 --> 00:03:48.720
wow to raging tax credit they've done you the courtesy of putting little tabby

00:03:46.799 --> 00:03:51.519
tabs on one side of some of these so it's a little bit easier to get the

00:03:50.080 --> 00:03:56.319
things off but this is going to be pretty tedious ladies and gentlemen here

00:03:53.360 --> 00:03:59.599
we go good luck everybody i have my doubts about this one of the

00:03:58.400 --> 00:04:04.480
other little tricks that i've learned over the years is to

00:04:02.239 --> 00:04:07.680
basically never ever trust a double-sided thermal tape if you were

00:04:06.480 --> 00:04:11.280
actually going to deploy this for long-term use i would strongly recommend

00:04:09.680 --> 00:04:15.920
investing the 13 or whatever it is in a high quality

00:04:13.920 --> 00:04:19.919
thermal epoxy like the one made by tech ingredients

00:04:16.959 --> 00:04:19.919
another youtuber

00:04:21.280 --> 00:04:26.880
really okay

00:04:24.400 --> 00:04:32.160
all right let's do the d where the heck does d go this is not a

00:04:29.680 --> 00:04:35.280
great diagram i believe d goes on the two large chips directly to the left of

00:04:34.000 --> 00:04:39.440
the man you've got to be kidding me that's a memory heatsink doesn't that

00:04:37.440 --> 00:04:42.720
just drive you crazy like as an engineering background at this point do

00:04:41.280 --> 00:04:47.199
you think this is harder or easier than water cooling oh

00:04:45.680 --> 00:04:51.759
you're gonna make me answer that are you uh i think that this is

00:04:49.520 --> 00:04:55.919
similar in difficulty compared to installing a GPU water block

00:04:54.479 --> 00:05:01.199
but you don't have to deal with the added complexity of all the other

00:04:59.040 --> 00:05:04.560
aspects of custom water cooling it's a good question though a valuable one for

00:05:03.120 --> 00:05:07.680
our viewers to have the answer to hard-hitting tech journalism right here

00:05:06.160 --> 00:05:11.520
folks here's a hard-hitting tech journalism question

00:05:09.199 --> 00:05:15.360
copper is a lot heavier than aluminum and we're using the same crappy adhesive

00:05:13.840 --> 00:05:19.440
that i think is barely holding on the aluminum how long will this stay here

00:05:17.840 --> 00:05:23.840
you know what's really mind-blowing to me about this is that this isn't the

00:05:21.440 --> 00:05:27.039
only heat sink that doesn't fit at all the copper one doesn't go all the way to

00:05:25.680 --> 00:05:31.919
the end of this bank of modules and this aluminum

00:05:29.919 --> 00:05:35.600
one is completely different from that other one it's double the width and the

00:05:33.600 --> 00:05:39.759
fins run the opposite way and it's like twice as long as it needs to be all

00:05:37.600 --> 00:05:44.320
three of these are for memory chips and these two are actually on two chips each

00:05:42.560 --> 00:05:48.960
the same number of chips like what am i looking at here

00:05:45.840 --> 00:05:50.880
this is the definition of a mix uh oh

00:05:48.960 --> 00:05:57.120
shoot which one's uh see this diagram is so

00:05:53.840 --> 00:05:59.759
shockingly bad two long bars uh go on

00:05:57.120 --> 00:06:03.840
like not only the tiny strip of the arms but also the bigger ones right next to

00:06:01.280 --> 00:06:09.520
them this just gets more random every time i put something on it

00:06:07.520 --> 00:06:13.840
i can't okay you got three more huge things to go we're we're getting there

00:06:11.120 --> 00:06:17.440
no i can't this is so stupid oh there's so many things that should have thermal

00:06:15.520 --> 00:06:22.639
pads according to nin video there okay well according to ragin tech

00:06:20.880 --> 00:06:25.919
we ready boys we're going to use their included

00:06:23.919 --> 00:06:30.080
thermal goop oh yeah it smells like generic

00:06:28.240 --> 00:06:33.840
let's just get that on there that should be oh that's uh that's

00:06:32.080 --> 00:06:36.880
really a lot it won't affect performance but i just don't feel like

00:06:35.360 --> 00:06:39.840
cleaning up the mess uh huh

00:06:40.960 --> 00:06:47.520
yeah so these are all gonna have to come off

00:06:45.039 --> 00:06:51.360
i just knocked off another heatsink okay yeah we're calling that good

00:06:49.520 --> 00:06:56.479
there goes that heatsink so we're going to be using our rog machine as our test

00:06:54.160 --> 00:07:02.479
bench it's got a 9900 ks and an rg maximus 11 motherboard with 64 gigs of

00:06:59.039 --> 00:07:04.880
3200 megahertz cl 14 and the RGB memory

00:07:02.479 --> 00:07:08.160
and well used to have an rog strix 2080 ti but we went with the founders edition

00:07:06.479 --> 00:07:13.599
because we wanted a pretty good chance of compatibility our GPU was pre-tested

00:07:10.479 --> 00:07:15.919
with the power limit slider crank to 112

00:07:13.599 --> 00:07:21.280
but without any clock speed adjustments and we'll be pairing our morpheus with

00:07:18.319 --> 00:07:26.080
these Noctua nf a12fs to see how it compares

00:07:23.199 --> 00:07:31.280
now if pci expansion was a priority we might choose something lower profile and

00:07:27.759 --> 00:07:33.759
conveniently the morpheus does include

00:07:31.280 --> 00:07:38.639
13 uh here we go 25 millimeter it does include brackets for low profile fans

00:07:36.319 --> 00:07:42.880
but here's the thing unless you're into high-speed networking or pci express

00:07:41.120 --> 00:07:46.639
storage the odds are that you don't have a ton of add-in

00:07:44.960 --> 00:07:49.919
cards so we're going with these because i think

00:07:48.080 --> 00:07:53.840
it's going to be a great combination of maximum performance

00:07:51.520 --> 00:07:57.120
and silence as long as you don't mind a little bit of

00:07:55.039 --> 00:08:01.840
GPU sag okay looks like these clip on to the

00:07:59.440 --> 00:08:05.919
bottom here nope it looks like they clip on to the top

00:08:03.360 --> 00:08:10.479
and then you just gotta drag them wait what that's it

00:08:08.560 --> 00:08:15.280
oh my goodness probably wouldn't be a terrible terrible idea to put the uh

00:08:13.199 --> 00:08:18.960
the i o bracket back on it does give you a little less leeway for

00:08:17.520 --> 00:08:24.080
jiggering when you're putting your cards in that is some next level saggage there

00:08:21.599 --> 00:08:27.599
that's like poor pc yeah

00:08:25.199 --> 00:08:32.560
okay we just lost another heat sink oh no we lost two more okay

00:08:30.400 --> 00:08:36.719
oh we close now okay well what like do we just fire it up or what

00:08:34.640 --> 00:08:40.320
just send it while you do that i'm just going to collect all the heat sinks that

00:08:38.159 --> 00:08:44.399
have fallen off so far is this an appropriate look how they massacred my

00:08:42.080 --> 00:08:48.959
boy moment now we're all set up we're logging

00:08:46.560 --> 00:08:53.360
temperature data from hardware info we've got afterburner cranked to 112

00:08:51.600 --> 00:08:58.000
power limit and we're going to run a GPU stress test in fur mark here we're

00:08:55.040 --> 00:09:02.160
running at 2560 by 1080 with four times msaa we're trying to really toast this

00:09:00.160 --> 00:09:05.519
graphics card no i don't want this warning

00:09:03.440 --> 00:09:10.000
give it to me those fans are barely running i can tell you right now that

00:09:06.959 --> 00:09:11.360
GPU is basically silent plugged into the

00:09:10.000 --> 00:09:14.640
motherboard headers that's steep but hold on give it time

00:09:13.200 --> 00:09:20.720
give it time 55. oh and with the panel closed

00:09:17.600 --> 00:09:20.720
closed let's get the panel closed

00:09:20.959 --> 00:09:29.040
does that just go on like that it sure does okay so we hit uh

00:09:25.279 --> 00:09:30.959
96 fpf average FPS average at 78 degrees

00:09:29.040 --> 00:09:36.440
celsius was our kind of that was our equilibrium point yes is

00:09:32.959 --> 00:09:36.440
there any coil one

00:09:38.800 --> 00:09:45.040
curiously not that i can detect which is funny to me because i went and i water

00:09:42.800 --> 00:09:49.279
blocked an rtx titan a little while back and it was just nasty in terms of coil

00:09:47.360 --> 00:09:54.000
one this looks like it's settled in now we're still running at 96 FPS if we go

00:09:51.760 --> 00:09:58.000
ahead and pop open hardware info we can see that we are at

00:09:55.839 --> 00:10:02.240
1920 so we've not lost any of our maximum turbo here

00:09:59.920 --> 00:10:05.120
and we have 1700 on the stock cooler that i remember

00:10:04.160 --> 00:10:10.959
correctly you got you went to seven dropped to 1700. i think so yeah oh well fantastic

00:10:09.519 --> 00:10:14.160
and that is a substantially better result because we're running at 72

00:10:12.560 --> 00:10:19.120
degrees celsius max turbo and

00:10:16.640 --> 00:10:23.200
these fans are like

00:10:20.399 --> 00:10:27.920
barely spinning you hear that those are at like 600 700 RPM

00:10:28.000 --> 00:10:36.720
if the question was does this outperform the stock cooler then the answer is yes

00:10:33.600 --> 00:10:38.800
but if the question is should i run out

00:10:36.720 --> 00:10:42.720
after having spent extra on a founder's edition card

00:10:40.320 --> 00:10:47.040
anyway and go buy this aftermarket accessory for it that has quite frankly

00:10:45.600 --> 00:10:53.360
not just an unintuitive but kind of a scary installation process

00:10:51.920 --> 00:10:57.600
and then just terrible quality when it comes to

00:10:55.200 --> 00:11:01.360
the thermal pad staying on the answer gets

00:10:59.120 --> 00:11:07.200
a little bit more complicated i think really the takeaway

00:11:04.800 --> 00:11:10.320
here is when we compare the results we were going to put the ragin tech on this

00:11:08.720 --> 00:11:13.600
strix card but now that we've seen these results i i

00:11:11.839 --> 00:11:17.440
don't think it makes any sense when we compare the results to just buying a

00:11:15.519 --> 00:11:21.920
card with a big fat beefy cooler on it in the first place

00:11:20.079 --> 00:11:29.079
the answer is pretty obvious especially because this doesn't cost an extra 70 80

00:11:25.040 --> 00:11:29.079
over a founder's edition

00:11:29.680 --> 00:11:36.399
if it's not obvious the answer is no and if it's not also obvious i gotta tell

00:11:34.320 --> 00:11:41.120
you about my sponsor right about now what do you get when you mix drop.com

00:11:38.320 --> 00:11:46.079
and thx you get the drop thx panda wireless headphones they feature

00:11:43.040 --> 00:11:47.200
audiophile quality with no wires they

00:11:46.079 --> 00:11:50.560
claim they're the world's most distortion-free wireless headphones and

00:11:49.120 --> 00:11:54.959
quite frankly from listening to them myself i have not heard anything that

00:11:53.200 --> 00:11:59.279
even comes close they're a closed back headphone they feature thx's aaa

00:11:57.279 --> 00:12:03.519
amplifier technology the sound quality is the same whether you run them with

00:12:00.880 --> 00:12:06.800
bluetooth or with a wire and you get 30 hours of wireless life pretty freaking

00:12:05.360 --> 00:12:10.000
awesome you can even get a detachable gaming mic for them so check them out

00:12:08.480 --> 00:12:13.680
today at the link in the video description if you guys enjoyed this

00:12:11.760 --> 00:12:18.560
video you might actually enjoy one of mine and colin's previous crazy GPU

00:12:16.240 --> 00:12:22.560
cooling experiments where we put a CPU tower cooler on a graphics card i think

00:12:20.560 --> 00:12:26.800
i would go as far as to say that one was more sensible significantly more

00:12:24.800 --> 00:12:31.839
sensitive i mean aside from just a power drill we didn't really need any crazy

00:12:28.880 --> 00:12:35.839
tools to mount the thing and i trusted it way more than this quite honestly

00:12:37.519 --> 00:12:40.240
that's true