Gadgetory

what is going on everyone Jase $0.02 here when I did the video about my 6700 K build last week a lot of folks pointing out the fact that there's a front-mounted radiator on there and you can't do that because it's going to heat up everything in your system and it's terrible and the world is going to end as far as we know it and it's just bad for all of your parts and I got to thinking I've been putting front mounted radiators on computers for years and years and not once has any of those terrible doomsday type of predictions ever happened so I thought maybe today we would just kind of go ahead and do for the sake of science a video here about comparing temperatures with the radiator mounted to the front and seeing if allowing warmer air to enter the front of the computer really makes a huge overall difference the mastercase 5 and mastercase 5 pro from Coolermaster combines modularity with creativity giving you the freedom to build it your way make it yours by clicking the link down in the description okay so hardware wise it is an intel 6700 K it is overclocked we'll probably be doing somewhere around 4.7 for this test at something like one point three volts somewhere around there we want to pump some heat we don't want to we want to really kind of go for some numbers here in this test we just run stock speeds and I don't think anything would really be affected which is actually a good thing but we want to see in kind of a worst-case scenario whether or not the claims of front mounted radiators are going to cause problems temperature-wise for your parts so it's a 6700 K overclocked corsair h 100 i we do have in here also a reference titan x of course they're all reference but it is a blower style cooler so it's going to be getting all of its air from inside the case the case cooling is not going to have any effect on the temperatures of the GPU which is why i'm using a blower style cooler all of its air it's getting is going to be atmospheric air inside the case which means if the front mounted radiator really is warming up the case then we can go ahead and see that by using a blower style cooler because that's the only place that's getting air if we use the atmospheric cooler where it's circulating the air into the case and then the case has to exhaust it then the case itself is going to play more of a factor here than where things are placed so that's why we're going with the blower cooler on the tight next we're going to tight necks because it is the biggest most powerful hottest GPU core right now that nvidia has on a single GPU yeah I mean yeah there's a Titan Z that has two GPUs but that again is a half exhaust half recirculating cooler which would not have worked very well for this test okay so just to do some baseline numbers here I have a laser temperature get probe as well as an air temperature meter so right now the ambient temperature here in the room that's going to restart this is sitting at 24 Celsius or 20 or 75 point well now 75.2 fahrenheit is what the air temperature is in this room so now before we start the test I'm going to go ahead and set this on top of the exhaust fan here I'm going to let it measure the temperature of the air exiting the system here for about 60 seconds or so and then that's our baseline that's where we are right now before running any tests and running any sort of heat through the system and letting things kind of balance out and then we'll run it through load get our numbers make sure the ambient temperature in the room is the same I've got the air running to keep it pretty consistent then we'll do it again and compare and they'll do the same thing with the GPU okay so the air coming out of the computer is measuring seventy seven point zero Fahrenheit and in see for those of you using Celsius that is twenty four point nine that's what it's showing there seventy six point eight because it comes down pretty quickly here so that's what we've got here is a baseline so let's go ahead and start our a 264 let this thing heat up the system and let's see how hot it gets to transition I've had the test going now for over ten minutes I stopped it for a second there and forgot to hit record on the camera anyway ambient temperature in the room is still seventy five point two and I have had the thermometer up here measuring the temperature exiting the case with this thing going a hundred percent and as you can see we actually did get a little bit of throttling here on the CPU we due to temperature we actually hit a hundred C on the first core several times so that's just to show you here that we are pushing the same for as much heat as we possibly can I don't recommend actually doing that and forcing your CPU to go all the way to throttling temps but I wanted to exacerbate that's exacerbate not the other word on the kind of exacerbate the situation as best I can to get these tests you know as concrete as possible so anyway we're going to remove this now and the air leaving the case is at seventy eight point six degrees Fahrenheit or a twenty six C so as you can see it's the I mean we can pretty much already call this conclusive if we wanted to where the air that is exiting the case is hardly heated up by the air entering the case at least with the H 100 I if I go ahead and set this thing on the front so that it's measuring the air that's coming down and it does if you guys are curious about whether or not the s 340 can actually pull air in very well or if it's too obstructed you can see this little blade right here will actually move due to the air flow and tell you how much air is actually going in there now I believe you guys will actually be able to see this it's double check here so if I set this thing on the front of the case right here check this out see that it is pulling air in just fine so we'll let that measure the intake temperature right now intake temperature going into the front of the case let that go for just a second here okay that takes time to actually measure and balance it out all right seventy seven point nine degrees Fahrenheit is what's entering the case and what is exiting the case seventy eight point six so do you see how the front intake radiator at least on a setup like this is not really doing a whole lot in terms of heating up the system there's that test let's go ahead and do this now let's go ahead and take in fact I'm going to skip taking the side panel off for the CPU test there's really no point in fact I'm go ahead and stop this now and I want to doing too much more throttling on my CPU see the temperatures came down really quickly back down to the 30s on the core let's let things cool off and then we're going to go ahead and run a GPU test on this and then we'll see if the GPU temperatures rise at all if we do it right after running a stressful CPU test I really don't think that's going to be the case just based on the numbers that we've already seen now I'm not using an opinion here I'm using a thermometer to do the test so I don't see how you know this could be an opinion thing this is just flat out what does the thermometer say anyway let's go from there okay so what I've done right now is I've loaded up GPU tests for Valley benchmark and we have overclocked the GPU one hundred megahertz left at stock voltage and I have also raised the thermal threshold for the GPU to nine to one degrees Celsius in MSI Afterburner and we're going to let it the test run and run and run until things have balanced out I think we're pretty much already there based on my previous test where we're hitting 84 degrees Celsius on the GPU so that's below our maximum temperature in GPU Boost 2.0 which means that the core right now is running a hundred percent it's not being thermal throttled at all and then we are going to heat up the H 100 I as best we can by running a 264 and then switching to Valley and seeing if we got any difference in temperatures and then if we did we'll take the side panel off and compare if we didn't and the GPU didn't throttle at all then we can go ahead and call this test complete I guess okay so I've had Valley going for I don't know how long now at least 15 minutes temperatures are very stabilized ambient temp is 78 Fahrenheit on the dot and that's 25 point 6 C so the temperature in the room went up a couple of degrees simply because of the graphics card that's important to note because remember when we did this with the CPU test temperatures didn't really rise at all temperatures exiting the case right now remember the CPU is under small load when you're doing gaming gaming does not put a huge load on the CPU so I don't expect this to be a whole lot different here especially since we have a blower style card I already explained why I went with that seventy nine point seven Fahrenheit case temperature and that's going to be twenty six point five C so as you can see only a few degrees like three degrees raised on the case temperature at all now what our maximum GPU temp get to our max temperature whoops our max temperature was uh it looks like we've got a max temperature of about eighty-five eighty-six we hit 86 once all right so the next thing we're going to do now let's go ahead and load up that CPU again let's make that nice and toasty and then it switched to a2 valley and see if we get any higher temperatures I don't think we're going to quite honestly I think the GPU is so hot it's going to create its own atmospheric issue where we're going to see more issues happening with the GPU than we are with anything CPU radiator wise causing any problems for anything in the system but if we open up Corsair link and look at the temperatures I mean shoot we've got 23 degrees Celsius on the on the hard drive 35 degrees Celsius on the motherboard 35 C on the other spot on the motherboard that it measures I mean 29 C on another motherboard temperature location I think I think we're starting to get our results here guys so Valley has been looping and looping and looping and looks like we're sitting at a maximum temperature once again of 86 degrees Fahrenheit or Celsius on the graphics card my bed and case temperature is sitting at 86 point one degrees Celsius or Fahrenheit Dan I can't get those numbers straight okay 86 Fahrenheit in the case for Celsius that's about 30 degrees Celsius so one thing to keep in mind here is that the ambient temperature in this room has gone up considerably because the graphics card is putting off a lot of heat so a lot of that raising temperature is actually attributed to the graph card and the amount of heat it's putting off more so than the fact that you the CPU and the front mounted radiator was affecting anything in fact if we look at the curve here for the CPU everything stays pretty consistent everything for the graphics card temperature-wise as you can see right here stays very very consistent on temperature in fact we even dip down a little bit and came back up slightly so there's a lot of things that could be factoring temperatures not just the location of your radiator one last thing I want to do before we talk about conclusions is I want to try and run both at the same time and see how much it really does affect things now I'm at that actually take my overclock off of the graphics card because I tried doing this once already and I ended up getting instability on the graphics card as a CPU is at a hundred percent load and so believe it or not your CPU stability can affect what appears to be your graphics card overclock but could also be your CPU causing problems as well ambient temperature in the room right now eighty four point five degrees Fahrenheit twenty nine point one C so you can see the temperature in the room is going up and equally the system temperature is going up as well it's kind of funny how that works huh anyone who understands temperature in physics and thermodynamics will know exactly why that makes sense okay so both valley benchmark and aida64 have been running for quite a while now you can see the cpu temp in the case there is at ninety C on the socket exhaust temperature on the top of the case is at eighty three point four degrees Fahrenheit it actually came down a little bit for Celsius that's twenty-eight point five now if you want to see something contrast actually we'll we'll take a quick reading of the rear by the graphics card and that'll give you some insight as to why the temperature in the room came up as I started the graphics card tests okay so I've been sitting here with my hand pretty much being toasted alive on the back of this Titan X remember this is not overclocked right now this is factory temperature graphics cards currently running at eighty-three Celsius we came down three degrees Celsius from the overclock and the rear temperature the air exiting the back of the graphics card one hundred and thirty six point four Fahrenheit 56 actually it's coming down obviously it's away from it so let's say 55 degrees Celsius is the temperature of the air coming off the back ooh that's warm so let's go ahead and talk about some final thoughts in all of this now one reason I didn't bother taking the side panel off whatsoever was the fact that our tests that we did first pretty much concluded that even though the CPU was running full speed even throttling a little bit that we were only getting a couple of degrees difference in temperature between the intake temp and the exhaust temp so to me that meant that the H 100 I was not really heating up the air enough to make any sort of difference I think we saw what three degrees Fahrenheit difference in temperatures it wasn't a lot now what you saw here was the fact the GPU alone makes more difference to the entire rooms temperature which is going to affect the entire system temperature than anything else in your system now I'm not saying that front mounted radiators won't affect your temperatures I can guarantee you right now that if I took my skunkworks system here and I put the GPUs on a radiator that's front mounted we would see a system pressure system temperature increase on all the components across the board and the reason for that is it comes down to watts displaced by the radiator there becomes a point where you have more radiator than is necessary so it only gets to a certain operating temperature and that's it but what happens is when you have an under sized radiator and too much wattage being pushed through that radiator then you end up having heat soak where it gets hotter and hotter and hotter now something else I also did to try and make this test as absolutely as worst-case scenario as I could was I ran the fan profiles on the H 100 I at balanced because I didn't want it to overcompensate by running a ton of air through it which would obviously cool it off and would make the case cooling much more of a factor than how much the H 100 I is heating up the temperature inside the case if that makes any sense to you guys so running it balanced and only run about 1,300 and maybe 1,400 rpm fluctuating back and forth and maximum RPMs of those static pressure fans now don't misquote me here because I am NOT saying front mounted radiators will not affect the overall system temperatures they we'll and as you saw here it affected us only a couple of degrees but if we had more heat going through that radiator let's say we had the graphics cards or the CPU and everything in a single loop then yeah the entire system temperature would rise but then again if the parts that are mostly being cooled by the radiator are part of that loop then the only things that are going to make any sort of difference would be like motherboard temperatures and MOSFETs and things like that which can be affected by those but unfortunately there's no way to really measure those in this test because it's an all-in-one cooling loop that's something that would be a much more in-depth test later on but there's physics involved here and physics cannot be twisted and changed by opinion and what I want to tell you now is if that you have a case like the s340 or other cases that really only offer front-mounted options for all the one water cooling loops or even custom loops don't be concerned about it causing any sort of temperature damage to your system it's not going to happen I've seen all kinds of misconceptions out there and flat-out wrong information being regurgitated and thrown up all over the community of people saying front-mounted radiators will cause your system to overheat and it's just a flat-out lie there's more to it than being cut and dry like that and in your most most situations it's not going to matter so building your case put the radiator in the front if that's where you want to pull it put it and stop worrying about it play some games have some fun watch some more of my videos if you want but I'm going to get out of here because I am heading to PDX land tomorrow morning today is Wednesday so this is the same day video hitting the PDX land I'll be hanging out with Jerry maybe we'll do another Q&A video or something those are always fun not doing another drunk Tech Talk don't ask don't even bother to waste your time anyway guys we'll see you there I'll be doing a huge live stream like all weekend long for PDX so make sure you guys are there to check that out it's going to be on my twitch channel my twitch name is Jay's VFX J ays VFX not Jays two cents anyway guys I'm gonna get out of here we will see you in the live stream follow on Twitter if you want to know when it's going live at JSU sense enough shameful plugs you know makes me happy when you guys follow me and say J we love you because I love you guys too anyway time to go we'll see you in Portland well Seattle first then Portland you know I got to pick up Jerry first someone's got to control that guy my god that guy that guy