[SOLVED] Front fans for Meshify C ?

[hornetmax]

Hi all,

I've got a Meshify C with a 2070 super and a 9900K, no overclock. Cooling-wise, I've a Thermalright Macho RevB, 2x Corsair AF140 front (intake) and 2x Fractal 120mm fans out (rear and top-rear).

Since the 9900K (had a 9600K before), CPU (and VRMs) temps go a bit high during heavy loads (e.g h265 encoding): VRMs hitting 98C, CPU cores overing at 88C, hitting (very occasioanlly) +95C and causing (again, very occasionally) thermal throttling. It's not a huge problem, but I'd be more comfortable with lower temps on both CPU and VRMs.

While under full load, I removed the front and top filter of the Meshify C (metal grid + foam): that gained me 6C on the CPU and 15C on the VRMs (!!).

Now I'm wondering: what shoud I do to have more flow from the two front fans ? Options I see:

1. Take the foam away from the front filter (keep only the metal grid).
2. Take the foam away from the front filter (keep only the metal grid) but put some good filtes on the front fans: things like Silverstone F143 or Demci Flex.
3. Swap the two front Corsair AF140 in the front for something else ? Higher static pressure to "overcome" the foam ? Which fan ?

Option 1: just a matter of judging fs the dust is manageable without the foam.
Option 2: I'm unsure how much better this would be compared to the foam. Any idea ?
Option 3: I'm OK with the fans being a bit loud if spinning fast, but at low loads I'd like them to be quiet (i.e. similar to the AF140 I have).

Any advice is welcome !

 

[Darkbreeze]

The first thing you should probably do is replace the stock Fractal fans with some fans that can actually move some air. I have a pile of those stock Fractal Dynamic X2 GP fans on the shelf above my workbench, and while they are quite good for included free fans, they don't compare at all to any of the better aftermarket PWM fans out there. Also, I'd replace the top-rear fan with a 140mm version as it will, obviously, generally be capable of moving more air at the same RPM or noise normalized load than a similar 120mm model.

Intake is a necessary, essential part of the cooling equation, but MORE intake beyond a point does not equal better cooling. More exhaust, just about ALWAYS results in an increase in cooling performance and it ALSO allows the front fans to work a lot more efficiently as there is less resistance to airflow into the case when there is a high level of negative pressure. But, since negative pressure tends to promote dust collection while positive pressure tends to suppress it, we usually try to go with a pretty much balanced, neutral pressure arrangement to get some of each benefit even though we don't get all of them both.

For purely performance related configurations a system where the cooling system is purely negative pressure would always outperform one that was tilted deep into positive pressure territory.

These, MIGHT make a difference. Should, actually, make a difference. And are a whole lot quieter than those X2 GP fans, so that even when running at full speed you can barely hear them from three feet away. The 140mm fan adds about an additional 30CFM of air movement OUT of the case, which is nearly as much as adding a whole extra fan by comparison. The 120mm Noctua fan doesn't add much, but keeps things uniform. If you want to use that fan to add additional performance and are willing to sacrifice a little bit of additional noise at higher RPMs, then the Noctua NF-F12 Industrial iPPC 2000rpm fans would add about an additional 20CFM air movement to the 30CFM gained with the 140mm fan which WOULD definitely be like add a whole other exhaust fan without the problems associated with botching the airflow path that occur when using a middle or front top exhaust fan on air cooled systems.

PCPartPicker Part List

Case Fan: Noctua F12 PWM chromax.black.swap 54.97 CFM 120 mm Fan ($22.90 @ Amazon)
Case Fan: Noctua A14 PWM chromax.black.swap 82.52 CFM 140 mm Fan ($24.90 @ Amazon)
Total: $47.80
Prices include shipping, taxes, and discounts when available
Generated by PCPartPicker 2021-02-14 17:04 EST-0500

Once you do THAT, THEN adding some additional performance through the front fans MIGHT be worth looking into. It might also not be a bad idea to think about a better CPU cooler. While the Macho rev.B IS a good cooler, it might actually be a bit lacking when it comes to your 9900k. Those big i7 and i9 CPUs with all those cores, these days, really need a twin finstack cooler, noting that the Macho IS a BIG finstack, but it does not have the surface area of any of the big twin finstack coolers.

 

[Karadjgne]

The Meshify C is about as good as it gets for an airflow case for an air cooler. From what I've seen, the best setup is just using the stock rear 120mm and 3x front fans, leaving the baffles intact on the top.

The difference in cpu/VRM temps with removing the top baffles is an easy explanation, if you understand how a fan works and air pressure. When a blade spins, it creates a low pressure area above it, the byproduct being the displaced air forced out the back. At higher rpm, the pressure is disproportionate, there's higher pressure on exhaust than intake, the stream is more condensed. Nature abhors a vacuum, so the largest, nearest, easiest path for air is outside being pushed by 14.7 psi into the case. Literally when you remove the top baffles, they become intake ports for the cpu and rear exhaust, bypassing most of the intake fans air which is drawn/blown towards the gpu.

So while you see a benefit in your mind as to cpu temps, reality is its a detriment to everything else as case air temps are higher. Like sucking on a straw that's got a crack in it, you get more flow, but less soda.

With the top intact, there's no choice. The cpu and rear exhaust have no options as the easiest path is whatever the intake fans supply. Having a high mount fan actually works in that case, as it pushes any ceiling heat towards the rear. With just the 2 lower fans in front, you get a Hotspot at the top/front, and thats a good portion of where the cpu cooler pulls from.

I can second Darkbreeze choice in fan replacements, I've got 4 of those in my own system, but also consider exactly how the air is moving, or not moving, and try different fan placements, different fan curves etc. Better fans will help existing good airflow, they rarely do much for bad airflow characteristics.

 

[hornetmax]

Thanks for the replies. Looking at the graphs I still think there's something bad going on at the front, not at the back. Let me explain.
Towards the end of the test I was running without the front and top filters. Both front (in) and rear+top rear (out) fans running at 100% as the VRM temps were above 80C.
At that point I have put the front filter on again (not the top one). What happened:

• Rear+top rear RPMs (the two fans are on a Y -splitter) went instantly down a bit. This is expected: with the front filter on now it's harder to pull air out of the case.
• Front fans RPMs went instantly up. That's weird no ?

All these fans have curves on the same probe (VRM temp).

It doesn't seem to be a fluke as I had the dual behavior when I removed the filters: rear+top fan (out) RPMs go upwhile front fans (in) RPM go down.

Why is this happening ?

 

[Darkbreeze]

It is very simple to determine if airflow through the case is the problem.

Remove the side panel and conduct your tests again. If the CPU temp doesn't drop by five or more degrees by simply taking the side panel off then nothing you do with airflow though the case is going to make any difference in that regard and you might as well shift your focus, at that point, to figuring out what else is going on such as not having the best thermal paste application, too much voltage somewhere, not good enough motherboard for the CPU model, etc.

 

[hornetmax]

It is very simple to determine if airflow through the case is the problem.

Remove the side panel and conduct your tests again. If the CPU temp doesn't drop by five or more degrees by simply taking the side panel off then nothing you do with airflow though the case is going to make any difference in that regard

But isn't this already confirmed ? As I said in my 1st post:

While under full load, I removed the front and top filter of the Meshify C (metal grid + foam): that gained me 6C on the CPU and 15C on the VRMs (!!).

That's why I was thinking about removing the front foam and/or having "stronger" fans (higher static pressure) at the front ...

Max

 

[Darkbreeze]

Obviously, everything you remove from the air path will improve performance. The less resistance, the better the airflow. No filters equals less resistance. No foam equals less resistance. And so on. But that might not BE the problem at all. I'd perform the tests with the side panel off and see what kind of results you get, and then go from there. Of course, you can always upgrade ANY of the fans, at any time, just because it is more like an investment in your system when you buy high quality fans and it usually results in a quieter one as well, but also for performance reasons.

I think one of the biggest things you can do is replace that 120mm top-rear fan that only moves 50cfm to a 140mm fan that moves almost 90. That should also help, to some degree or other, with the VRM temps as well.

 

[Owterspace]

Yes I removed all filters and foam. Meshify is a great case for air cooling, but you need strong fans. 60cfm is not strong. Look for something around 75cfm and up..

 

[Karadjgne]

You do realize what cfm is? Cubic feet per minute. A cubic foot is 1' x 1' x 1' of air. 50 cfm is enough airflow to replace half the air in a garage in 1 minute. Shoved through a box that's on average ony a total of about 1 cubic foot.

Roughly, a single input and single output fan can replace all the air in that case 50x in 1 minute, or a little under total replacement every second.

That's plenty of possible movement. Which is why you'll see seriously diminishing returns at anything over 4x 120mm, (2x in, 2x out) for 2x reasons. First being the case air doesn't get much opportunity to heat up and build up, the second being that low case air temp can't get any colder than the ambient air introduced, so has no further affect on heatsinks to reduce temps.

Air is one thing, air flow is another entirely. Replacing all the air in a case roughly every second or every other second, is plenty of flow, but it has to actually move in/out and not enter a circulatory pattern. The strongest area of vacuum is closest to the fan blades, by opening vents close to the fan, they become intakes for air. So the cpu cooler front fan is drawing air from that top, unrestricted hole. Consequently it's getting a cooler ambient air source and the cpu temps go down.

But just like having a hole in a straw, that means less air is used from the intakes, which allows buildup of heat in other areas as case air looses momentum. Your concern for cpu/VRM temps is admirable, but they are not everything. There's also Sata chipsets, Northbridge (pcie chipset), super i/o etc that are not temp reported, but just as vital to pc health. You sacrifice one at the expense of the other.

Which can affect stability, performance, boost levels of cpu/gpu, drive health and life etc. It's more important to maintain middling temps on everything, moving all the air in the case out, than aiming for performance out of 1 or 2 components. Better fans will increase airflow potential, but aimed wrong can end up doing exactly nothing.

View: https://m.youtube.com/watch?v=OupiPZybyBE


Perfect example. It's not so much the quality, you could put 3000rpm Industrial Noctua in that case, but simply the lack of Flow that destroys everything, the air goes nowhere.

 

[Darkbreeze]

That comment, would seem to have been directly intended as a reply to the post from Owterspace, except Owterspace is not the OP so it's kind of weird.

 

[Karadjgne]

Eh, somewhat. More of a clarification. But as long as it helps Op see the bigger picture than just a cpu temp, I'm good with it. A case full of Noctua is a wonderful thing, but ultimately useless if they are not doing the job the right way. Optimizing the airflow is priority, the better fans are icing on the cake.

 

[Owterspace]

Indeed. I was just saying more airflow is helpful. I run 120x38s in just about every hole of my case, works great. Not the quietest but its very good, if not excellent.. Gets a bit dusty but Datavacs are good. Anyways.. sorry to interrupt.

 

[hornetmax]

But just like having a hole in a straw, that means less air is used from the intakes, which allows buildup of heat in other areas as case air looses momentum. Your concern for cpu/VRM temps is admirable, but they are not everything. There's also Sata chipsets, Northbridge (pcie chipset), super i/o etc that are not temp reported, but just as vital to pc health. You sacrifice one at the expense of the other.

All other temps are already OK, that's why I didn't mention them.

Better fans will increase airflow potential, but aimed wrong can end up doing exactly nothing.

Can't disagree with that, but what are you implying ?
I have 2 front 140mm intake and 2 rear/top-rear 120mm exhaust: you mean I'm aiming them wrong ?

I know that it's to be expected to have lower temps if I remove any filter, what I can't explain is that if I remove the filters, the exhaust fans instantly drop some RPMs while front fans instantly gain some RPMs. Can't figure out why ...

Anyway, as the temps go down when I remove the filters, this means that the system could benefit from a bit more flow coming in. 3 ways to obtain that:

1. Get rid of the filters.
2. Keep the filters and have "stronger" intake fans
3. Keep the filters and have "stronger" exhaust fans

I was thinking more about 2 (as the AF140 have low static pressure and hence they may struggle with the front foam filter) but it's true that the top-rear could benefit from a 140mm (could help the VRMs). Maybe I'll just buy 2 Arctic P14 for the front (as they have slightly higher CFM but much higher static pressure than the AF1450s) and move one AF140 as top-rear (the filter there is just a mesh, no foam).

 

[Karadjgne]

Filters are a personal choice, for most they are a benefit, my pc has none, but doesn't need them.

Be careful with stats. They don't tell the full tale. One fan might have better cfm or sp than another, but that's only at maximum rpm. Says nothing for the space in between. Noctua is actually perfect for that example. The NF-F12 has quite high sp, the NF-S12 has high cfm. Both ends of the scale. The A12x25 doesn't have as high as either, but it's performance curve is convex whereas the S/F fans are concave. So if you look at the curves superimposed, at @ 1/3-2/3rds rpm, the A12x25 has much higher performance than both the S and F fans in both SP and cfm.

More air and stronger, right where the fans spend the majority of their rpm life-cycle, not at maximum, but in the middle ranges between idle speeds and average case temps. And that's a common phenomenon, often the multipurpose designed fans exceed the performance of dedicated fans everywhere except at the stated maximums.

The Arctics are very quiet, and very good for their price, if you run them flat out, it's when running them at a more average rate, they fall flat on their face, performance drops significantly.

Replacing fans isn't cheap, my 4x Noctua cost $70 (on sale), so for most ppl it's a 1 shot deal, that means looking for the right fan for the intended purpose, but also must take into account all the variables. Just looking at maximum stats can be misleading and entirely possible you'll lose performance vrs the stock fans and most ppl are clueless as to why that happened.

Using HWInfo (sensors only) as part of optimizing fan curves, seeing what happens to all its reported temps, at what speeds, vs your tolerance levels for noise, vs changes in fan or baffle placements etc. Some of us are lucky (like Darkbreeze) and have access to multiple fans of different types and designs, so we get to flip-flop fans and see differences they can make. So when someone like Darkbreeze recommends a certain course of action, like the Noctua he recommended, you can take it on faith that there's a very good reason for such. We've already done the homework for you.

But the rest is on you. Setting curves, sizes, placement etc. For the Meshify I prefer 3x 120mm in front, 1x 120mm rear, baffles in place, top sealed. That may not be your choice or an available option so you'll need to play around with things to get the best performance from your pc for your own needs.

 

[hornetmax]

OK, thaks a lot for all the answers.

3x120 front not doable: I don't want to take away the psu cover, so must stick with 2x140 front + 1x120 rear (140not possible) and 1x120/140 top-rear (140 a bit of a tight fit due to MB cables but it should be OK).

Noctua's: no doubt they are excellent, but price is a bit steep and some tests I've found around (see below) seems to indicate that the Arctic P14 seems good, at least when the price is factored in: I can have the A14 PWM Chroma for 25E but the Arctic P14 PWM is only 12E (10E without PWM).

Independently on the models I'll buy, current dilemma is if I should buy:

1. 2 new 140mm fans to be put in the front, one of my AF140 will go as top-rear replacing one of the X2 GP and the other will sleep in a drawer (rear fan will still be a X2 GP).
2. 1 new 140mm and 1 new 120mm, to replace respectively the top-rear and rear X2 GP (I'll keep the AF140 on the front, as right now).

My understanding is that you seem to suggest to go for option 2, but my fear is that if it's the frot that is choking, having better exhaust fans won't help much.

MaX.

 

[Darkbreeze]

Same argument we always hear, regarding Arctic. Yes, they have some ok fans for the price and if price is the biggest concern then they are (so far as we know so far since mostly these fan models haven't been around long enough to make any determinations about longevity) a decent enough choice.

Personally, if cost is the biggest hindrance, I'd much prefer to go with these than anything made by Arctic, and I'm not saying that because I think Arctic is crap, just that Noctua is superior and exemplary in just about every way a fan could be.

This is about half the cost of the other two fans I recommended. They aren't AS good as those other fans, and maybe not the preferred color scheme, but they are very good fans with good performance and quality.

PCPartPicker Part List

Case Fan: Noctua P14s redux-1500 PWM 78.69 CFM 140 mm Fan ($14.95 @ Amazon)
Case Fan: Noctua P12 redux-1700 PWM 70.75 CFM 120 mm Fan ($13.90 @ Amazon)
Total: $28.85
Prices include shipping, taxes, and discounts when available
Generated by PCPartPicker 2021-02-15 18:20 EST-0500

 

[Karadjgne]

Arctic MTBF (Mean Time Before Failure) is 50k hours. That's 6 years of constant use. Or closer to 10 year warranty under normal ish usage.

Noctua MTBF is 150k hours. Meaning 17.5 years of 24/7/365 use. Under normal pc use, you'll be past retirement and in an old folks home long before that fan is expected to quit. Warranty period of 5 years, they could realistically put it at 20 years and not be joking.

While that claim might be overkill, it's a good measure of exactly what it means when you get what you pay for. It means every time I get a new pc, if I had Arctic fans, they need replacing. And Arctic is not the only one, most companies have similar MTBF.

It means that even if I get lax on cleanings and kill half the expected lifespan, I'll still expect a good 10 years out of a Noctua, or 3 years from Arctic.

 

[hornetmax]

Same argument we always hear, regarding Arctic. Yes, they have some ok fans for the price and if price is the biggest concern then they are (so far as we know so far since mostly these fan models haven't been around long enough to make any determinations about longevity) a decent enough choice.

Personally, if cost is the biggest hindrance, I'd much prefer to go with these than anything made by Arctic, and I'm not saying that because I think Arctic is crap, just that Noctua is superior and exemplary in just about every way a fan could be.

This is about half the cost of the other two fans I recommended. They aren't AS good as those other fans, and maybe not the preferred color scheme, but they are very good fans with good performance and quality.

Thanks !
Given the small price difference with the Arcitc and some returns about motor noise at some RPMs and redux's acceptable color scheme, I'm OK with the cheapo Noctuas
What would you suggest between the two options:

1. I buy 2x NF-P14 redux 1500 PWM and put them as front intake, movig one of my Corsair AF140 to top-rear (replacing one of the X2 GP) and keeping my other X2 GP as rear.
2. I buy 1x NF-P14 redux 1500 PWM that will go top-rear and 1x NF-P12 redux 1700 PWM that will go rear (so replacing both my X2 GP) , keeping my Corsair AF140 as front intake.

 

[Karadjgne]

The Corsair are AF, that's airflow. As in high cfm. Those type of fans work far better as exhausts, there's no real impediment or need for static pressure at exhaust position, just a need to move air. The Noctua have equitable cfm but higher sp, so are much more suited for getting air To the gpu area and up under the cpu cooler.

Static pressure being the force of air, cfm being the amount of air. Not the same thing.

So I'd go with 2x Redux at intake and move the Corsair to the exhaust position. Option #1. You can always opt for a Corsair AF120 at the rear but that's more for symmetry or aesthetics appeal. I'd also adjust the fan curves so that the Noctua maintain at least a 1/3rd higher rpm for normal usage, especially if using the foam

 

[Darkbreeze]

Replace them all with cheapo Noctua fans. Get three of the 140mm P14s 1500rpm Redux fans and one of the P12 1700rpm Redux fans, and then sell your Corsair fans to make up for it. Or, if you want to stay with Corsair and simply improve in the area of airflow and static pressure, the Corsair Maglev fans are pretty good too, if you get the Non-RGB Maglev model. The RGB model, like most RGB fan models, has reduced specifications compared to the non-RGB model. Less CFM, lower static pressure, probably potentially higher noise level, etc.

Myself, I've given up looking at any other options really. Cheapo Noctua or else the A14 or F12 chromax.black.swap fans if it needs to be a premium fan. Even the "earth tones" models are fine with me really. For really over the top performance the iPPC industrial Noctua fans, but you have to know what you are getting into with those because they DO move a lot of air, they DO have insanely high static pressure ratings and they DO have a significantly higher noise level than the non-industrial NF-A14 1500rpm variety. But they'll cool the crap out of something too.

Honestly I don't recommend ever using those anymore unless it's in an application where you won't be sitting by it, like in a cabinet or server room. The regular Noctua (And other brands) type fans are more than enough for 99% of systems.

 

[hornetmax]

Thanks to both for the replies. Went with two P14 1500 redux: I'll slam these in (as in option 1 above) and see what I get, then ponder if more is needed !

 

[hornetmax]

Hi again, new fans in and ... well, I'd say no significant change in temps and now when under full load, the system makes more noise (guess that's due to the P14s going up to 1500rpm where my AF140 were only going up to 1150rpm).

Even with the P14s front, if I remove the front and top filters the CPU and VRMs temps go down significantly, roughly like with the AF140. Most of the gain comes with removing the front filter (mesh+foam, whil etop is mesh only).

So what's the conclusion here ? Aside taking the filters out (when needed, h265 encoding for me) and other more radical options (better CPU cooler, better MB), what's left ? Undervolting the 9900K and see if I can have better temps for the same stock settings ?

 

[Darkbreeze]

The conclusion is, nothing surprising. When you remove obstacles and obstructions, air moves easier. That should be easy enough to absorb. The more things you get out of the way, the better airflow and cooling are able to work. That is, and always will be, the case.

As far as the fans go, I told you already in the beginning that the exhaust fans were likely to be the place where gains could be had by adding or changing fans, not the front intakes. Exhaust = cooling performance. Intake = dust suppression. Except that, you NEED at least moderately decent intake airflow to resupply the inside of the case with cool air once the EXHAUST fans have removed some portion of hot air from the case. A vacuum is no better, and likely much worse, at cooling, than hot air is, so exchanging hot for cool and keeping it all moving, is how you reduce thermals or increase thermal efficiency.

There are very few occasions really where increasing intake pressure will equal better cooling performance. One is when there IS no intake, and you add some intake. Another is when you lack sufficient intake, and you increase your intake capacity, but at some point intake's value diminishes if you do not equally or more greatly in fact, increase the exhaust performance as well.

Keep in mind, the VAST majority of systems out there that have ever COME with cooling preinstalled, whether it's a case or a whole system, if they have only ONE fan installed, it is ALWAYS going to be the rear exhaust fan. That alone should show you where the priority lies when it comes to fan placement priorities and pressure offsets.

It still might not make a HUGE difference though. I'd probably get rid of the foam and leave the mesh if it were me. At least the mesh allows somewhat high airflow rates and will trap MOST of the incoming dust, while the foam will greatly reduce airflow and cooling performance.

I'm sure it's also worth mentioning, although I'd guess you have already as you're no dummy, to make darn sure that each fan is actually faced the correct direction. Front mounted fans should have the fan blades pointed towards the outside while rear and top fans should have their fan blades facing into the case. That way front is bringing air in and top or rear are taking it out. Simple to overlook sometimes.

Getting a case that will allow you to install 140mm fans in ALL locations, including the rear exhaust, like my Define S does (And most Fractal cases like the Define S2, Define 6, etc.), would be helpful as well. 140mm fans, especially in exhaust locations, can typically move a lot more air than an otherwise decent 120mm fan.

 

[hornetmax]

Yeah of course less obstacles is better, that was kind of clear even before

However putting "better" fans at the front and replacing the top-rear exhaust 120mm X2 GP with an AF140 didn't make any difference (at least not on the temps I was looking at, CPU and VRMs under heavy loads): that seems kind of counter-intuitive (to me), given that removing the front filter drops 10C -15C.

Maybe I'll get a PF12 for the rear and see if this helps. Or before that, moving a PF14 top-rear exhaust and the AF140 back to front. Or maybe I'll just pop the filters when I have to use handbrake, nice and easy.

Final remarks;
- I noticed that a lot of air intake happens in the top-front fan slot (not used at the moment).
- I can see that no 140mm rear option is a bit of a bummer but tbh I do like the fact the Meshify is pretty compact (as it's sitting next to my screen, on the desk).


P.S.
And yes, all fans are properly oriented thanks for asking. Checked the arrows on their frames and put some very thin paper floating in front of each fan, just to be sure

 

[Karadjgne]

Most ppl assume that the intake side of a fan acts the same as the exhaust. It doesn't. Just the opposite. The low pressure area created is just that, an area and it covers 180+° from the face of the fan upto 270° ish. So if you take a tower cooler and it's fan, the closest available source of unrestricted air, at a lower value than outside sir pressure, is the top. Especially the top-front.

Your intakes are putting air in. That air is going through gpu radiated heat, chipset radiated heat, ram radiated heat, drives, anything that creates heat in the front area of the case, before it gets to the cpu fan. Top vent isn't.

So if case air temps are 40°C, that's what's getting shoved through the cpu, hitting the VRM's etc.

If ambient outside air is 23°C, then that's what's getting shoved through the cpu cooler when the top vent is accessible.

Warmer air through a heatsink is less effective than cooler air. That's why cpu and VRM temps are dropping when you remove the top, your cooler is more effective.

At the cost of the cpu fan not picking up nearly as much case air, which allows case air to get even warmer, which then affects everything else but that which is affected by the cpu fan.

Cpu temps are but 1 part of the process, not the answer. Sacrificing case temps to drop cpu temps isn't the smartest move, but it's all too common because ppls priorities are assumed importance.

For best cpu temps put an intake fan on top near the cpu, but by that measure you blow any rising case heat back into the case, defeating the entire purpose of airflow.

Jump on dropping 10° off the cpu all you want to, but don't complain when the Sata chipset is pushing a toasty 95°C+, or the super i/o chipset is slowly roasting at over 100°C.

The air needs to flow. In then out. Just moving it around does nothing, especially under extended loads. Your total temp tests should be done for an hour or more at high loads, just to acclimate temps, a 5 minute test does nada.