1 00:00:00,160 --> 00:00:01,960 So you're building a brand new PC 2 00:00:01,960 --> 00:00:05,220 and it's time to choose a cooler for your CPU. 3 00:00:05,220 --> 00:00:07,720 There are so many options out there, 4 00:00:07,720 --> 00:00:09,940 but before you even really get into the weeds, 5 00:00:09,940 --> 00:00:11,960 you've gotta make one choice. 6 00:00:11,960 --> 00:00:15,520 Do you cool your PC with a large air cooler 7 00:00:15,520 --> 00:00:18,220 or with an all-in-one liquid cooler? 8 00:00:18,220 --> 00:00:20,140 Now, if we can trust the marketing, 9 00:00:20,140 --> 00:00:22,400 AIOs are the clear choice. 10 00:00:22,400 --> 00:00:25,200 Water transfers heat so much better than air, right? 11 00:00:25,200 --> 00:00:27,620 But is that the entire story? 12 00:00:27,620 --> 00:00:29,080 Maybe, actually. 13 00:00:29,080 --> 00:00:29,920 I don't know. 14 00:00:29,920 --> 00:00:31,820 Alex has been working on this in isolation 15 00:00:31,820 --> 00:00:34,520 and I don't know the test results. 16 00:00:34,520 --> 00:00:47,060 So I'm just as eager to find out as you guys are. 17 00:00:47,060 --> 00:00:49,460 Now, before you run out and buy a new CPU cooler, 18 00:00:49,460 --> 00:00:51,720 you need to first figure out if your CPU 19 00:00:51,720 --> 00:00:54,480 is gonna get any benefit from additional cooling. 20 00:00:54,480 --> 00:00:55,740 Now, in the old days, 21 00:00:55,740 --> 00:00:57,480 inadequate cooling was a recipe 22 00:00:57,480 --> 00:00:59,780 for instant catastrophic failure 23 00:00:59,780 --> 00:01:03,180 because the modern protection or throttling mechanisms 24 00:01:03,180 --> 00:01:04,820 that prevent them from exceeding 25 00:01:04,820 --> 00:01:07,440 the generally accepted maximum, 26 00:01:07,440 --> 00:01:09,680 of a hundred degrees Celsius, 27 00:01:09,680 --> 00:01:11,700 didn't exist yet. 28 00:01:11,700 --> 00:01:15,780 But just because both Intel and AMD seem to agree 29 00:01:15,780 --> 00:01:17,640 that their processors will last 30 00:01:17,640 --> 00:01:21,480 at least through their warranty period at that temperature, 31 00:01:21,480 --> 00:01:24,080 doesn't mean having it there is a great idea. 32 00:01:24,080 --> 00:01:26,280 The hotter an integrated circuit runs, 33 00:01:26,280 --> 00:01:28,000 the more power it will consume 34 00:01:28,000 --> 00:01:29,800 and the shorter its lifespan will be. 35 00:01:29,800 --> 00:01:32,980 And furthermore, when your CPU does approach its limit, 36 00:01:32,980 --> 00:01:36,320 you can experience greatly reduced performance. 37 00:01:36,320 --> 00:01:37,320 That throttling mechanism. 38 00:01:37,320 --> 00:01:39,000 That I alluded to before. 39 00:01:39,000 --> 00:01:43,500 So if your only goal is not sacrificing performance, 40 00:01:43,500 --> 00:01:46,640 then temperatures that don't exceed 90 degrees, 41 00:01:46,640 --> 00:01:48,700 it's honestly a good spot to start 42 00:01:48,700 --> 00:01:52,000 with 65 to 75 under full load 43 00:01:52,000 --> 00:01:54,960 being the conventional wisdom target, 44 00:01:54,960 --> 00:01:55,960 as well as contributing 45 00:01:55,960 --> 00:01:58,280 to a quieter computing experience overall. 46 00:01:58,280 --> 00:02:01,340 So finally getting to our PC then, 47 00:02:01,340 --> 00:02:04,120 in here, we've got an RTX 2070 48 00:02:04,120 --> 00:02:06,620 and an Intel Core i7-9700K. 49 00:02:06,620 --> 00:02:08,940 But since any decent cooler 50 00:02:08,940 --> 00:02:13,520 can keep the stock 95 watt 9700K at bay, 51 00:02:13,520 --> 00:02:15,700 provided you don't live in the Sahara, of course, 52 00:02:15,700 --> 00:02:18,420 we cranked it up to a blistering five gigahertz 53 00:02:18,420 --> 00:02:22,060 on all eight cores at a voltage of 1.33 volts, 54 00:02:22,060 --> 00:02:24,860 bringing it up to 160 watts. 55 00:02:24,860 --> 00:02:30,080 Now, we definitely need something a bit more robust. 56 00:02:30,080 --> 00:02:31,040 Or do we? 57 00:02:31,040 --> 00:02:32,880 This overclock was actually achieved 58 00:02:32,880 --> 00:02:33,620 using just a Noctua NFU12S, a mid-range air cooler, 59 00:02:33,620 --> 00:02:34,460 coming from Intel. 60 00:02:34,460 --> 00:02:35,540 And that's what we're going to do today. 61 00:02:35,540 --> 00:02:36,380 We're going to test it out. 62 00:02:36,380 --> 00:02:36,720 We're going to test it out. 63 00:02:36,720 --> 00:02:39,020 We have a 10 gigahertz controller coming in at just $60. 64 00:02:39,020 --> 00:02:40,520 And with just one fan, 65 00:02:40,520 --> 00:02:43,220 it was able to keep us at a max of 84 degrees. 66 00:02:43,220 --> 00:02:45,740 Now, that is a little toasty. 67 00:02:45,740 --> 00:02:47,240 And with our fan at full blast, 68 00:02:47,240 --> 00:02:48,740 it was a little loud, 69 00:02:48,740 --> 00:02:52,120 at least by Noctua standards, at 45 decibels. 70 00:02:52,120 --> 00:02:56,120 So well, now it's time to go water cooling then, right? 71 00:02:57,650 --> 00:03:02,270 Maybe, but not until we've adjusted our test setup. 72 00:03:02,270 --> 00:03:06,110 The thing is, 43 decibels is clearly audible 73 00:03:06,110 --> 00:03:09,350 and bordering on annoying to have right next to your head. 74 00:03:09,350 --> 00:03:14,070 But for context, the average office is 50 decibels. 75 00:03:14,070 --> 00:03:17,330 So because noise is just as important 76 00:03:17,330 --> 00:03:19,190 as cooling performance, 77 00:03:19,190 --> 00:03:20,870 we had to create an environment 78 00:03:20,870 --> 00:03:23,630 where we could properly evaluate both. 79 00:03:23,630 --> 00:03:26,150 Okay, so why don't you show me what you've done here? 80 00:03:26,150 --> 00:03:28,650 Yes, welcome to my room of quietness. 81 00:03:28,650 --> 00:03:31,080 This used to be an office, you know. 82 00:03:31,080 --> 00:03:35,260 Yes, but now it's uncomfortably silent. 83 00:03:35,260 --> 00:03:37,440 So we know this is kind of ghetto, 84 00:03:37,440 --> 00:03:40,340 but it works via the same principles 85 00:03:40,340 --> 00:03:43,780 as our recent server room sound deadening video. 86 00:03:43,780 --> 00:03:47,500 Basically, if you take a bunch of loosely packed 87 00:03:48,700 --> 00:03:52,240 permeable mass and you put it all around 88 00:03:52,240 --> 00:03:55,580 where you're trying to block the sound, 89 00:03:55,580 --> 00:03:57,180 you don't get any standing waves. 90 00:03:57,180 --> 00:03:59,360 It ends up absorbing a lot of the sound 91 00:03:59,360 --> 00:04:00,360 that's trying to come in 92 00:04:00,360 --> 00:04:03,460 and preventing any reflections and bouncing. 93 00:04:03,460 --> 00:04:05,630 So basically, 94 00:04:05,630 --> 00:04:07,890 because we can easily control the HVAC, 95 00:04:07,890 --> 00:04:10,350 this room has its own dedicated thermostat. 96 00:04:10,350 --> 00:04:12,470 And because we've got all these beanbags set up, 97 00:04:12,470 --> 00:04:14,940 we're able to achieve, what was it? 98 00:04:14,940 --> 00:04:16,220 26 decibels. 99 00:04:16,220 --> 00:04:18,480 A 26 decibel noise floor in here. 100 00:04:18,480 --> 00:04:22,140 That means that we can comfortably measure devices 101 00:04:22,140 --> 00:04:26,600 like this computer that are not much above 30. 102 00:04:26,600 --> 00:04:28,700 Now this computer has been modified 103 00:04:28,700 --> 00:04:30,820 from our original configuration a little bit. 104 00:04:30,820 --> 00:04:33,820 So it's running four of Noctua's NF-A12X25FLX fans. 105 00:04:33,820 --> 00:04:35,000 So it's running four of Noctua's NF-A12X25FLX fans. 106 00:04:35,000 --> 00:04:35,220 So it's running four of Noctua's NF-A12X25FLX fans. 107 00:04:35,220 --> 00:04:35,660 So it's running four of Noctua's NF-A12X25FLX fans. 108 00:04:35,660 --> 00:04:37,320 With their ultra low noise adapter 109 00:04:37,320 --> 00:04:38,480 coming off the power supply. 110 00:04:38,480 --> 00:04:40,840 So these are at just 900 RPM. 111 00:04:40,840 --> 00:04:44,700 That power supply is a prime fanless titanium 112 00:04:44,700 --> 00:04:45,900 from Seasonic unit. 113 00:04:45,900 --> 00:04:47,780 So it's not making any noise. 114 00:04:47,780 --> 00:04:50,660 The graphics card, there's not much we can do. 115 00:04:50,660 --> 00:04:53,120 Your system's gonna have a graphics card in it, but- 116 00:04:53,120 --> 00:04:54,440 It does have a silent mode. 117 00:04:54,440 --> 00:04:55,760 It does have, oh, it does. 118 00:04:55,760 --> 00:04:56,600 Oh, the fans are off. 119 00:04:56,600 --> 00:04:58,600 It has a switch on it. 120 00:04:58,600 --> 00:04:59,440 Perfect. 121 00:04:59,440 --> 00:05:01,540 Thank you, ROG Strix. 122 00:05:01,540 --> 00:05:05,180 So what this means then is that our only variable 123 00:05:05,180 --> 00:05:06,020 is our CPU cooler. 124 00:05:06,020 --> 00:05:09,180 Now, some of you have probably noticed 125 00:05:09,180 --> 00:05:11,900 that the spot where we mounted the radiators is in the top 126 00:05:11,900 --> 00:05:13,680 because we've got case fans in the front here. 127 00:05:13,680 --> 00:05:15,120 And that was for a reason. 128 00:05:15,120 --> 00:05:17,260 So on the one hand, yes, 129 00:05:17,260 --> 00:05:19,400 drawing in fresh air for your radiator 130 00:05:19,400 --> 00:05:21,340 is going to give it an advantage 131 00:05:21,340 --> 00:05:23,780 versus drawing in warmed up air 132 00:05:23,780 --> 00:05:25,600 and exhausting it out the top of the system. 133 00:05:25,600 --> 00:05:28,840 However, what we found was that the disadvantage 134 00:05:28,840 --> 00:05:30,720 from an acoustic standpoint 135 00:05:30,720 --> 00:05:33,240 of having it right there in the front of the case 136 00:05:33,240 --> 00:05:35,820 was much greater than the advantage 137 00:05:35,820 --> 00:05:39,020 it gained from having access to fresh airflow, 138 00:05:39,020 --> 00:05:40,600 especially because you guys got to consider, 139 00:05:40,600 --> 00:05:43,940 this is not exactly a little tiny mid-tower chassis. 140 00:05:43,940 --> 00:05:45,880 We've got three intake fans. 141 00:05:45,880 --> 00:05:48,220 So the ambient temperature inside the case 142 00:05:48,220 --> 00:05:49,980 is really not that much warmer. 143 00:05:49,980 --> 00:05:51,780 So to keep things as real world as possible, 144 00:05:51,780 --> 00:05:53,800 we're using the blender classroom test. 145 00:05:53,800 --> 00:05:55,900 So that takes about 17 minutes 146 00:05:55,900 --> 00:05:59,620 and that's enough time to heat up most AIOs. 147 00:05:59,620 --> 00:06:01,880 And we were having issues where the whole room 148 00:06:01,880 --> 00:06:04,680 was heating up two to three degrees. 149 00:06:04,680 --> 00:06:05,700 And given the fact that we had to do that, 150 00:06:05,700 --> 00:06:07,040 given that's more than the variation 151 00:06:07,040 --> 00:06:08,200 between all of our coolers, 152 00:06:08,200 --> 00:06:11,840 we had to have HVAC on and the door open 153 00:06:11,840 --> 00:06:15,190 and then run in at about 16 minutes, 154 00:06:15,190 --> 00:06:19,910 turn the HVAC off, measure the acoustics and boom. 155 00:06:19,910 --> 00:06:21,670 So with that out of the way, 156 00:06:21,670 --> 00:06:23,990 we're ready for the results. 157 00:06:23,990 --> 00:06:25,350 That we are. 158 00:06:25,350 --> 00:06:26,470 Are you excited? 159 00:06:26,470 --> 00:06:28,610 I actually am. 160 00:06:28,610 --> 00:06:31,490 Which one of these do you think is the best? 161 00:06:31,490 --> 00:06:32,750 Cooling, purely cooling. 162 00:06:32,750 --> 00:06:35,950 Am I supposed to play devil's advocate or am I? 163 00:06:35,950 --> 00:06:36,810 Actually guess. 164 00:06:36,810 --> 00:06:39,700 Actually guess. 165 00:06:39,700 --> 00:06:42,120 I would guess that it's gonna be 166 00:06:42,120 --> 00:06:44,420 the triple 120 millimeter red 167 00:06:44,420 --> 00:06:49,200 from a sheer mass and surface area perspective. 168 00:06:49,200 --> 00:06:50,840 Not even in second. 169 00:06:50,840 --> 00:06:52,260 This guy here. 170 00:06:52,260 --> 00:06:53,100 Really? 171 00:06:53,100 --> 00:06:54,100 Yup. 172 00:06:54,100 --> 00:06:55,100 Which one do you think? 173 00:06:55,100 --> 00:06:55,940 So hold on. 174 00:06:55,940 --> 00:07:00,590 We're talking raw performance at a given noise level. 175 00:07:00,590 --> 00:07:01,970 Just raw performance. 176 00:07:01,970 --> 00:07:03,570 Straight up raw performance. 177 00:07:03,570 --> 00:07:04,410 Yup. 178 00:07:04,410 --> 00:07:05,390 This beats this. 179 00:07:05,390 --> 00:07:06,230 Yup. 180 00:07:07,390 --> 00:07:10,260 78 degrees, 81. 181 00:07:10,260 --> 00:07:11,540 Wow, that's not even that close. 182 00:07:11,540 --> 00:07:12,380 No. 183 00:07:12,380 --> 00:07:13,220 Okay. 184 00:07:13,220 --> 00:07:15,820 And guess what sound level they were both at then? 185 00:07:18,890 --> 00:07:22,590 Okay, this one right here was at 36. 186 00:07:22,590 --> 00:07:23,870 Okay, 36 decibels? 187 00:07:23,870 --> 00:07:25,010 Yeah. 188 00:07:25,010 --> 00:07:26,450 Okay. 189 00:07:26,450 --> 00:07:28,550 I would say this with only two fans 190 00:07:28,550 --> 00:07:30,730 is gonna have to be a little louder, 38. 191 00:07:30,730 --> 00:07:31,870 34. 192 00:07:31,870 --> 00:07:32,910 Really? It's quieter too. 193 00:07:32,910 --> 00:07:33,750 Yeah. 194 00:07:35,780 --> 00:07:39,200 And the worst part is that this right here. 195 00:07:39,200 --> 00:07:40,580 Yeah. 196 00:07:40,620 --> 00:07:43,880 It performs the 360. 197 00:07:43,880 --> 00:07:45,540 And isn't the 360 the one that comes 198 00:07:45,540 --> 00:07:47,220 with their new redesigned cold plate? 199 00:07:47,220 --> 00:07:48,140 Yeah. 200 00:07:48,140 --> 00:07:50,420 So the Ace Attack one, 201 00:07:50,420 --> 00:07:52,380 it's only a two degree difference, 202 00:07:52,380 --> 00:07:56,900 but also like 79, 81. 203 00:07:56,900 --> 00:07:57,880 So. 204 00:07:57,880 --> 00:07:59,950 So it's quieter as well. 205 00:07:59,950 --> 00:08:00,790 Yeah. 206 00:08:00,790 --> 00:08:03,730 And this right here, garbage, 86 degrees. 207 00:08:03,730 --> 00:08:06,410 This guy fared pretty well. 208 00:08:06,410 --> 00:08:07,250 So. 209 00:08:07,250 --> 00:08:09,190 Okay, well now let's talk price to performance then. 210 00:08:09,190 --> 00:08:10,450 So we know this is our least expensive offering. 211 00:08:10,450 --> 00:08:12,970 This was like our mid-tier air cooler. 212 00:08:12,970 --> 00:08:14,130 Yeah. 213 00:08:14,130 --> 00:08:16,150 This is probably more like 80 bucks, right? 214 00:08:16,150 --> 00:08:17,070 70 bucks. 215 00:08:17,070 --> 00:08:19,990 The one that I really don't understand 216 00:08:19,990 --> 00:08:22,290 is how this performs so well. 217 00:08:22,290 --> 00:08:23,830 I guess this is really a testament 218 00:08:23,830 --> 00:08:26,340 to how well heat pipes perform then. 219 00:08:26,340 --> 00:08:30,210 And also just Noctua, like they just kill it. 220 00:08:30,210 --> 00:08:31,710 They know their shiz. 221 00:08:31,710 --> 00:08:34,270 But if you think about it, like yes, 222 00:08:34,270 --> 00:08:37,330 water transfers heat very quickly, 223 00:08:37,330 --> 00:08:40,170 but heat pipes are also fluid heat transfer. 224 00:08:40,170 --> 00:08:43,410 So this is actually phase change cooling on a certain level. 225 00:08:43,410 --> 00:08:45,170 So then one point of clarification though, 226 00:08:45,170 --> 00:08:48,130 do you think this is unique to Corsair's AIOs? 227 00:08:48,130 --> 00:08:49,790 Like, are they just bad? 228 00:08:49,790 --> 00:08:51,970 No, I think they're pretty good. 229 00:08:51,970 --> 00:08:54,290 They're just like the largest company for AIOs 230 00:08:54,290 --> 00:08:57,170 and they all use basically the same design. 231 00:08:57,170 --> 00:08:59,430 So it's the most representative. 232 00:08:59,430 --> 00:09:00,270 Right. 233 00:09:00,270 --> 00:09:01,870 Cause that's what people are probably gonna buy. 234 00:09:01,870 --> 00:09:04,250 I'm gonna get you to take these two away since they're done. 235 00:09:04,250 --> 00:09:07,010 So things don't look great for our water coolers, 236 00:09:07,010 --> 00:09:09,910 but there is more to the story than performance. 237 00:09:09,910 --> 00:09:14,300 And performance, noise levels and price to performance. 238 00:09:14,300 --> 00:09:16,300 Let's run through some more pros and cons. 239 00:09:16,300 --> 00:09:17,860 So one big problem, 240 00:09:17,860 --> 00:09:21,300 an additional problem for water coolers is actually the pump. 241 00:09:21,300 --> 00:09:25,840 Not only does that add additional noise and cost, 242 00:09:25,840 --> 00:09:29,020 it also adds a point of failure and a nasty one too. 243 00:09:29,020 --> 00:09:32,500 Since not only are pumps more likely to fail than a fan, 244 00:09:32,500 --> 00:09:36,180 but when they do, it's more catastrophic for your cooling 245 00:09:36,180 --> 00:09:37,600 and it's harder to detect. 246 00:09:37,600 --> 00:09:39,740 Now, if your CPU fan dies, 247 00:09:39,980 --> 00:09:43,780 then you can probably see that it isn't spinning 248 00:09:43,780 --> 00:09:47,280 and then slap a case fan on there until a new one arrives. 249 00:09:47,280 --> 00:09:48,520 With a pump, 250 00:09:48,520 --> 00:09:51,020 you better hope that you have your software monitoring 251 00:09:51,020 --> 00:09:53,420 enabled and then when it does bite the bullet, 252 00:09:53,420 --> 00:09:56,820 you pretty much just need to pony up for an entire new unit. 253 00:09:56,820 --> 00:10:00,300 An unattractive prospect since AIOs, like I said, 254 00:10:00,300 --> 00:10:03,600 extra components tend to be more expensive than heat sinks 255 00:10:03,600 --> 00:10:05,480 with similar performance. 256 00:10:05,480 --> 00:10:10,740 But that doesn't mean that AIOs don't have 257 00:10:10,740 --> 00:10:12,180 some advantages. 258 00:10:12,180 --> 00:10:14,080 Because it's full of water, 259 00:10:14,080 --> 00:10:18,040 an AIO has greater thermal mass than a heat sink. 260 00:10:18,040 --> 00:10:21,060 That means that even though after a long load, 261 00:10:21,060 --> 00:10:24,280 its performance might not be as impressive, 262 00:10:24,280 --> 00:10:28,440 during short bursts, like while your CPU is turboing up, 263 00:10:28,440 --> 00:10:32,280 that extra heat can be absorbed then dissipated later 264 00:10:32,280 --> 00:10:35,940 without such stark swings in temperatures. 265 00:10:35,940 --> 00:10:38,480 Also, if you regularly move your PC, 266 00:10:38,480 --> 00:10:40,580 like to LAN parties in search, 267 00:10:40,580 --> 00:10:44,920 transportation is arguably much safer with an AIO. 268 00:10:44,920 --> 00:10:47,220 I mean, we are talking for this guy, 269 00:10:47,220 --> 00:10:52,220 12 points of contact versus just two for this. 270 00:10:52,920 --> 00:10:56,400 Big bulky air coolers have a lot of weight 271 00:10:56,400 --> 00:10:59,000 hanging directly off of your CPU socket, 272 00:10:59,000 --> 00:11:02,340 which in some extreme cases can actually lead 273 00:11:02,340 --> 00:11:05,180 to motherboards bending or even breaking. 274 00:11:05,180 --> 00:11:08,280 Now, this isn't as much of an issue with brands 275 00:11:08,280 --> 00:11:09,920 that have good mounting hardware, 276 00:11:09,920 --> 00:11:10,840 which is one of the reasons 277 00:11:10,840 --> 00:11:13,080 that we so strongly recommend Noctua. 278 00:11:13,080 --> 00:11:15,680 But even with Noctua mounting, 279 00:11:15,680 --> 00:11:18,960 if I needed to entrust my PC to an airline 280 00:11:18,960 --> 00:11:23,240 or a parcel carrier, I would equip it with an AIO. 281 00:11:23,240 --> 00:11:26,400 Another good point for AIOs is the coolness factor, 282 00:11:26,400 --> 00:11:28,420 not coolness for your CPU, 283 00:11:28,420 --> 00:11:30,100 or even for your motherboards VRM, 284 00:11:30,100 --> 00:11:32,880 which can benefit from incidental airflow 285 00:11:32,880 --> 00:11:35,320 that bleeds away from the heat sink. 286 00:11:35,320 --> 00:11:37,520 But I mean, coolness for showing off your PC. 287 00:11:37,520 --> 00:11:38,400 Now, naturally this comes down to personal preference, 288 00:11:38,400 --> 00:11:39,360 but I mean, coolness for showing off your PC. 289 00:11:39,360 --> 00:11:40,580 But I mean, coolness for showing off your PC. 290 00:11:40,580 --> 00:11:43,740 Some people really like the look of big air coolers. 291 00:11:43,740 --> 00:11:44,760 Yeah. 292 00:11:44,760 --> 00:11:46,140 Thank you, Alex. 293 00:11:46,140 --> 00:11:47,380 But most of the time, 294 00:11:47,380 --> 00:11:50,520 it's easier to get a cleaner overall look 295 00:11:50,520 --> 00:11:52,940 to your build with water. 296 00:11:52,940 --> 00:11:54,940 Now, one thing we haven't mentioned yet 297 00:11:54,940 --> 00:11:56,400 is custom liquid cooling. 298 00:11:56,400 --> 00:11:59,280 With a top of the line block from EK or Alpha Cool, 299 00:11:59,280 --> 00:12:02,380 a huge radiator and a D5 pump, 300 00:12:02,380 --> 00:12:05,220 you can get measurably better performance 301 00:12:05,220 --> 00:12:09,080 than what is possible even with the beefiest of air coolers. 302 00:12:09,080 --> 00:12:11,260 The thing is, at that point, 303 00:12:11,260 --> 00:12:13,480 it isn't really a fair comparison anymore, 304 00:12:13,480 --> 00:12:15,320 since a custom water cooling setup 305 00:12:15,320 --> 00:12:17,560 is only for very advanced users, 306 00:12:17,560 --> 00:12:19,240 requires regular maintenance, 307 00:12:19,240 --> 00:12:20,900 and is incredibly expensive. 308 00:12:20,900 --> 00:12:23,260 Like just the CPU block alone 309 00:12:23,260 --> 00:12:25,860 can cost more than our NHU 12A. 310 00:12:25,860 --> 00:12:28,360 That's not to say custom water cooling isn't awesome 311 00:12:28,360 --> 00:12:30,980 and that we don't love it for ourselves. 312 00:12:30,980 --> 00:12:32,220 It's just not the kind of thing 313 00:12:32,220 --> 00:12:35,300 that we're comfortable recommending to people who, 314 00:12:35,300 --> 00:12:35,560 well, enjoy using their computer as opposed to, you know, 315 00:12:35,560 --> 00:12:39,220 well, enjoy using their computer as opposed to, you know, 316 00:12:39,240 --> 00:12:41,140 tinkering with their computer. 317 00:12:41,140 --> 00:12:42,200 So thanks for watching, guys. 318 00:12:42,200 --> 00:12:44,220 If you just liked this video, you can hit that button. 319 00:12:44,220 --> 00:12:46,740 But if you liked it, hit like, get subscribed, 320 00:12:46,740 --> 00:12:47,860 and maybe consider checking out 321 00:12:47,860 --> 00:12:49,300 where to buy the stuff we featured 322 00:12:49,300 --> 00:12:51,400 at the link in the video description. 323 00:12:51,400 --> 00:12:52,860 Also down there is our merch store, 324 00:12:52,860 --> 00:12:56,570 which has cool shirts like this one. 325 00:12:56,570 --> 00:12:58,440 Bottles, there's some. 326 00:12:58,440 --> 00:13:02,860 And our community forum, which you should totally join. 327 00:13:02,860 --> 00:13:05,320 Nope. 328 00:13:05,320 --> 00:13:08,320 You know what's funny is I remember back when 329 00:13:08,320 --> 00:13:11,620 these 120 millimeter Asetek units first launched, 330 00:13:11,620 --> 00:13:13,800 like 10 years ago or whatever it was, 331 00:13:13,800 --> 00:13:16,900 the initial reviews for the Asetek LCLC, 332 00:13:16,900 --> 00:13:20,900 go find reviews for it, overwhelmingly negative. 333 00:13:20,900 --> 00:13:24,140 But then when the Corsair H50 turned up, 334 00:13:24,140 --> 00:13:26,720 which was literally exactly the same product 335 00:13:26,720 --> 00:13:30,640 with a different fan on it, the reviews get positive. 336 00:13:30,640 --> 00:13:34,240 And I'm not accusing, I know the people at Corsair very well. 337 00:13:34,240 --> 00:13:37,370 I know for a fact they didn't pay off anybody. 338 00:13:37,370 --> 00:13:41,870 But like, it's just funny how a brand can, 339 00:13:41,870 --> 00:13:43,650 like the brand name and brand loyalty 340 00:13:43,650 --> 00:13:45,450 might subliminally influence people. 341 00:13:45,450 --> 00:13:46,850 Like with the Asetek ones, people would be like, 342 00:13:46,850 --> 00:13:49,090 yeah, it's not great. 343 00:13:49,090 --> 00:13:50,590 And then with Corsair, they'd be like, 344 00:13:50,590 --> 00:13:52,530 yeah, it's like, it didn't perform as good, 345 00:13:52,530 --> 00:13:54,130 but like, here's all these other benefits. 346 00:13:54,130 --> 00:13:57,670 I don't know, it's a funny thing. 347 00:13:57,670 --> 00:13:58,510 Ha! 348 00:13:58,510 --> 00:13:59,330 Ha!