WEBVTT

00:00:03.000 --> 00:00:11.120
well I wanted to take some time to compare the reference cooler for the GTX

00:00:08.320 --> 00:00:15.679
560 and I realize I'm holding entirely the wrong prop right now but the

00:00:12.719 --> 00:00:21.359
reference cooler for the GTX 560 so this is a flowerstyle heat pipe cooler so you

00:00:19.520 --> 00:00:24.320
can see if I I guess if I spin that you might be able to see a little bit better

00:00:22.880 --> 00:00:29.080
you can see it looks kind of like a stock Intel LGA 1155 heat sink and then

00:00:27.720 --> 00:00:32.920
there's three heat pipes that come out of the top you can you see in there okay

00:00:31.560 --> 00:00:38.600
what if I turn like this can you see the heat pipes no okay how about now can you

00:00:35.680 --> 00:00:38.600
see the heat pipes through the

00:00:38.760 --> 00:00:45.399
top really you can see the heat pipes okay

00:00:43.039 --> 00:00:51.440
one two three heat pipes like that so those are spreading out into a uh like

00:00:48.960 --> 00:00:55.520
fin Arrangement on either side so it's actually kind of like the MSI Cyclone

00:00:53.520 --> 00:00:59.239
cooler in design but it's covered with a shroud now some of the exhaust is going

00:00:57.480 --> 00:01:02.000
to come out here some of it's going to come out here some of it's going to come

00:01:00.719 --> 00:01:05.920
out here and some of it's going to come out here so while it does have a rear

00:01:04.439 --> 00:01:12.640
exhaust vent for all intents and purposes we can consider this a sort of

00:01:08.439 --> 00:01:16.439
70 to 80% internal exhaust card now I

00:01:12.640 --> 00:01:19.759
wanted to compare that to the MSI GTX

00:01:16.439 --> 00:01:22.560
560 twin frozer 2 now this is not that

00:01:19.759 --> 00:01:26.799
card this is a GTX 460 Hawk the twin frozer 2 card is actually on the test

00:01:24.479 --> 00:01:30.960
bench right now preparing for my idle temperature measurement but I'm going to

00:01:28.920 --> 00:01:35.040
compare both temper temp and noise levels of the twin frozer 2 heat sink

00:01:33.439 --> 00:01:38.920
which looks like this now this is an almost entirely internal exhaust heat

00:01:37.200 --> 00:01:43.960
sink as well you can see we're going to get some incidental air flow from this

00:01:41.680 --> 00:01:48.240
fan coming out here but other than that there's a lot ex of exhaust coming right

00:01:46.079 --> 00:01:52.719
through the card here in the gaps right through here there's going to be some

00:01:50.040 --> 00:01:57.280
going out this way uh the heat pipes including uh two super pipes that is

00:01:55.439 --> 00:02:02.960
thicker than standard heat pipes and two normal ones so there are four total heat

00:01:59.000 --> 00:02:05.039
pipes are part of the overall design it

00:02:02.960 --> 00:02:10.440
does use a dense fin arrangement with dual fans in order to achieve what MSI

00:02:07.560 --> 00:02:16.480
claims is up to 20° better cooling under load than a stock cooler

00:02:13.640 --> 00:02:22.160
now it's become a little bit challenging with recent gpus to measure maximum load

00:02:19.840 --> 00:02:27.160
so instead of using furmark which was my traditional favorite um because of the

00:02:25.280 --> 00:02:31.280
way they've locked them down to actually underperform in furmark so that you

00:02:29.120 --> 00:02:35.480
don't blow your GPU by benchmarking furmark all day which I can kind of

00:02:33.280 --> 00:02:40.400
understand what I'm actually going to do is I'm going to Loop the heaven

00:02:37.040 --> 00:02:42.280
Benchmark and then just find the highest

00:02:40.400 --> 00:02:45.959
temperature recorded during the entire thing this will be lower than normal so

00:02:44.319 --> 00:02:51.239
probably the Delta will be lower than we'd normally see but I guess it'll give

00:02:48.159 --> 00:02:52.800
us a more realistic look at what kind of

00:02:51.239 --> 00:02:56.599
noise and what kind of temperatures we're going to get with these cards in a

00:02:54.640 --> 00:03:02.040
gaming scenario so right now I'll show you guys my testing methodology I've got

00:02:59.440 --> 00:03:06.400
the GTX 560 on an Open Bench bear in mind Open Bench is going to be up to 5

00:03:04.159 --> 00:03:10.480
to even 10° cooler than what you're going to see in a closed case either

00:03:08.840 --> 00:03:14.560
that or the fan will ramp up more so it will also be in some ways quieter now

00:03:12.959 --> 00:03:18.760
I've got my ambient temperature here which is about 23 Dees and I have a

00:03:16.959 --> 00:03:25.159
temperature probe right here in front of the fan intake So based on a 23° intake

00:03:21.720 --> 00:03:28.959
temperature and a 27° idle temperature

00:03:25.159 --> 00:03:31.319
on my card I will call that a Delta of

00:03:28.959 --> 00:03:35.720
4° for my GTX 560 so I'm going to do some

00:03:34.080 --> 00:03:39.879
load temps and I'm going to do some testing on the reference design and then

00:03:38.439 --> 00:03:44.040
I'll come back and do a conclusion for you guys that to stress the card as much

00:03:42.799 --> 00:03:50.120
as possible I am running the tessellation on Extreme as well as 8xaa

00:03:47.640 --> 00:03:57.799
and no it's not counterfeit I just changed my Hardware all the time AA uh

00:03:52.640 --> 00:04:00.239
8x8 and 16x AF otherwise looks like this

00:03:57.799 --> 00:04:04.480
now in terms of temperatures uh hold on I'm going to turn my sensor back on so

00:04:02.680 --> 00:04:09.680
the room temps have gone up a little bit because I did have the doors closed so

00:04:07.000 --> 00:04:13.439
we're at 26° in the room now and the maximum and I'm going to take the

00:04:11.439 --> 00:04:19.919
absolute Peak maximum we saw during the heaven Benchmark was

00:04:16.600 --> 00:04:23.919
54° so that was the highest that was

00:04:19.919 --> 00:04:25.720
seen okay so then that means uh 54 - 26

00:04:23.919 --> 00:04:29.960
is I'm not very good at Mental Math so I'll be recording that on a graph and

00:04:28.080 --> 00:04:34.560
then I'll show you guys the reference cooler in a moment okay so just getting

00:04:32.720 --> 00:04:41.039
an idle temperature reading from my reference card so we're going to call

00:04:35.960 --> 00:04:43.720
that 23° ambient and we'll call

00:04:41.039 --> 00:04:50.039
that an in complete inability of my camera to focus on anything well I'll

00:04:47.080 --> 00:04:53.600
tell you what it says then it says 28 so there is the reference card on the bench

00:04:52.039 --> 00:05:00.720
and there is another attempt to focus come on come on come on come on come on yeah you can kind of see that 28 so

00:04:57.199 --> 00:05:03.919
that's a Delta of only about five ° so

00:05:00.720 --> 00:05:05.479
back with the load results so we can see

00:05:03.919 --> 00:05:12.039
here on the graph what our Max temperatures were while running the

00:05:07.680 --> 00:05:16.000
heaven Benchmark we peaked at around 70°

00:05:12.039 --> 00:05:17.840
so that is a Delta of uh actually we

00:05:16.000 --> 00:05:22.400
here I'll do the math

00:05:19.639 --> 00:05:26.639
um okay because our room temperature is now

00:05:23.720 --> 00:05:32.000
23 actually it was 24 a moment ago but I opened the door so we're going to go

00:05:27.800 --> 00:05:33.479
with 24 uh I way here is the data I'm

00:05:32.000 --> 00:05:40.479
not going to bother graphing this for you guys I'm sorry so with the hawk we

00:05:36.400 --> 00:05:42.960
had a 4° Delta from the room temperature

00:05:40.479 --> 00:05:48.639
uh up to the temperature of the GPU at idle so a 5 Dee Delta on the reference

00:05:46.240 --> 00:05:56.160
card but then when we look at the load temperatures it is almost as advertised

00:05:52.080 --> 00:05:59.280
on the MSI box almost a 20° difference

00:05:56.160 --> 00:06:02.319
between the four 560 twin frozer 2 and

00:05:59.280 --> 00:06:06.560
the 560 reference card so you can see

00:06:02.319 --> 00:06:08.199
that MSI does advertise 20° C cooler now

00:06:06.560 --> 00:06:12.720
uh if we were running a more extreme scenario we probably might see 20° as a

00:06:10.840 --> 00:06:18.479
worst case scenario something else to bear in mind is this is an OC Edition

00:06:14.840 --> 00:06:20.919
card so naturally the twin frozer 2 card

00:06:18.479 --> 00:06:24.360
will run a little bit hotter so if we clocked it down to the same speeds as

00:06:22.680 --> 00:06:29.039
the reference card we might actually see even better temperatures so thank you

00:06:26.400 --> 00:06:35.800
for checking out my video of the GeForce GT TX 560 TI thermals and Acoustics as

00:06:33.560 --> 00:06:41.639
well as the comparison between the reference card and the twin frozer 2

00:06:39.240 --> 00:06:47.039
Edition card from MSI don't forget to subscribe to Linus techtips for more

00:06:43.160 --> 00:06:47.039
unboxings reviews and other computer

00:06:48.039 --> 00:06:57.120
videos so this is my noise reading for the gtx560 ti twin frozer 2 going to go

00:06:55.039 --> 00:07:03.240
about 6 in from the fan and let you guys take a little listen and then I'm going

00:06:58.840 --> 00:07:03.240
to cut in the reference

00:07:15.039 --> 00:07:22.879
cooler and here's my load test on the twin frozer 2 so I'm just going to move

00:07:19.560 --> 00:07:22.879
my normal distance from the

00:07:23.319 --> 00:07:28.160
card and I'll cut in with the reference

00:07:28.319 --> 00:07:31.319
cooler