lexii miller vertical placement would probably decrease the gpu temps by ~1-2C for the open air gpu because it pulls the exhausting air from the gpu away more.
This is what I also think. The blower style cooled GPU gets hotter and radiates more heat and since the CPU and GPU coolers are drawing their air from the intakes it makes a still air pocket below the CPU cooler allowing the heat radiate from GPU to CPU and MB (so, there's none to little airflow between GPU and CPU cooler).
That doesn't really makes sense. GPU vrams etc. should run as hot generating same amount of heat in both tests. The reason why you see higher temps in reference cooled cards is because those stock coolers and heat sinks simply struggle with transferring the heat from the chipset. But all this released energy in form of heat still has to go somewhere, ergo if you have a better cooler capable of transferring more heat from the card, the gpu itself will have lower temps but the relocated heat should mean higher temps on other pc components that are getting blasted by all that circulating hot air. Only thing I can think off is difference in voltage and core/mem clocks which means more emancipated heat but if anything FE should produce less energy than non-ref cards with additional pins and higher boost clocks. All of this leads me to believe that it's a margin of error or some unknown law of aerodynamics or thermodynamics that I simply don't know :/
Suggestion - Rerun the tests with the strange results, you might well find they change. Did you have enough (or too much) thermal interface material? Did you have just a poor seating of the heatsink on the CPU? Yes, I'm sure you know how to mount a cooler, I'm not doubting that, but mistakes happen, I know I've made them. Of course you might have already done this and gotten the same result, but it is the first thought that came to my mind, was to rerun the test with everything remounted fresh.
it's pretty interesting to see you commenting on an airflow video but when someone mentioned you didn't have any exhaust on your son's pc (your rear fan was oriented to be intake) during a live upgrade stream a few weeks ago you said "why does it matter? worry about your own pc" to the guy. rude af.
1080p? A gtx 1070 would do it for most (but not all) games. Gamersnexus just did a great video with a heavily overclocked gtx 1080 and 7700k and found that it wasn't enough to get 120 fps in most games at 1440p though. Really need to get into SLI at that point.
Not even my gtx 1080/i7 4790K 4.7ghz combo gives me 144 fps ultra 1080p in somewhat up to date AAA games. If you really want to have constant 144fps to profit off your 144hz monitor, and you want to play on ultra/very high, you'll need at least a gtx 1080, as I said not even that is enough.
Think this has to do with the direction the air is circulating in the chassis.. This is something you may have to test using a glass side panel case and some smoke to see where the air flow is going..
Jesse Dodge he mean like a fog machine. Of course it would be bad to use actual smoke on a computer. There are fog machine to test the airflow in your PC
There are so many myths people claim when it comes to computers and this is one of them, I knew it would not make much difference, I got flamed in one of my builds for doing the same thing, so I rotated the heatsink for a 3 degree difference in temps. Good video Kyle.
Have mine with a upwards flow as well. First for the simple reason to geht the hot air away from my GPU so it doesn't reduce it's boost. MWith my new card, I'm just to lazy to reassamble all that stuff. (Heavy cooler and a delidded chip: nothing to be done in 5 mins) Besides that I have 2x 140mm in the top and 1x 120mm in the back, so clearly the case supports the vertical airflow. And there would be not nearly as much fresh air coming from the front with all my drive bays, compared to the case he used.
Same. Some foo went after me for mounting my dark rock pro 3 with a horizontal airflow pattern. Good luck fitting that vertically, let alone the mounting hardware doesn't even allow for it.
*Theories!* *Test #1* - Blower was stealing all the cool air. Could test with a PC-safe "smoke machine". *Test #2" - The Zotac card seems to have a more restrictive "Open" design forcing the bulk of the exhaust to the back and bottom allowing for less heat to be ejected into the case compared to an ASUS Strix for example. Also it may not have been stealing all the fresh air allowing the 212 to move more air across the fins. Could test with a more open-air designed card. *Test #3* - "Direct Contact Heat-Pipes" may have lined up with the hot-spot on the IHS better where the die is mounted vertically. Could test using non-direct style cooler with each test. (Would take alot of time though...) *Test #4* - The Zotac's backplate may get hotter then the "Founder's" card due to added power deliver or design changes and the backplate may be heating up the air at the 212's intake. Could test by measuring the air temp above both backplates. (may be a bit difficult to do though...) I know there's that one person out there that's gonna have a rage-attack, but remember, they're just theories that were thought up immediately after watching the video. ^_^ (*Edit*) Went to the -Thought Chamber- restroom and thought of something else. I wonder if an anomaly caused utilizations to differ between tests. Were the frame-rates consistent and the GPU/CPU loads reported equally? Maybe a background Windows update or other process was causing CPU load to vary.
Absolutely, I thought about the air being hotter from the backplate but completely ignored the fact that the open-air GPU is still exhausting hot air. That and the broadside of the CPU cooler in test #4 is also making it much easier for the cooler to intake that exhaust while at the same time "tripling" the surface area that the warm air is passing though. It will be quite interesting if a follow-up is done for this video, I realize it would take a good amount of time to do but your input makes me want to know the answers even more. Heck, this type of "Guess what caused it" video conclusion could even be it's own series of videos. Even if only once a month or so I find it quite interesting and enjoy the participation side of it.
WRONG. It would only be meaningless if all the results were identical. They are not, so obviously blower and orientation of the CPU cooler have an impact.
I notced same thing when change my GPU from "blower style" to "open". But final results for my setup: Less fans is requied (2 removed from top) to make same temp. at CPU now. Blower style cards "stealing" air from case, makes CPU cooling less effective.
I noticed that as well, this is why I added a fan underneath the blower cooler of my GPU. The blower was sucking the air from the front intake fans (and was basically a wall separating top and bottom) and was radiating heat from the backplate up toward the CPU. Adding that bottom mounted fan didn't prevent the GPU from stealing the front intake airflow, but it cooled the card, which helped the CPU stay a bit cooler. Pretty interesting how that all works out.
it could be that the horizontal CPU & open GPU are actually avoiding each other... in a sense. The GPU is pulling air straight up and the air is likely blooming away and above the the CPU intake pressure drop. Causing cooler air to collect for the CPU. Where vertical is in the direct path of hot GPU air... Great idea for a video.
The little tests that you and Jay do are really are service to the community and I appreciate all the time and effort that you put into your videos! Thanks once again for the info!
I believe its as simple as just good airflow, because as soon as you turn the cpu cooler towards the gpu and away from the front fans temps rose up, and since in-closed gpus are mainly always more hotter than open gpus the front fans werent really helping the in-closed gpu dissipate its radiating heat , but the open gpu even though its was dissipating warm air in the case the airflow was pushing it out
possibly the blower style since it's worse as dispersing heat a good portion of the hot air surrounding the card stuck around, while the open air one was better at getting hot air out and away.
@5:45 Just maybe this will explain scenario 2. CPU cooler blows Left, GPU blows upwards, with case flans blowing left and up. Heat is dissipated more evenly (maybe more efficiently) between the top and left of the case. This differs from option 4. However, this is very similar to option 3. CPU cooler blows upwards, while GPU blows Left. Heat going both directions in options 2 and 3. This is contrasted to option 1 ( Majority of air is dissipated left) and option 4 (majority of air is dissipated upwards) which results in higher CPU temps. An analogy for scenario 2 would be having 2 straws in a beer: If you had to drink the whole beer using straws, then it would be able to get the same amount of liquid out of the can using two straws compared to pushing that same amount of liquid through mostly 1 straw.
This is also based on the concept of better cooling with a greater proportion of fresh air vs recycled air (air already cooling one component to cool another component).
Hot air rises, a vertical config with open gpu pulls air directly into the case. Also, I noticed if the cpu, gpu airflow direction matches, temps are lower. Open gpu has vertical airflow, closed has horizontal.
Maybe since the open shroud GPU has multiple fans resulting in relatively more air circulation and creating a higher positive pressure unlike the blower style GPU... correct me if I am wrong
horizontal mounting uses more fresh air from the front intake, vertical mounting the cooler is sucking the hot air from the gpu in = higher cpu temps with open style gpu cooler...nothing more...thus horizontal mounting the cpu cooler is better if you have an open gpu cooler...
Walt Findley .... if the case was fully sealed then yes a internal fan inside sealed cabinet will not suck air from outside .... but in a pc cabinet intenal fan can create a positive pressure "i think so" just like if you put two same fans on top of each other the total air moved will be double the air moved by single fan.
I think the horizontal position cpu cooler lowered the temp because the air flow is in a more consistent direction from to the intake fans airflow. Thus creating a more positive air pressure effect.
Ohhh let me take a stab. The blower gpu is taking air from within the case and pushing the air laterally across the gpu and out the back, so it never hits the cpu cooler. You'd think that's good for avoiding hot air circulating in the case and hitting the cpu cooler, but maybe that's bad for total airflow in the case overall. The less air circulating in the case, the less usable air there is for the cpu cooler. The open air gpu has air all around it, which in turn creates more airflow within the case with the same air. More airflow near the cpu cooler provides more air to cool the cpu. Just a theory.
LOL no the case would have to be operating at a negative pressure vacuum in order for that to be the cause. There is no side panel which makes this whole thing pointless anyway. Its likely the cause was either the difference in mounting (remember he's is unmounting and remounting everytime) position, pressure or some other aerodynamic factor.
One thing that everyone underestimates is interaction between heatspreader and heatsink. It's better to have the heatpipes passing perpendicularly on the "die" of the Intel CPUs. In fact in the best conditions (for the CPU) with blower GPU, the temperature is lower, but open GPU has negatively influenced the results. Anyway horizontal+open GPU gives a strange result.
My thoughts on this are that I have a different case, different fans, different graphics card, different cpu cooler, so I should do my own damn tests if I want relevant results.
I think you changed air flow characteristics by using the lower style fan. The blower style fan would effect the pressure of the case. The blower style has a larger negative pressure zone an then pulls more fresh air away from the CPU cooler? You could check this by using a smoke machine.
I was really tired building my PC and accidentally installed the Cooler facing up and didn't fully realize until I put it in the case. Glad to see it doesn't make too much difference and need to waste the paste to fix it.
I'd actually expect there to be a slight (perhaps not as much as 5 degrees C) difference based on where the exhaust fans are. Having both of them might mean they're pulling the airflow path slightly differently. If that's the situation, then the rear mounted fan may be expelling more GPU heat without it going through the vertically amounted (horizontal air flow path) CPU cooler. You may want to consider adding some more tests to this idea, specifically air flow pathing by way of adjusting the exhausting case fans.
When I saw the thumbnail I wondered why Trigglypuff was so upset about CPU cooler mounting. ...and sleeve bearing fans should be mounted vertically (horizontal air flow).
i've heard that if you have air cooled horizontal cpu cooler the blowing fan at the top of the case should be in the front mounting holes because the air pushing from the cooler should go directly to the back of the case fan and pushing the hot air out of the case easier.
i think the air flow from the frontal fans is pushing more air faster than the CPU cooler can absorb, like if the evo 212 where taking the cooler air thats coming from the front while the air from the unrefrigerated backplate is being push back and up by the air coming from the front and out by the fan in the back, also the third party vga could be running cooler so there is less heat coming out from it
There are actually alot of great coolers in the same range of the 212 EVO. Starting with the Hyper T4, going over the Hyper 212X, H7, Brocken ECO, Matterhorn Pure Gammaxx S40, and all the way to the Macho 90 and True Spirit 120. The thing ist just, that the 212 EVO has been established as a cooler fitting to all recent sockets,, relatively easy to mount for a tower cooler, and with a good price to peformance ratio. In the end it all depends on availability and space inside the case.
ziljin harder to get your hands on, a couple bucks more pricy, but significantly better. Mounting is easy, it has much better clearance in every way, and it performs a lot better. I've used both and I always advise people to spend the extra $5-10 to get the H7
The blower fan was catching cool air from the front fans which made the temperature the lowest. The other config had the cpu blower pumping in warm air from the cpu shroud and exhausting warm air, not to mention the case air flow was obstructed by the side of the fan shroud rather than pushing cool air through while immediately exhausting it out the back.
so nobody said anything about it,i think you got higher temp on the cpu cuz the backplate was releasing heat right on the heatsink,while the other gpu back plate would not get as hot and increase airflow and in the second test the result were different cus the cpu cooler was cooling the backplate of the blower gpu(that was heating up the cpu in the first test)
+Reo! Vendredi Funny/annoying fact. My Intel 3970x from 2012 with my Gigabyte GTX 1080 Xtreme handles everything in VR, even using 2x supersampling via the Oculus debug tool, JUST FINE ....YET the stupid Oculus software has that dumb warning stating my hardware doesn't meet recommended spec that never goes away!
It's a real thing, I even upgraded to Kaby for it. In games with higher CPU requirements (physics, terrain generation, etc) you need to account for the bottleneck. If any component in your PC is causing a lower-than-acceptable framerate, it needs to be changed. (Even the case applies, if airflow is bad you'll need to fix it)
For the first test, as there are 3 intake fans and 2 exhaust, blower card reduces amount of air pulled through those 2 exhaust fans, therefore less is being pushed through the horizontally opposed CPU cooler.
Blower style card is creating an airflow dead zone and trapping heat between the the cpu cooler and the gpu. With vertical orientation of the CPU cooler eliminates the dead zone...you could probably illustrate it with some smoke
5:13, the difference is that the 2nd card adds to the airflow of the case. The net amount of heat being exhausted through the first card, doesn't help the ambient heat generated immediately around the 1 st card. 2nd card piggybacks onto the air flow you have.
I actually use almost that same config right now (with open shroud GPU). I have SilentiumPC Pax M70 case with the config: 2x120 intake fronts, 120+140 intake tops and only one 120 exhaust on the back. It gives me a lot of positive pressure (good dust control) and I achieve the lowest temperatures from amongst any other config (but with horizontal CPU airflow).
One thought as to why the shroud + horizontal was hotter, perhaps is because the video card's radiant heat was rising and directly warming the heat sink. Being that the CPU heat sink fan was directing air horizontally and not drawing air upward directly, less thermal dissipation was occurring in that area resulting in higher temperatures. One thing you can do if you revisit this, is get your hands on a FLIR camera so you can visibly observe any heat pooling.
Now I don't know everything about PC building, or how this heat physics stuff work. But would the Backplate on the Zotac affect the temps? Even by a bit?
Back in the day, I used to run the back fan blowing in directly into the CPU heatsink fan and fans in the front of the case exhausting all the heat from inside the case. No clue if it worked better or not. Now I just custom water loop and don't worry about it.
The horizontal mounting probably pulled the cold air from the front intake fans, creating a thermocline (A buffer of colder air preventing the warmer GPU exhaust from mixing with the air pulled through the CPU fan). Like Jay said, can be seen if you use a smoke test.
With the horizontal mount and the open fan gpu, 100% of airflow goes through or near the CPU cooler, and the fan is not intaking directly off the hot back of the GPU. All three front fans are inline with the fan drawing air through the cooler, so I imagine there is a LOT of air going through there, and most of it straight from the outside of the case. In the horizontal mount, the blower fan actually steals airflow because it provides an alternate escape path for air. With the vertical mount, it comes down more to case temp, because the cooling fins are perpendicular to the incoming air from the front of the case, therefore all air going through the CPU fan has either circled the case or is coming directly off the back of the hot GPU. With the blower fan, you lower the case temps by a few degrees, probably the entire difference you are seeing.
I suppose that the blower style card pulls air out, and by acting as a powerful exhaust fan it makes the pressure inside the case negative, and if the pressure is negative then the cpu gets less dense air going through the cooler to cool it...so that's my theory.
I can confirm that moving my h115i water cooler's radiator from the top of my Corsair 750D case to the front dropped my CPU temperatures from a max of 98C on full load using Aida64 to 68C. Absolutely incredible.
I think something to keep in mind is that board partner coolers are usually more powerful, larger fans, better technological advances, etc. With the cooler horizontal, it might be that more surface area of the cooler is in the way of the airflow from the card and therefore heating it up further. Not really sure
@allupro I actually did that config some time ago. I had a terrible case (a little bit old) with very stupid fan placement, so I reversed the air flow in the whole case as you described (with open shroud GPU). With no other changes (only reversing cpu airflow and reversing intake and exhaust case fans) i dropped down about 15 degrees on CPU and about 10 degrees on GPU. Keep in mind that it was unusual situation (stupid case fan placements) so you may not get any positive effect.
presuming that heat generation is equal in all tests what is happening is either the configurations affect total airflow into/out of the case or they affect airflow inside the case. to test this you could measure air pressure going into the case and measure the GPU, Mobo temps and put some thermometers inside the case
It could be that since blower-style coolers are less effective, the gpu was warmer and then the air AROUND the GPU heated as a result. This air then rose up into the cpu cooler, increasing temps. Or it could just be within margin of error. :P
Kim Jong-Un yes it does. With the vertical blower the heat from the back plate does not get to the cooler fan intake. Real question is what is happening to GPU temperature in the four cases. Case 3 is best for CPU, but Case 2 is probably much better for GPU.
Blower style coolers are MOST effective. They extract the heat DIRECTLY outside the case. Which keeps the ambient temp inside the case lower, which benefits all devices inside the computer.
Wrong! Literally every review of blower style coolers- like Nvidia Founders Edition and AMD Vega models- have proven that those cards run MUCH hotter that AIB cards (even though the fans exhaust into the case). That's why the FE cards are horrible for overclocking without liquid cooling; they get to 85°C in no time and then throttle down performance to keep temps in check. There are tons of examples of this on the internet. It's not even debatable. With a decent case with good airflow, it's not an issue. There's no way a single 80mm (?) blower cooler is keeping up the THREE 100mm fans on my Aorus 1080 Ti.
shantfack Wrong. Change the fan speed curve and the cards wont even reach 60c under most circumstances. Your personal experience, and poor unexperienced reviews dont equal facts. As I said, the advantage of a blower style cooler is having the heat extracted directly out of the case. Your 3 100mm fans do not do that. They extract the heat into the case. Which in turn, keeps all of the heat inside the case and raises your entire case temps which increases card temps in return. A single blower fan can move around 50 cfm of air, or about the same as your average 120mm case fan
Maybe the weird results have to do with opposing fan currents? Maybe the open style GPU cooler disperses air through the case in a way that allows the CPU cooler to pull in more outside (cooler) air from that empty fan spot in the top of the case, resulting in cooler CPU temps.
The only thing i can think of is with the horizontal fan it is easier for it to suck in more hot air and blow it out pulling from 2 sides (below and to the right) while the vertical would have a harder time pulling air from the right side, and with the open shroud it allows more heat to escape which in turn giving more air for the fan to blow out of the case... just a thought
concerning your horizontal scenario hot air rises up, so it passes through the cpu heatsink, which results in potential airflow through the cooler, that could be the answer because it won´t do that with a blower style cooler because the hot air gets ejected out the back
The cool air from the front intake is being dragged across the fins by the rear exhaust fan, where as when the HSF is oriented to the front, the normal way, the front fan really isn't doing that much. It's like you have fans on all four faces of the heatsink instead of just the normal two. Turn the fans on the front off and you will see.
Random thought... Anyone who has a $400 GPU, a $200 motherboard, and a $350 CPU who wants to use a $30 CPU cooler needs to reexamine their PC part choices... Nice idea for a test, but not very real world in my opinion, try this again with a NH-D15 or other high end air cooler... Or... you know, just liquid cool it and be done with it. :)
Cooper Zirbel nothing wrong with the cooler, but it's just highly unbalanced for the build. More power = more heat, and cheaping out on the cooling makes no sense
You couldn’t be more wrong there because its very real world. You don’t need anything more than a H212 when it comes to air cooling because the next thing you should be looking at is an AIO cooler. Just like a stock cpu doesn’t need more than its factory cooler but people still buy aftermarket air coolers for them lol. Back when the 4790k was still new, the rig I built was at these similar price points and it has been running @4.8ghz with a hyper 212 and its still going strong. The temps stay in the high 60’s currently even with this heat wave. Sure I *could* have gone with an AIO and most likely clocked over 5ghz but I saw no point the rig was plenty fast and still handles everything I throw at it with no thermal issues. Just because you like to throw in an expensive cooler on every rig you make doesn’t mean it needs it lol.
I have a core i5 6600k OC'ed at 4.6ghz super stable on a hyper 212 turbo that stays around 55c even on full load. People have time and again talked about their core i7 overclocking like a beast and achieving stable core clocks on a Hyper 212 and never needing anything more. So please, explain to me again how one would have to reexamine their PC part choices if they use a $30 cooler on a $350 CPU?
Heat rises, adding the extra airflow from the fans allowed it to circulated more vs the blower fan., hence the increased dispassion from the extra fans allowed the heat to redistribute throughout the case more rather then it being concentrated. Also it has more resistance to suck the hot air out. pushing it out the back,its due to having a exhaust fan in the top of the case really in the end and the heat is less concentrated..
first test has to be that the fan is getting all the cool air from the front intakes..the open air card probably throws the hot air out the sides and it gets countered by the cool air. but the direction of the fan and heat sink. it does make a lot of sense as the cool air has a stronger presence
With this heat sink, I did a horizontal position and did a the push pull style. And got crazy results. This Is by far my favorite cooler. So I suggest doing to same thing for even better results, do 2 fans on the heatsink instead of just 1.
Kyle, Blower style cards don't have direct airflow on their backplates, making the surrounding air hotter. With the open shroud design, the cooler had direct airflow access to the backplate allowing the backplate to cool down resulting in cooler temperatures surrounding the tower, and therefore reducing CPU temperatures. Not only that but if you have mismatching fan speeds on the tower and top case fans you may actually blow heat back into that pocket. Also, I believe positive pressure, while being dust friendly, actually helps create small circulating pockets of heat in certain parts of some cases. Here with your choices, the bottom front case fan would feed the blower, and the top two would gravitate more towards the top of the case, NOT towards the backside of the gfx card (the majority of the air would be flowing towards the path of least resistance, and there is less resistance where the exhaust fans are flowing due to less pressure being there). With the open shroud, the gfx fans could push the surrounding air around back of the backplate and disrupt the pocket to allow for better case fan operation. But hey, that's just a theory, a possibly complete BS theory. Thanks for reading. AND if you do happen to read this, Kyle, test the statement, "positive pressure, while being dust friendly, actually helps create small circulating pockets of heat" because the flipside to it is that with negative pressure you have no other choice but to bring in cooler air into the case.
hypothesis: blower fan reduces pressure that the cooler can generate. open air fan rebounding air actually helps push cool air into the cooler. the reason it doesnt help in the vertical scenario is because the vertical fan takes in the warm air instead of the cool.
Sounds like air flow is the key even if you use already heated air for the CPU. The open GPU seems to have created a veil of hot air streaming up that stopped the fresh air from the front of the case reaching the CPU fan. I've read that having the two rear exhaust fans pointing back and up so close makes them compete for the same air in the corner.
When it was horizontal it was pulling all the fresh air from the front of the case, that's why it was cooler. Vertical only had 1 to pull from and it was moving against it with flow.
The only explanation for the higher heat with a horizontal mount with a blower style gpu is that it takes in air from the ram fins directly, and when using the open air gpu, the air is slightly directed upward and the front fans can reach it with cooler air.
The founders card during the horizontal test was pushing hot air on too and up the motherboard unlike the open GPU which allows for heat to scape from every side of the card
did not read through the comments but if that case has a clear side panel install it and run the system with smoke flowing through. you may be able to see the air flow and how each card and positioning of the CPU cooler affects the overall flow. it is possible that there are different pressures based on which way the fans are blowing.
it's about the overall flow of the air. you should put cameras inside for a cross section view then get a smoker or a smoke/fog machine. That will show you how the flow of air is actually tracking through the case. But it seems the orientation for the cpu cooler is also dependent on what kind of GPU cooling solution you end up getting.
Try it with negative pressure only, meaning only fans blowing outward. Two up top, and one in the back. That's what I still run, as I never changed anything since I bought my case. Also make sure your room is closed. Sure your house might be temperature controlled, but different areas of the house are still relatively warmer or cooler depending on what's in them.
Because the founders edition is a closed blower gpu then possibly the hot air being pushed through the enclosure to the end increases the overall temp of the backplate. The open blower gpu is taking heat from all over the gpu and sending it to the whole system, not just predominantly the backplate. The horizontal heatsink being closer to the backplate could explain the higher temp for the closed blower as its backplate is probably a bit hotter than the open blower.
It's all about airflow. CPU cooler in horizontal position is taking in cooler air as the open air video card is disrupting the cool airflow from the front fans; instead of going under the video card (as it is with blower card) more cool air (from front fans) is travelling over the top of open air video card and through CPU cooler.
if u can use smoke to check the air flow it might explain the 1st test.... the open shroud might give the tower a better air flow even if its a bit hotter.... unlikely but possible?
Put the side panel back on, with the open air design its able to freely exhaust the hot air out of the side where as the blower style has to force it out of the back. Also you should realize blowers tend to run hotter than open air in general.
On your horizontal fan i think in the cooler it creates an air pocket within the fins not allowing the air from the open graphics card to get in... the enclosed graphics card is maybe radiating hotter from the rear components... you'll need a laser temp gauge to test this or even better a thermal camera if you can borrow one.
Well just a hunch, when the open gpu lets out all the hot air since hot air rises it will naturally go up but since the horizontal cpu cooler is blowing air out the back of the case there is a negative pressure zone so all the hot gpu air is rushed out the back into that zone which therefor kind of causes air from the cpu cooler leave faster so the overall cpu temp drops as the cooler is releasing air faster. // correct me if im wrong its late idk
there is a possibility that the increase in air pushed into the case leads to greater circulation that then reduces the time the hot air has to heat the cpu
the first two results is due to the front case fans feeding the cpu cooler with fresh cold air, which is also dissipating the gpu heat. as where in the second result, the cpu cooler is sucking up the hot air from the zotac card
The reason is actually more about flow paths. In the horizontal stream, the CPU is pulling the heat up and off the vid, but, not into the cpu cooler. It is channeling it out the back by creating a negative pressure area that sucks the heat out the back. With the enclosed vid card, it is not adding to the "streams" of air in the system the same way (open cards blow air into the card, then around it, so it can slip stream if you have enough pull /push from the front and rear fans) Since the enclosed card is pushing it's heat out the back, you get the radiant heat from the back of the card that eddies and flows boack into the cpu cooler When it is facing the other way, you pull the heated air from the open gpu into the heatsink for the cpu, increasing temps, but, since more of the heat from the closed gpu goes out the back in this configuration(and the cpu fan blows the radiant heat out faster) the enclosed will be cooler. You should have checked GPU temps in both configs too.
Maybe with the Blower style GPU, you end up with lower Positive air pressure? In that configuration you effectively have 3 exhaust fans, whereas with the open style GPU, you only have 2 exhausts... I guess to put it another way, perhaps the CPU fan and Blower fan were competing for the same intake air..?
The amount of fans isnt as important as the air they can move. Most case fans are around 50cfm, which is pretty low. Stock blower fans move about half of that, or about the same at best.
the reason why the open air had lower temps compared to the blower, i believe, is due to the fans pushing more air through the case. Even though it may be hotter than the ambient air in the case it allows more air to get pushed through the cooler and cool the CPU more.
The fan was pulling cool air over the RGB lighting, that's why the temps were lower. RGB makes everything cooler.
ThisIsAPrettyLongUserNameWhyTheFuckWouldHeDoThatSoThatEveryoneWouldThumbsUpIt nice
ThisIsAPrettyLongUserNameWhyTheFuckWouldHeDoThatSoThatEveryoneWouldThumbsUpIt lul
ThisIsAPrettyLongUserNameWhyTheFuckWouldHeDoThatSoThatEveryoneWouldThumbsUpIt top kek
I was going to like your comment, but then I read your name. Tisk, tisk, sir.
ThisIsAPrettyLongUserNameWhyTheFuckWouldHeDoThatSoThatEveryoneWouldThumbsUpIt ThisIsAPrettyLongUserNameWhyTheFuckWouldYouDoThat?
5:47 you're welcome.
Maybe, if you could use a tempered glass side panel with some smoke, to better visualize the airflow, would make a great part 2
Wouldn't the moisture harm the parts
Smoke isn't the same as water vapour
I was about to suggest this. Great minds think alike.
kyle kyle.....KYYYLE KYYYYLE .....yea this guys has an idea tht you might want to hear ......yep, yep errr his name is nOTpASS
nOTpASS Nice idea
Would have been good to see GPU temps as well, to see how much they're affected by cpu cooler placement.
lexii miller vertical placement would probably decrease the gpu temps by ~1-2C for the open air gpu because it pulls the exhausting air from the gpu away more.
maybe founders edition is hotter in general so heat can transfer to cpu
the gpu is the same. 1070 is a 1070 no dif in the gpu itself.its all from Nvidia
different coolers make different temperatures.... The blower style cooler is *almost* always warmer than the other types
This is what I also think. The blower style cooled GPU gets hotter and radiates more heat and since the CPU and GPU coolers are drawing their air from the intakes it makes a still air pocket below the CPU cooler allowing the heat radiate from GPU to CPU and MB (so, there's none to little airflow between GPU and CPU cooler).
husssein But then that doesn't explain the vertical solutions.
That doesn't really makes sense. GPU vrams etc. should run as hot generating same amount of heat in both tests. The reason why you see higher temps in reference cooled cards is because those stock coolers and heat sinks simply struggle with transferring the heat from the chipset. But all this released energy in form of heat still has to go somewhere, ergo if you have a better cooler capable of transferring more heat from the card, the gpu itself will have lower temps but the relocated heat should mean higher temps on other pc components that are getting blasted by all that circulating hot air. Only thing I can think off is difference in voltage and core/mem clocks which means more emancipated heat but if anything FE should produce less energy than non-ref cards with additional pins and higher boost clocks. All of this leads me to believe that it's a margin of error or some unknown law of aerodynamics or thermodynamics that I simply don't know :/
Suggestion - Rerun the tests with the strange results, you might well find they change. Did you have enough (or too much) thermal interface material? Did you have just a poor seating of the heatsink on the CPU?
Yes, I'm sure you know how to mount a cooler, I'm not doubting that, but mistakes happen, I know I've made them. Of course you might have already done this and gotten the same result, but it is the first thought that came to my mind, was to rerun the test with everything remounted fresh.
hi tech deals
Hello lmao
You can't have 'too' much thermal paste, it's been proven.
it's pretty interesting to see you commenting on an airflow video but when someone mentioned you didn't have any exhaust on your son's pc (your rear fan was oriented to be intake) during a live upgrade stream a few weeks ago you said "why does it matter? worry about your own pc" to the guy. rude af.
@@traaaaan Rude people love to give advice, but don't like taking it.
see how much GPU power you need to get games to run at 144hz at different resolutions
Jake Conner four peanuts and two butters²
wtf is the point, dumbest thing ever. just look at % usage of gpu and u can estimate power draw
he is talking about gpu graphical power not wattage pulled. dummy..
1080p? A gtx 1070 would do it for most (but not all) games. Gamersnexus just did a great video with a heavily overclocked gtx 1080 and 7700k and found that it wasn't enough to get 120 fps in most games at 1440p though. Really need to get into SLI at that point.
Not even my gtx 1080/i7 4790K 4.7ghz combo gives me 144 fps ultra 1080p in somewhat up to date AAA games.
If you really want to have constant 144fps to profit off your 144hz monitor, and you want to play on ultra/very high, you'll need at least a gtx 1080, as I said not even that is enough.
Think this has to do with the direction the air is circulating in the chassis.. This is something you may have to test using a glass side panel case and some smoke to see where the air flow is going..
Joe Jackson smoke is horrable on computer components......
Jesse Dodge he mean like a fog machine. Of course it would be bad to use actual smoke on a computer. There are fog machine to test the airflow in your PC
Team Ninja horrible”
Aaron Garza words
Team Ninja hahaha.
Did you re apply the thermal paste? Perhaps you did it better the second time over the first?
if you have years of experience and hundreds of builds behind you i dont think the thermal paste can be issue
+Gothin Fuklio Yeah i figured, never know though. Could be a lazy mistake.
lol you did see him piss coolant all over his rig.
Linus did a test showing the way you put on the thermal paste makes no difference unless it is way to little.
this
Can you make a video on how to get a wife?
WOPPER he needs to make a video on how to get cinnamon rolls
McMexicans
Step #1 - Get a wife
with RGB
become famous or rich
I only get girls in video games like The Witcher 3 and the Dragon Age franchise.
There are so many myths people claim when it comes to computers and this is one of them, I knew it would not make much difference, I got flamed in one of my builds for doing the same thing, so I rotated the heatsink for a 3 degree difference in temps. Good video Kyle.
by mi mixtapo
Have mine with a upwards flow as well.
First for the simple reason to geht the hot air away from my GPU so it doesn't reduce it's boost.
MWith my new card, I'm just to lazy to reassamble all that stuff. (Heavy cooler and a delidded chip: nothing to be done in 5 mins)
Besides that I have 2x 140mm in the top and 1x 120mm in the back, so clearly the case supports the vertical airflow. And there would be not nearly as much fresh air coming from the front with all my drive bays, compared to the case he used.
Same. Some foo went after me for mounting my dark rock pro 3 with a horizontal airflow pattern. Good luck fitting that vertically, let alone the mounting hardware doesn't even allow for it.
Desert eagle ace
I have it pointing upwards too, because otherwise my RAM won't fit
*Theories!*
*Test #1* - Blower was stealing all the cool air. Could test with a PC-safe "smoke machine".
*Test #2" - The Zotac card seems to have a more restrictive "Open" design forcing the bulk of the exhaust to the back and bottom allowing for less heat to be ejected into the case compared to an ASUS Strix for example. Also it may not have been stealing all the fresh air allowing the 212 to move more air across the fins. Could test with a more open-air designed card.
*Test #3* - "Direct Contact Heat-Pipes" may have lined up with the hot-spot on the IHS better where the die is mounted vertically. Could test using non-direct style cooler with each test. (Would take alot of time though...)
*Test #4* - The Zotac's backplate may get hotter then the "Founder's" card due to added power deliver or design changes and the backplate may be heating up the air at the 212's intake. Could test by measuring the air temp above both backplates. (may be a bit difficult to do though...)
I know there's that one person out there that's gonna have a rage-attack, but remember, they're just theories that were thought up immediately after watching the video. ^_^
(*Edit*) Went to the -Thought Chamber- restroom and thought of something else. I wonder if an anomaly caused utilizations to differ between tests. Were the frame-rates consistent and the GPU/CPU loads reported equally? Maybe a background Windows update or other process was causing CPU load to vary.
Absolutely, I thought about the air being hotter from the backplate but completely ignored the fact that the open-air GPU is still exhausting hot air. That and the broadside of the CPU cooler in test #4 is also making it much easier for the cooler to intake that exhaust while at the same time "tripling" the surface area that the warm air is passing though. It will be quite interesting if a follow-up is done for this video, I realize it would take a good amount of time to do but your input makes me want to know the answers even more.
Heck, this type of "Guess what caused it" video conclusion could even be it's own series of videos. Even if only once a month or so I find it quite interesting and enjoy the participation side of it.
Yeah, the retarded amount of case fans makes these figures meaningless.
WRONG. It would only be meaningless if all the results were identical. They are not, so obviously blower and orientation of the CPU cooler have an impact.
I notced same thing when change my GPU from "blower style" to "open". But final results for my setup: Less fans is requied (2 removed from top) to make same temp. at CPU now. Blower style cards "stealing" air from case, makes CPU cooling less effective.
I noticed that as well, this is why I added a fan underneath the blower cooler of my GPU. The blower was sucking the air from the front intake fans (and was basically a wall separating top and bottom) and was radiating heat from the backplate up toward the CPU. Adding that bottom mounted fan didn't prevent the GPU from stealing the front intake airflow, but it cooled the card, which helped the CPU stay a bit cooler. Pretty interesting how that all works out.
it could be that the horizontal CPU & open GPU are actually avoiding each other... in a sense.
The GPU is pulling air straight up and the air is likely blooming away and above the the CPU intake pressure drop. Causing cooler air to collect for the CPU.
Where vertical is in the direct path of hot GPU air...
Great idea for a video.
The little tests that you and Jay do are really are service to the community and I appreciate all the time and effort that you put into your videos! Thanks once again for the info!
d00d nice video, what camera are you using to record yourself?
I believe Kyle uses the Panasonic Lumix GH4
correct!
Will you be upgrading to the GH5?
like a 1000$ camera Why I wish
Because the GH5 lets you do 4K at 60 FPS.
I believe its as simple as just good airflow, because as soon as you turn the cpu cooler towards the gpu and away from the front fans temps rose up, and since in-closed gpus are mainly always more hotter than open gpus the front fans werent really helping the in-closed gpu dissipate its radiating heat , but the open gpu even though its was dissipating warm air in the case the airflow was pushing it out
*The blower style GPU acts as the 3rd exhaust fan, there will be less air coming through top and back exhaust fan **---**> hotter CPU*
possibly the blower style since it's worse as dispersing heat a good portion of the hot air surrounding the card stuck around, while the open air one was better at getting hot air out and away.
@5:45 Just maybe this will explain scenario 2. CPU cooler blows Left, GPU blows upwards, with case flans blowing left and up. Heat is dissipated more evenly (maybe more efficiently) between the top and left of the case. This differs from option 4. However, this is very similar to option 3. CPU cooler blows upwards, while GPU blows Left. Heat going both directions in options 2 and 3. This is contrasted to option 1 ( Majority of air is dissipated left) and option 4 (majority of air is dissipated upwards) which results in higher CPU temps.
An analogy for scenario 2 would be having 2 straws in a beer: If you had to drink the whole beer using straws, then it would be able to get the same amount of liquid out of the can using two straws compared to pushing that same amount of liquid through mostly 1 straw.
This is also based on the concept of better cooling with a greater proportion of fresh air vs recycled air (air already cooling one component to cool another component).
freaking 4:30 minutes to start the video!
Tozixc
I use a hyper 212 evo to cool my i5 6600k @4.6ghz and the highest temp I've seen is 57°C
Marley I bet you didnt use Prime or did you?
I did, that's the weird part
Marley pics or didnt happen. Friend of me is using a 6600k with a beefy Dark Rock Pro 3 as cooler and doesnt even maintain 60 at 4
5 GHz.
maybe it's because my room is always 16°C
if your room temp is 16c then why bother buying cooler XD
Hot air rises, a vertical config with open gpu pulls air directly into the case.
Also, I noticed if the cpu, gpu airflow direction matches, temps are lower. Open gpu has vertical airflow, closed has horizontal.
Maybe since the open shroud GPU has multiple fans resulting in relatively more air circulation and creating a higher positive pressure unlike the blower style GPU... correct me if I am wrong
Indrajeet Pala air circulation from internal components can't magically pull air in from outside the case and raise the internal pressure.
horizontal mounting uses more fresh air from the front intake, vertical mounting the cooler is sucking the hot air from the gpu in = higher cpu temps with open style gpu cooler...nothing more...thus horizontal mounting the cpu cooler is better if you have an open gpu cooler...
But the blower style cooler is designed to pull air out of the case, so there would be less positive pressure.
Walt Findley .... if the case was fully sealed then yes a internal fan inside sealed cabinet will not suck air from outside .... but in a pc cabinet intenal fan can create a positive pressure "i think so" just like if you put two same fans on top of each other the total air moved will be double the air moved by single fan.
the components on the vga pcb, they also heat a little, so even with the blower style cooling, some minimal heat disipation will occur in that area
Linus: the sponsor guy
Dimitry: the professional editor
Jay: the graphic snob
Paul: the budget friendly tech tuber
Kyle: the comedian tech tuber
Did you make sure to put the case side panel on when you did these tests?
He always does for the tests, he leaves them off for the "glamour shots"
Neon_Lights12 the only way to do it imo. makes it look sexy
I think the horizontal position cpu cooler lowered the temp because the air flow is in a more consistent direction from to the intake fans airflow. Thus creating a more positive air pressure effect.
Ohhh let me take a stab. The blower gpu is taking air from within the case and pushing the air laterally across the gpu and out the back, so it never hits the cpu cooler. You'd think that's good for avoiding hot air circulating in the case and hitting the cpu cooler, but maybe that's bad for total airflow in the case overall. The less air circulating in the case, the less usable air there is for the cpu cooler. The open air gpu has air all around it, which in turn creates more airflow within the case with the same air. More airflow near the cpu cooler provides more air to cool the cpu. Just a theory.
LOL no the case would have to be operating at a negative pressure vacuum in order for that to be the cause. There is no side panel which makes this whole thing pointless anyway. Its likely the cause was either the difference in mounting (remember he's is unmounting and remounting everytime) position, pressure or some other aerodynamic factor.
Zero One yep plus the shitty 212 evo is shit, nobody cares about it
I'm gonna like her comment only cause bewbs. 😌
@@AdityaSingh97 wtf is wrong with you?
I see people have the cpu fan facing the rear fan, explain that people. 😫
One thing that everyone underestimates is interaction between heatspreader and heatsink. It's better to have the heatpipes passing perpendicularly on the "die" of the Intel CPUs.
In fact in the best conditions (for the CPU) with blower GPU, the temperature is lower, but open GPU has negatively influenced the results. Anyway horizontal+open GPU gives a strange result.
6:44 case closed, pun intended?
i once commented about a similar pun like that, and got massive hate lol.
@@NineS5 because its fucking cringe worthy.
@@jefftraboulsy8631 ikr i sounded like a plonker
My thoughts on this are that I have a different case, different fans, different graphics card, different cpu cooler, so I should do my own damn tests if I want relevant results.
I think you changed air flow characteristics by using the lower style fan. The blower style fan would effect the pressure of the case. The blower style has a larger negative pressure zone an then pulls more fresh air away from the CPU cooler?
You could check this by using a smoke machine.
I was really tired building my PC and accidentally installed the Cooler facing up and didn't fully realize until I put it in the case. Glad to see it doesn't make too much difference and need to waste the paste to fix it.
I usually like your videos before watching them. Just know that the content will be top notch.
I'd actually expect there to be a slight (perhaps not as much as 5 degrees C) difference based on where the exhaust fans are. Having both of them might mean they're pulling the airflow path slightly differently. If that's the situation, then the rear mounted fan may be expelling more GPU heat without it going through the vertically amounted (horizontal air flow path) CPU cooler.
You may want to consider adding some more tests to this idea, specifically air flow pathing by way of adjusting the exhausting case fans.
When I saw the thumbnail I wondered why Trigglypuff was so upset about CPU cooler mounting.
...and sleeve bearing fans should be mounted vertically (horizontal air flow).
i've heard that if you have air cooled horizontal cpu cooler the blowing fan at the top of the case should be in the front mounting holes because the air pushing from the cooler should go directly to the back of the case fan and pushing the hot air out of the case easier.
Funny how when he said horizontal the fan was in the vertical
Position and vice versa.
i think the air flow from the frontal fans is pushing more air faster than the CPU cooler can absorb, like if the evo 212 where taking the cooler air thats coming from the front while the air from the unrefrigerated backplate is being push back and up by the air coming from the front and out by the fan in the back, also the third party vga could be running cooler so there is less heat coming out from it
I heard Cryorig H7 is a great budget CPU cooler.
yes it is
Better than the 212 for a bit more. The H7 is less available around the world than the 212 though.
There are actually alot of great coolers in the same range of the 212 EVO.
Starting with the Hyper T4, going over the Hyper 212X, H7, Brocken ECO, Matterhorn Pure Gammaxx S40, and all the way to the Macho 90 and True Spirit 120.
The thing ist just, that the 212 EVO has been established as a cooler fitting to all recent sockets,, relatively easy to mount for a tower cooler, and with a good price to peformance ratio.
In the end it all depends on availability and space inside the case.
ziljin harder to get your hands on, a couple bucks more pricy, but significantly better. Mounting is easy, it has much better clearance in every way, and it performs a lot better. I've used both and I always advise people to spend the extra $5-10 to get the H7
ziljin get the h5 i got it and under 55° on i5 6600k at 4.5ghz
The blower fan was catching cool air from the front fans which made the temperature the lowest. The other config had the cpu blower pumping in warm air from the cpu shroud and exhausting warm air, not to mention the case air flow was obstructed by the side of the fan shroud rather than pushing cool air through while immediately exhausting it out the back.
I really like these experiment videos, keep it up
so nobody said anything about it,i think you got higher temp on the cpu cuz the backplate was releasing heat right on the heatsink,while the other gpu back plate would not get as hot and increase airflow
and in the second test the result were different cus the cpu cooler was cooling the backplate of the blower gpu(that was heating up the cpu in the first test)
Seriously?
"VR compatible" even on CPU?!
>.
Yeah and say me also that we need VR compatible HD ready RGB tempered side glass panel...
+Reo! Vendredi Funny/annoying fact. My Intel 3970x from 2012 with my Gigabyte GTX 1080 Xtreme handles everything in VR, even using 2x supersampling via the Oculus debug tool, JUST FINE ....YET the stupid Oculus software has that dumb warning stating my hardware doesn't meet recommended spec that never goes away!
DMan 9971 I'm sure you can erase that message in registry.
Arek R. Damn right you do! but you're missing one thing. fire to make the CPU go fastet
It's a real thing, I even upgraded to Kaby for it. In games with higher CPU requirements (physics, terrain generation, etc) you need to account for the bottleneck. If any component in your PC is causing a lower-than-acceptable framerate, it needs to be changed. (Even the case applies, if airflow is bad you'll need to fix it)
For the first test, as there are 3 intake fans and 2 exhaust, blower card reduces amount of air pulled through those 2 exhaust fans, therefore less is being pushed through the horizontally opposed CPU cooler.
ayy lmao, that's me!
Eggsr2bcrushed yes
That is correct
I was in there too. Support group?
Blower style card is creating an airflow dead zone and trapping heat between the the cpu cooler and the gpu. With vertical orientation of the CPU cooler eliminates the dead zone...you could probably illustrate it with some smoke
I love when Kyle gets all sciencey
5:13, the difference is that the 2nd card adds to the airflow of the case. The net amount of heat being exhausted through the first card, doesn't help the ambient heat generated immediately around the 1 st card. 2nd card piggybacks onto the air flow you have.
What about pulling cold air from the top of the case. And let the cpu cooler blow air down on the gpu?
I actually use almost that same config right now (with open shroud GPU). I have SilentiumPC Pax M70 case with the config: 2x120 intake fronts, 120+140 intake tops and only one 120 exhaust on the back. It gives me a lot of positive pressure (good dust control) and I achieve the lowest temperatures from amongst any other config (but with horizontal CPU airflow).
Heat rises so it may be a wise choice to find a flow with that in mind.
One thought as to why the shroud + horizontal was hotter, perhaps is because the video card's radiant heat was rising and directly warming the heat sink. Being that the CPU heat sink fan was directing air horizontally and not drawing air upward directly, less thermal dissipation was occurring in that area resulting in higher temperatures. One thing you can do if you revisit this, is get your hands on a FLIR camera so you can visibly observe any heat pooling.
Now I don't know everything about PC building, or how this heat physics stuff work.
But would the Backplate on the Zotac affect the temps? Even by a bit?
Back in the day, I used to run the back fan blowing in directly into the CPU heatsink fan and fans in the front of the case exhausting all the heat from inside the case. No clue if it worked better or not.
Now I just custom water loop and don't worry about it.
Is it me or is the volume lower on this video.
Uiriamu Busujima i think it is lower,also.
The horizontal mounting probably pulled the cold air from the front intake fans, creating a thermocline (A buffer of colder air preventing the warmer GPU exhaust from mixing with the air pulled through the CPU fan).
Like Jay said, can be seen if you use a smoke test.
Is it just me or does his mic quality sound a little worse in this video?
its you
Nova its just you
I have a hyperx cloud 2 and it sounds better on almost any other video.
He seems to have some bad processing going on compared to normal.
It's like, really compressed this timed around.
With the horizontal mount and the open fan gpu, 100% of airflow goes through or near the CPU cooler, and the fan is not intaking directly off the hot back of the GPU. All three front fans are inline with the fan drawing air through the cooler, so I imagine there is a LOT of air going through there, and most of it straight from the outside of the case. In the horizontal mount, the blower fan actually steals airflow because it provides an alternate escape path for air.
With the vertical mount, it comes down more to case temp, because the cooling fins are perpendicular to the incoming air from the front of the case, therefore all air going through the CPU fan has either circled the case or is coming directly off the back of the hot GPU. With the blower fan, you lower the case temps by a few degrees, probably the entire difference you are seeing.
******[QUESTION] CPU TEMP WITH SAME LENGTH RADIATOR AT DIFFERENT THICKNESS'S******
I suppose that the blower style card pulls air out, and by acting as a powerful exhaust fan it makes the pressure inside the case negative, and if the pressure is negative then the cpu gets less dense air going through the cooler to cool it...so that's my theory.
I can confirm that moving my h115i water cooler's radiator from the top of my Corsair 750D case to the front dropped my CPU temperatures from a max of 98C on full load using Aida64 to 68C. Absolutely incredible.
I think something to keep in mind is that board partner coolers are usually more powerful, larger fans, better technological advances, etc. With the cooler horizontal, it might be that more surface area of the cooler is in the way of the airflow from the card and therefore heating it up further. Not really sure
Can you test with the CPU cooler horizontal, but in reverse so it draws air from the back of the case?
@allupro I actually did that config some time ago. I had a terrible case (a little bit old) with very stupid fan placement, so I reversed the air flow in the whole case as you described (with open shroud GPU). With no other changes (only reversing cpu airflow and reversing intake and exhaust case fans) i dropped down about 15 degrees on CPU and about 10 degrees on GPU.
Keep in mind that it was unusual situation (stupid case fan placements) so you may not get any positive effect.
presuming that heat generation is equal in all tests what is happening is either the configurations affect total airflow into/out of the case or they affect airflow inside the case. to test this you could measure air pressure going into the case and measure the GPU, Mobo temps and put some thermometers inside the case
Very informative video. Thanks although I could have definitely lived without the after breakfast sound effects.
Wow reale impressive how close You are to your Community! Big respect!
It could be that since blower-style coolers are less effective, the gpu was warmer and then the air AROUND the GPU heated as a result. This air then rose up into the cpu cooler, increasing temps.
Or it could just be within margin of error. :P
But that doesn't really explain the 2nd test
Kim Jong-Un
yes it does. With the vertical blower the heat from the back plate does not get to the cooler fan intake. Real question is what is happening to GPU temperature in the four cases. Case 3 is best for CPU, but Case 2 is probably much better for GPU.
Blower style coolers are MOST effective. They extract the heat DIRECTLY outside the case. Which keeps the ambient temp inside the case lower, which benefits all devices inside the computer.
Wrong!
Literally every review of blower style coolers- like Nvidia Founders Edition and AMD Vega models- have proven that those cards run MUCH hotter that AIB cards (even though the fans exhaust into the case). That's why the FE cards are horrible for overclocking without liquid cooling; they get to 85°C in no time and then throttle down performance to keep temps in check. There are tons of examples of this on the internet. It's not even debatable.
With a decent case with good airflow, it's not an issue. There's no way a single 80mm (?) blower cooler is keeping up the THREE 100mm fans on my Aorus 1080 Ti.
shantfack Wrong. Change the fan speed curve and the cards wont even reach 60c under most circumstances.
Your personal experience, and poor unexperienced reviews dont equal facts.
As I said, the advantage of a blower style cooler is having the heat extracted directly out of the case. Your 3 100mm fans do not do that. They extract the heat into the case. Which in turn, keeps all of the heat inside the case and raises your entire case temps which increases card temps in return.
A single blower fan can move around 50 cfm of air, or about the same as your average 120mm case fan
Maybe the weird results have to do with opposing fan currents? Maybe the open style GPU cooler disperses air through the case in a way that allows the CPU cooler to pull in more outside (cooler) air from that empty fan spot in the top of the case, resulting in cooler CPU temps.
Alternative results, yo'.
Amun 'Omii' Omar lmao
The only thing i can think of is with the horizontal fan it is easier for it to suck in more hot air and blow it out pulling from 2 sides (below and to the right) while the vertical would have a harder time pulling air from the right side, and with the open shroud it allows more heat to escape which in turn giving more air for the fan to blow out of the case... just a thought
Hint: measure the temp of the back of the GPU.
concerning your horizontal scenario
hot air rises up, so it passes through the cpu heatsink, which results in potential airflow through the cooler, that could be the answer because it won´t do that with a blower style cooler because the hot air gets ejected out the back
The answer is buy a FLIR one.
The cool air from the front intake is being dragged across the fins by the rear exhaust fan, where as when the HSF is oriented to the front, the normal way, the front fan really isn't doing that much. It's like you have fans on all four faces of the heatsink instead of just the normal two. Turn the fans on the front off and you will see.
Random thought... Anyone who has a $400 GPU, a $200 motherboard, and a $350 CPU who wants to use a $30 CPU cooler needs to reexamine their PC part choices...
Nice idea for a test, but not very real world in my opinion, try this again with a NH-D15 or other high end air cooler...
Or... you know, just liquid cool it and be done with it. :)
Is there something wrong with that cooler?
Cooper Zirbel nothing wrong with the cooler, but it's just highly unbalanced for the build. More power = more heat, and cheaping out on the cooling makes no sense
You couldn’t be more wrong there because its very real world. You don’t need anything more than a H212 when it comes to air cooling because the next thing you should be looking at is an AIO cooler. Just like a stock cpu doesn’t need more than its factory cooler but people still buy aftermarket air coolers for them lol. Back when the 4790k was still new, the rig I built was at these similar price points and it has been running @4.8ghz with a hyper 212 and its still going strong. The temps stay in the high 60’s currently even with this heat wave. Sure I *could* have gone with an AIO and most likely clocked over 5ghz but I saw no point the rig was plenty fast and still handles everything I throw at it with no thermal issues.
Just because you like to throw in an expensive cooler on every rig you make doesn’t mean it needs it lol.
I have a core i5 6600k OC'ed at 4.6ghz super stable on a hyper 212 turbo that stays around 55c even on full load. People have time and again talked about their core i7 overclocking like a beast and achieving stable core clocks on a Hyper 212 and never needing anything more.
So please, explain to me again how one would have to reexamine their PC part choices if they use a $30 cooler on a $350 CPU?
Wrong! Hyper 212 is a very capable cooler
Heat rises, adding the extra airflow from the fans allowed it to circulated more vs the blower fan., hence the increased dispassion from the extra fans allowed the heat to redistribute throughout the case more rather then it being concentrated. Also it has more resistance to suck the hot air out. pushing it out the back,its due to having a exhaust fan in the top of the case really in the end and the heat is less concentrated..
Hey bitwit, can I safely oc my i7 7700k to 5ghz with the hyper 212 evo?
Andrew Robinson Depends of the quality of your CPU.
Andrew Robinson highly unlikely
+Fresh Dumbledore ...i7 7700k...
+freestuffgood dang. What do you think I can I get to?
Andrew Robinson Every CPU is a bit different. Even if its the same modell. Some i7 7700k's can be oc'd better than others ^^. Also, whats your PSU?
couldnt tell you but 5 seems like alot for a non watercooled build
first test has to be that the fan is getting all the cool air from the front intakes..the open air card probably throws the hot air out the sides and it gets countered by the cool air.
but the direction of the fan and heat sink. it does make a lot of sense as the cool air has a stronger presence
I blame Obama...Global warming
With this heat sink, I did a horizontal position and did a the push pull style. And got crazy results. This Is by far my favorite cooler.
So I suggest doing to same thing for even better results, do 2 fans on the heatsink instead of just 1.
Test how much coke you can snort verticaly and horizontali, and using metal and copper snorting tube. Compare the results
Kyle,
Blower style cards don't have direct airflow on their backplates, making the surrounding air hotter. With the open shroud design, the cooler had direct airflow access to the backplate allowing the backplate to cool down resulting in cooler temperatures surrounding the tower, and therefore reducing CPU temperatures. Not only that but if you have mismatching fan speeds on the tower and top case fans you may actually blow heat back into that pocket.
Also, I believe positive pressure, while being dust friendly, actually helps create small circulating pockets of heat in certain parts of some cases. Here with your choices, the bottom front case fan would feed the blower, and the top two would gravitate more towards the top of the case, NOT towards the backside of the gfx card (the majority of the air would be flowing towards the path of least resistance, and there is less resistance where the exhaust fans are flowing due to less pressure being there). With the open shroud, the gfx fans could push the surrounding air around back of the backplate and disrupt the pocket to allow for better case fan operation.
But hey, that's just a theory, a possibly complete BS theory. Thanks for reading.
AND if you do happen to read this, Kyle, test the statement, "positive pressure, while being dust friendly, actually helps create small circulating pockets of heat" because the flipside to it is that with negative pressure you have no other choice but to bring in cooler air into the case.
hypothesis: blower fan reduces pressure that the cooler can generate. open air fan rebounding air actually helps push cool air into the cooler. the reason it doesnt help in the vertical scenario is because the vertical fan takes in the warm air instead of the cool.
Sounds like air flow is the key even if you use already heated air for the CPU. The open GPU seems to have created a veil of hot air streaming up that stopped the fresh air from the front of the case reaching the CPU fan. I've read that having the two rear exhaust fans pointing back and up so close makes them compete for the same air in the corner.
When it was horizontal it was pulling all the fresh air from the front of the case, that's why it was cooler. Vertical only had 1 to pull from and it was moving against it with flow.
The only explanation for the higher heat with a horizontal mount with a blower style gpu is that it takes in air from the ram fins directly, and when using the open air gpu, the air is slightly directed upward and the front fans can reach it with cooler air.
The blower style backplate gets hotter than aftermarket.
The founders card during the horizontal test was pushing hot air on too and up the motherboard unlike the open GPU which allows for heat to scape from every side of the card
did not read through the comments but if that case has a clear side panel install it and run the system with smoke flowing through. you may be able to see the air flow and how each card and positioning of the CPU cooler affects the overall flow. it is possible that there are different pressures based on which way the fans are blowing.
it's about the overall flow of the air. you should put cameras inside for a cross section view then get a smoker or a smoke/fog machine. That will show you how the flow of air is actually tracking through the case. But it seems the orientation for the cpu cooler is also dependent on what kind of GPU cooling solution you end up getting.
Try it with negative pressure only, meaning only fans blowing outward. Two up top, and one in the back. That's what I still run, as I never changed anything since I bought my case. Also make sure your room is closed. Sure your house might be temperature controlled, but different areas of the house are still relatively warmer or cooler depending on what's in them.
Because the founders edition is a closed blower gpu then possibly the hot air being pushed through the enclosure to the end increases the overall temp of the backplate. The open blower gpu is taking heat from all over the gpu and sending it to the whole system, not just predominantly the backplate. The horizontal heatsink being closer to the backplate could explain the higher temp for the closed blower as its backplate is probably a bit hotter than the open blower.
It's all about airflow.
CPU cooler in horizontal position is taking in cooler air as the open air video card is disrupting the cool airflow from the front fans; instead of going under the video card (as it is with blower card) more cool air (from front fans) is travelling over the top of open air video card and through CPU cooler.
if u can use smoke to check the air flow it might explain the 1st test.... the open shroud might give the tower a better air flow even if its a bit hotter.... unlikely but possible?
Put the side panel back on, with the open air design its able to freely exhaust the hot air out of the side where as the blower style has to force it out of the back. Also you should realize blowers tend to run hotter than open air in general.
On your horizontal fan i think in the cooler it creates an air pocket within the fins not allowing the air from the open graphics card to get in... the enclosed graphics card is maybe radiating hotter from the rear components... you'll need a laser temp gauge to test this or even better a thermal camera if you can borrow one.
Well just a hunch, when the open gpu lets out all the hot air since hot air rises it will naturally go up but since the horizontal cpu cooler is blowing air out the back of the case there is a negative pressure zone so all the hot gpu air is rushed out the back into that zone which therefor kind of causes air from the cpu cooler leave faster so the overall cpu temp drops as the cooler is releasing air faster. // correct me if im wrong its late idk
there is a possibility that the increase in air pushed into the case leads to greater circulation that then reduces the time the hot air has to heat the cpu
the first two results is due to the front case fans feeding the cpu cooler with fresh cold air, which is also dissipating the gpu heat. as where in the second result, the cpu cooler is sucking up the hot air from the zotac card
The reason is actually more about flow paths. In the horizontal stream, the CPU is pulling the heat up and off the vid, but, not into the cpu cooler. It is channeling it out the back by creating a negative pressure area that sucks the heat out the back. With the enclosed vid card, it is not adding to the "streams" of air in the system the same way (open cards blow air into the card, then around it, so it can slip stream if you have enough pull /push from the front and rear fans) Since the enclosed card is pushing it's heat out the back, you get the radiant heat from the back of the card that eddies and flows boack into the cpu cooler
When it is facing the other way, you pull the heated air from the open gpu into the heatsink for the cpu, increasing temps, but, since more of the heat from the closed gpu goes out the back in this configuration(and the cpu fan blows the radiant heat out faster) the enclosed will be cooler. You should have checked GPU temps in both configs too.
Maybe with the Blower style GPU, you end up with lower Positive air pressure? In that configuration you effectively have 3 exhaust fans, whereas with the open style GPU, you only have 2 exhausts... I guess to put it another way, perhaps the CPU fan and Blower fan were competing for the same intake air..?
that's a good theory
The amount of fans isnt as important as the air they can move. Most case fans are around 50cfm, which is pretty low. Stock blower fans move about half of that, or about the same at best.
the reason why the open air had lower temps compared to the blower, i believe, is due to the fans pushing more air through the case. Even though it may be hotter than the ambient air in the case it allows more air to get pushed through the cooler and cool the CPU more.