Gadgetory

what's up guys Jays two cents here you might be wondering why I'm wearing this ridiculous wristband is because it is very hot in my studio right now we came in this morning and the a/c wasn't working we had a massive heat wave come through California the other day I saw a max of 126 degrees freedom units which is like 52 see right now in my studio it's cooled off a bit since the weekend but it's still 86 degrees freedom units which actually works out to exactly 30 degrees Celsius so we've changed our video today specifically to see how much water cooling actually helps in a hot environment we've already shown how water cooling in a very controlled environment is very close to air cooling when the air cooler is actually built to a proper standard but what happens now when you live in a hot climate and you don't have a/c we're gonna see exactly what that Delta is now between air and water spoiler alert the water is gonna be cool but just how much with its adjustable pmw 3367 16,000 dpi optical sensor for RGB zones 9 customizable buttons and offering both wired and wireless operation are only a few reasons why the corsair dark or RGB should be on your shortlist for your next mouths purchase check out its full list of features by clicking the link in the description below so here's how this test is going to work I'm actually using a 1080 TI Poseidon from Asus which is actually an air-cooled I take the lid off which is actually an air-cooled and a water-cooled GPU you can just leave it as is and run the air cooler or you can actually just plumb it upto tubing like I have here and the water block is built into it now admittedly the Poseidon is not a hundred percent efficient or best at either air or water cooling but it's good enough to get us a controlled result if we took two different cards one on air and one on water even if they're identical cards the founders Edition cards there is variance between cards Silicon Lottery and all that various mumbo-jumbo and junk so at least here we are using the same core the same everything not being touched and the way we're gonna run this test is I'm going to have the fans on the to recurve and then we're gonna run the like a heaven benchmark and just loop because it's again it's a controlled benchmark we're gonna see what our max temperature is then we're gonna ramp the fans up to a hundred percent and compare what our max temperature is and that tells us what our max cooling capability would be in our current 30 C environment and the cooler maxed out then we're gonna turn the fans off and we're gonna turn on the water cooling pump and see how far it comes down that is the way we're gonna be doing our test so without further ado I've got to fill this loop and then we are going to office lis put up our monitor and all that stuff so I know maybe Phil can just do a quick little Philly montage filling Phil a filling Mont now let's talk about what we already expect to happen we expect it to hit its target temperature which in this case is 84 C that's factory once we hit 84 see the way GPU boost works is it's going the closer it gets to that temperature the more it's gonna start slowing down the core and the way it keeps itself from exceeding that target temperature is it will slow down the core and the bring down the voltage to try and keep it from going above that we should not see above 84 C unless we had this like outdoors in the Sun in a hundred plus degrees then you might see it exceed that and then actually thermal shutdown so what we're gonna be looking here for here so we're gonna be looking for today is how much core speed are we losing with air cooling in this hot ambient environment which we are then going to compare with our water cooling so without further ado we're gonna run heav'n benchmark now I know this is a synthetic benchmark I am well aware of the fact that games are probably a bit more stressful the problem is it's very difficult to run an actual game and get a very consistent result because of the fact that you are free roaming and moving around the world and it's very hard to get consistency between tests alright so here we are in the 4k test in heaven benchmark we are just chugging along in the 30s on fps we just hit 40 because with that 8x msaa and 4k that's a bit ridiculous but we are now working this GPU for every it's got point out though it is running at 2,000 mega Hertz 100 percent load 55c 0% fan speed you can see they're not turning because we are in the factory fan curve which has that up there they go which has that cool and quiet mode where it won't turn on the fans until it exceeds a certain temperature at 58 is where it kicks on and then this will speed up based on the factory curve but where you're gonna notice is that's going to start dropping over time so you see it fluctuating from 2000 now the higher this gets towards 84 the lower this numbers gonna be so what we're looking for here is how much speed do we lose in this 30 C ambient temperature versus water so we're gonna go ahead and just let this roll until we find out where our level work levels out then we're gonna run the fans at 100% level it out and then switch to water I know I explained that twice but you know sometimes you just gotta so about 15 minutes has gone by and we actually have an interesting discovery here so we decided to turn on the power limit indicator so we can see when we're hitting power limit which is why you've seen it bounce on and off because every time we hit power limit it will pop up on here and say power limit because what we realized is we're not hitting 84c we had expected to hit 84c we're not hitting it and yeah I know it's really annoying watching it bounce around our boost clock still came down from 2000 quite a bit but we are still significantly overclocked on the self boost GPU boost 3.0 100% CPU 76% fan 79 C on the temperature so we decide we're gonna go ahead and do the rest is test is we are going to max out the power limit because it's set at its default 100 but these cards especially the Poseidon's are designed for overclocking so we are going to go ahead and bump that to 100% so that we can see now I guess our differential between air and water now obviously we know water is going to be lower than this but and we obviously had to think about this decision because we don't want to manipulate the test in any way but I don't think that that's gonna cause us any problems here so all we're gonna do right here is we're gonna go ahead and take the power limit and the temp limit which are linked these are typically linked together which our new temperature limit now is 9 DC and 120 percent or 20 percent above factory TDP is where we are setting our temperature and our power limit that's what our limitation was before it was temperature even in this hot environment and before it was even like core clock or any of that stuff well obviously core clocks related to temperature because it's a sliding scale we actually just hit thermal limit or we actually just hit power limit before thermo Steven became a problem so what we're gonna see now is this will probably go up to 85 ish but now we have to let it run for a little while longer here and see what happens in fact it might not even go any higher because we're surprised at just how beefy this air cooler is and how good of a job it's doing in this hot environment with the factory curve we didn't touch the fan curve at all okay we're not getting any hotter we're pretty much locked at a DC 1949 megahertz ever since we raised the power limit this has not changed this one up a couple of degrees in fact it it bounces between 80 and 79 now doesn't go any higher fans seems to capped out about 79 percent so what we're gonna do now is we are just going to bump this fan to a hundred I don't expect the results to be that much different simply because of the fact that aunt Clinton because of the let me do it because we're only getting about an extra twenty one percent fan speed so I don't think it's gonna make a major difference so we're just gonna let it go for a couple of minutes we did go up one degree fahrenheit in here though so it's gonna be like thirty and a half C now something like that we're gonna keep an eye on the temperature as best we can as we as I said the AC in here is not working so I can't not control the environment as much so as it heats up throughout the day it's gonna get hotter in here that's how we're trying to get this test done quickly so we've settled at 72 C it's not changing we only came up one step on our boost clock you guys remember it moves in steps of 13 so we went from 1949 to 1962 fans are still at 100% this is actually pretty good considering the fact that we were almost 31 C inside of here inside this room that's a I expected it to be much harder than that obviously I thought with a 1080i 250 watt TV P more than that one overclocked obviously I thought it would be higher but I wanna I want to show you guys something here so obviously the system is filled with water it's not running the pump is not running the fans as you can see right here have not been running and I'm not gonna turn these fans on right now but I want to show you something here if I flip on this pump and just with the fact that the block and the heat sink are somewhat connected the amount of passive cooling in a loop like as long as there is water flow is pretty insane so even with those fans on and these fans off watch what happens to the temperature the moment I click on this pump look how fast this will drop so the pump is on it is now moving look at that with that temp right there I don't even turn on the radiator yet folks so we're gonna do now is we're gonna go in here we are going to zero out the fan keep our test running with the fans off the radiator off and just the pump going and passive cooling just passive cooling through this radiator in this 31c environment it's already down to 52 C and holding back up to 2,000 mega Hertz now it's pretty obvious once I plug in the radiator and get it flowing air that that's going to come down significantly we'll test how far that is and we'll talk about why that is and this is kind of my rebuttal to some of the folks that were upset that I said that air cooling of water cooling are closely matched in my last video obviously there's limitations to that and that's exactly what we're talking about today so stick around for the conclusion now that my fan is running and there's a lot of heat in that loop holy crap there's a lot of heat coming through that radiator already that's how much it had just soaked just touching the system okay so we're pretty much heat soaked we're seeing 50 C we were seeing 51 it came down to 50 and there's 51 again but we came all the way back up to 2012 megahertz that is the max core clock that this graphics card will actually allow based on its P States and its boost clock and all that 0% fan speed you know Phil was right so I put it back to Ottawa to go off because it's below that that speed thing so again feel smarter than me on that one but we are 30 C about 31 C lower than where we were on the fans on the factory curve an approximately 21 C lower than where we were with the air cooling on at a hundred percent now we are using a 240 millimeter radiator and there was a lot of heat coming out of here the the fluid temperature on this is definitely soaked so we are fortunate that we are in a big enough space to where this is not affecting the overall temperature of the space if you were in a small 8 foot by 8 foot bedroom or something this would indeed get hotter over time you might even see that get all the way up to 60 C if you're in a small but the reason why you're seeing the results come down like this even though we are in such a hot environment versus the previous test where we had a controlled 72 or 71 degrees Fahrenheit in this space is because of the efficiency and the efficacy of how water is able to pull air out or pull temperature out of already heat saturated air versus the efficiency of air cooling now air cooling is obviously gonna get the job done you saw this cooler had no problem getting it done in fact the water cooling loop here might put more heat into your room making you less comfortable because of the fact that you are more efficiently pulling the heat directly from the die to this heat exchanger to the environment aka your gaming space but I just want to kind of show here with this really up this hot room as our AC broke to kind of make lemonade out of Satan's ball sweat basically to show you guys where water cooling really does indeed benefit you and we're going to do now is we're going to set the fans to 100% with the radiators on and the water cooling going to see if this number comes down below 50 I don't really anticipate that it will if it does it might only go one or two degrees but that's it guys thanks for watching today's video hit this like button if you appreciate the fact that instead of walking in and just going home which I legitimately consider doing and texting Phil and me like hey paid day off it's hot I bet you wish I did that huh Phil yeah yeah just a little bit anyway guys thanks for watching if you want to know more about the site in cooler I do have a review on this and we compare it to the Founders card and all that stuff I'll put a link in the description below you guys can go and check out that review on this cooler and as always guys we will see you in the next one hey 48c so obviously when they work in conjunction there is benefits to be had even in this temperature spoiler alert though we actually tested that when we did the card review and I already knew that I forgot something downstairs woman Barry Beck I can't go any lower no you're not supposed to tilt the camera if you want to learn more about the