Rear exhaust fan necessity

[Kasenumi]

Hi everyone

So I switched my old BeQuiet Dark Base 700 to new Fractal Design Torrent case, designed for maximum airflow.

Also I am switching a cheap 360mm AIO to Noctua NH D15.

My new case has 2 big 180mm fans on front (intake) and 3x 140mm fans on bottom (intake).

Do I need a rear exhaust fan if I have Noctua NH D15 CPU cooler, which already has 2 fans? 

Will it have any positive effect on the airflow, or will it influence the temperatures inside the case? 

[Levent]

Do you even have any exhaust fans? Whatever CPU cooler you got, you need exhaust fans.

[Jurrunio]

Problem is not the number of fans, but the airflow pattern. If there are fans pulling air in from all other directions, having a rear fan to blow them out will be very helpful.

[Kasenumi]

31 minutes ago, Levent said:

Do you even have any exhaust fans? Whatever CPU cooler you got, you need exhaust fans.

No, this case didn't include a rear exhaust fan

[chebsy]

Just use only the middle fan on the D15, and put the other one as the exhaust fan.  The front of the CPU tower gets bombarded with so much air in the Torrent you really don't need a fan there. 

[Daethz]

You would be better off putting the second dh-15 fan on the back of the case

Without it I imagine a lot of the hot air will miss the exit and end up cycling back into the case

[Jeppes]

It has a huge amount of positive pressure. You dont need exhaust fans there like you can see from reviews.

[XVB]

15 hours ago, Kasenumi said:

Hi everyone

So I switched my old BeQuiet Dark Base 700 to new Fractal Design Torrent case, designed for maximum airflow.

Also I am switching a cheap 360mm AIO to Noctua NH D15.

My new case has 2 big 180mm fans on front (intake) and 3x 140mm fans on bottom (intake).

Do I need a rear exhaust fan if I have Noctua NH D15 CPU cooler, which already has 2 fans? 

Will it have any positive effect on the airflow, or will it influence the temperatures inside the case? 

you need a fan that can push air faster than the fans on the D15 otherwise it will actualy impede cooling since airflow becomes slower

[thrasher_565]

16 hours ago, Kasenumi said:

Hi everyone

So I switched my old BeQuiet Dark Base 700 to new Fractal Design Torrent case, designed for maximum airflow.

Also I am switching a cheap 360mm AIO to Noctua NH D15.

My new case has 2 big 180mm fans on front (intake) and 3x 140mm fans on bottom (intake).

Do I need a rear exhaust fan if I have Noctua NH D15 CPU cooler, which already has 2 fans? 

Will it have any positive effect on the airflow, or will it influence the temperatures inside the case? 

in the torrent its not needed but any other case it is.

[Dogzilla07]

9 hours ago, XVB said:

you need a fan that can push air faster than the fans on the D15 otherwise it will actualy impede cooling since airflow becomes slower

This isn't exactly true. U can always slow D15 fans instead. Fans need to be in sync (slightly in favour of intake).. If u you push air from the front/bottom intake too fast into the CPU/GPU cooler (the front/bot fans are spinning way faster than Heatsink/GPU fans), your airflow becomes worse. If the intake fans are slower than CPU/GPU Cooler fans, the airflow will becomes worse.

 

What comes in, must come out (Exhaust is an area regardless if it has fans or not), PC case is a "closed" system, and the air inside it behaves like a fluid. (In general air is a fluid, in the case, outside the case, air is a fluid, it behaves like one, being in a closed system and having fans in place exaggerates the behaviour). It's more a kin to aquarium/hose/submarine that it looks at first glance.

 

If u don't want to change anything about the Fractal Torrent, your best bet is too play around with that extra front NH-D15 fan, see if u get lower temps with it as stock, or if u get lower temps with it strapped on the back of the heatsink instead pulling from the heatsink. Those are the only fan positions for the extra NH-D15 fan that make sense, don't put that fan as exhaust on the case itself, it's pointless in the Torrent.

 

And of course tune the case fans of the Torrent so that they aren't spinning way too much faster than the NH-D15 fans and vice versa. The funny thing is, physics is gonna help you save time here xD. Torrent fans are worse when noise-normalized than the NF-A15 on the Noctua, so the noise is gonna give you a nice upper limit to more easily sync up all your fans.

[td.]

11 hours ago, Dogzilla07 said:

Torrent fans are worse when noise-normalized than the NF-A15 on the Noctua

If you are talking about the front fans, I tried 3x NF-A14 on the front, the stock 180mm are a lot better.  Way more airflow at a much better noise pitch. 

[Dogzilla07]

16 hours ago, td. said:

If you are talking about the front fans, I tried 3x NF-A14 on the front, the stock 180mm are a lot better.  Way more airflow at a much better noise pitch. 

Was talking about all the fans together as a set, (those 3x140mm Fractal really bring the average down). Nice to know, Noctua noise pitch/weird noise is one of the worst compared to the competition. NF-A15 is barely passable (better than NF-A14). Only A12x25 has good overall noise characteristics.

 

I am a bit skeptical the dynamic GP-18 having better airflow when noise-normalized though, but i guess it's possible (it did not look like that from GamersNexus numbers, which is where I'm pulling from). Better noise characteristics yes definitely, but overall dBA vs temps not sure (Did you have the bottom 140mm fans running with both sets of front fans ?) They could have meshed wrongly with the 3 NF-A14, it's 6 fans, and they're 140mm, and that can really stack up in the noise department (especially weird frequency noise).

 

What RPM where the dynamic GP-18 running at ?

[td.]

5 hours ago, Dogzilla07 said:

Was talking about all the fans together as a set, (those 3x140mm Fractal really bring the average down).

I did try an all Noctua setup with 7 A14s, it was just super apparent from basic observations (like holding your hand behind the case and feeling the airflow) that 140mm is just not in the same league as the 180mm.  With subjective noise normalization, you notice some mild waffs out the back with 140s.  With the 180s, the blinds on the window behind my PC shake.  

 

Then as far as the 140mm Fractal fans, there are 2 main issues I have with them.  There is what I would describe as a PWM buzz, and they also don't support a 0% PWM signal.  The latter is pretty important to me, particularly in a bottom mount position, where I would prefer the fans to turn off if my GPU isn't doing anything. If they fix the buzz and add 0% support, I would have no reservations using them, because otherwise they seem pretty good.  And ultimately, any fan you put on the bottom of the Torrent is going to have blade pass frequency issues at a certain point due to the design of the bracket, so fan differences become largely meaningless aside from niche features like 0% support.

 

5 hours ago, Dogzilla07 said:

Noctua noise pitch/weird noise is one of the worst compared to the competition. NF-A15 is barely passable (better than NF-A14). Only A12x25 has good overall noise characteristics.

This is subjective, but the opposite is true for me.  Noctua seems to have a consistently pleasant noise pitch for any given flow rate, and I own quite a few different fans.  The A15 is also identical to an A14 except with a lower max rpm, so if you control for RPM then both will be the same.  And as much as I like the A12x25, it's still a 120mm fan, and all 120mm fans have a higher pitch hum at speed than a 140 would at an equivalent flow rate. 

 

6 hours ago, Dogzilla07 said:

I am a bit skeptical the dynamic GP-18 having better airflow when noise-normalized though, but i guess it's possible (it did not look like that from GamersNexus numbers, which is where I'm pulling from)

The problem with objective measurements like sound pressure is that it doesn't say much about the human experience  The best we can do is apply A-weighting, but even that can't tell you how subjectively annoying something is.  I can sleep soundly next to a 55dba box fan, but a 55dba PC case fan would drive me insane.  

 

So compared to something like my T30, the 180mm fractal fans blow it out of the water.  Even at 3000rpm, the T30 doesn't compete.  And while a sound meter may measure a GP-18 generating more sound pressure at 1200rpm than a T30 would at 3000rpm, I can tell you that the noise profile is far more pleasant with the GP-18.  And I think my observations would be outside the realm of subjectivity for most people.  

 

6 hours ago, Dogzilla07 said:

Better noise characteristics yes definitely, but overall dBA vs temps not sure

 

Noise characteristics with respect to airflow is all I care about, and even that can be a bit pointless at times, I can give you an example.  Starting with the CPU, there is no workload (aside from p95 small instructions) I can throw at it that exceeds the cooling capacity of my setup at minimum speeds.  My front intake fans running at their minimum (300 rpm) and my D15 fan running at its minimum (also 300rpm) can comfortably handle anything I throw at it.  And this is with an overclocked 10850k, not the coolest cucumber.  So talking about noise vs performance here becomes a bit academic when we are already well under our ambient noise floor, with our fans set as low as they can go.  The only "improvements" you could make would be entirely theoretical and would have no real-world benefit. 

 

Throwing a GPU into the mix at least makes things interesting, but introduces its own set of problems.  The GPU becomes your new noise floor when 3D workloads are introduced.  So the goal here is to find a setup that provides the most fresh air to the GPU, moves warm exhaust air out of the case as quickly as possible, and does so while staying at or below the GPU noise floor.  

 

Applying these requirements to the Torrent, you will find that the default configuration is ideal.  140mm fans on the bottom provide the GPU with the most air while staying under the noise floor, compared to 120mm and 180mm (I have tested this).  The 180mm fans really struggle on the bottom, likely because of the huge amount of bracket material overlapping the fan blades.  Moving on, 180mm fans on the front will provide the quickest exchange of air in and out of the case compared to 120 and 140mm, again staying at or below the GPU noise floor.  Any modifications to this configuration (aside from changing brands on the 140mm fans) will lead to some losses, and I would not recommend doing it.  But I do encourage you to do your own testing to validate my findings. 

[Dogzilla07]

Having 7 fans in is worse than having less, the full stock configuration of the fractal is also worse than having less fans, though different (from both a noise-normalized perspective and a general one, but especially for weird specific frequency noise). The two setups are different enough that you would change the airflow pattern in such a way to invalidate the holding hand behind test. 

 

Subjective noise-normalization is ok with experience, but not if you're very tolerant to high dBA "clean" air gushing noise, and not if you're sensitive to certain frequencies. you'd be normalizing for them, not the barebones dBA vs dBA. Which is ok, but again too subjective and prone to error (You can figure out which fan has the least motor/bearing/resonance specific frequency noise for you, which is useful, you can experience why filling a case full with fans is gonna make more noise for not enough performance, etc, ... There's a ton of useful stuff u can find out, and comparisons to make, like u already did). But u'd to finish with it/do it after using a device to baseline non-subjective noise-normalization and then use ur ears to fine-tune which fan/motor is best for you.

 

I'm confident any of the high quality LCP fans will not produce the frequency noise due to the bottom bracket (If the problem is what i think it is, too soft material making resonance noise with the material of the bracket, impellers flapping, motor not stabilized enough). If it's more than than, then, yeah u'd probably need standoffs.

 

As for nf-a15/a14, yes I agree completely subjective, every Noctua i've tried other than a15 (and i haven't a14, but was comparing with sounds from the net, so definitely could be wrong about the specific frequency noise), has been consistently unpleasant at the vital RPM ranges.

 

11 hours ago, td. said:

The problem with objective measurements like sound pressure is that it doesn't say much about the human experience 

Exactly, that's why i keep talking about weird frequency noise issues as separate from dBA comparison. Everybody is more sensitive (or less) to different things and it's quite subjective. But I think in your case, your experience applies to a very small % of people. If I'm correctly assuming you're fine with deafening amounts of noise as long as it's "clean" (one that is too much above the noise-floor of your living/work space) ?

 

Comparing T30 vs GP-18 how ?, at what speeds ?, outside of the case /inside of case ? instead of AF-14 ?, u've given to little details for this (for instance if it was outside of case and close to your ear, or in case and again very close to your ear, yes, 120mm will have the the higher pitch hum, but only cause you're really close, and the angle is wrong. A fan in the case would be a different angle and further apart, and there'd be metal between u and the fan. The specific hum from this would quickly disappear. You could always have very sensitive and good ears that it takes a longer distance that for an average person xD I'm assuming it was all in case. I'm assuming it was all in the case, and just fully stocked with different sets of fans.

 

LCP fans produce less resonance and generally punch above their weight (speed and size wise). and it's highly possible that the GP-18 would have better specific noise at high speeds, and even more so at close range, compared to even t30 or a12x25 at high speeds. But when the GP-18 is like 500-700RPM and T30/a12x25 is 1000-1300RPM, that's what I'm skeptical about.

11 hours ago, td. said:

introduces its own set of problems.  The GPU becomes your new noise floor when 3D workloads are introduced.  So the goal here is to find a setup that provides the most fresh air to the GPU, moves warm exhaust air out of the case as quickly as possible, and does so while staying at or below the GPU noise floor.  

Yeah, exactly, go down enough with fans, HDD is the problem, switch to SSD, GPU is the problem, lower the general noise of the GPU, coil whine becomes the problem. So yes, I completely agree with you, with a lack of luck, having pleasant sounding fans spinning faster and/or isolated case is better if a person can't get used to the coil whine.

 

To solve GPU heat/airflow problem, zip-tie-ing fans after de-shroud, removing back PCI-E slots, experimenting with less bottom intake fans to remove vortices. All of this helps. And produces more effect, than base-line changing fans around like you did. You've proving my point with your experience (that u find so little difference, just solidifies that simply throwing in a lot of fans, or keeping the setup the same as baseline is gonna be within the margin of error/tolerance) If i'm correct in assuming u only tested in your case by fully populating all/almost all mounting spots with different fans and combination of fans.

 

And again correct me I'm wrong (and i agree with you when not taking into account LCP fans). GP-18 yes their static pressure to noise can very well be very competitive (need to see more proper tests, hear some recordings of it), but it's highly possible (I'm only pulling them from GamersNexus, and that's not enough info about them)

 

It's the 140 non-LCP vs 120 LCP what I'm most skeptical about. If you could please write out how exactly Phanteks t30-120 were mounted and how u tested them compared to the others, that'd be very helpful.

 

I'm fully certain for your ears and what u tested, that what you say holds subjectively true.  But i'm not certain how transferable and by which amount that would be to someone else's setup / subjective experience. Don't get me wrong If you really like the subjective noise-profile of the GP-18s and they successfully drown out all the other weird noise in your PC, you're golden, you've hit the jackpot, everything else is irrelevant. And if you're happy with your temps double whammy, use and enjoy it. half the people on overlock forums wanted to break Phanteks MP fans and swore to SW3, the opposite, even though testing showed them all silent enough xD And that's from people who hate Noctua sound, and u had plenty who were more than happy with NF-A14.

 

Phanteks T30-120 is just the most perfect fan choice objectively speaking (and it's gonna hold true for the highest percentage of users). In very rare cases someone's setup with them is annoying them noise wise, and lowering speed, lowering fan number, doesn't solve it, yes, something else might be a better fit. (Or just similar enough like stock torrent, because to get more tangible differences, yes u need to do more tinkering with a specific case sometimes when not using stock setup).

[td.]

1 hour ago, Dogzilla07 said:

Having 7 fans in is worse than having less, the full stock configuration of the fractal is also worse than having less fans, though different (from both a noise-normalized perspective and a general one, but especially for weird specific frequency noise). The two setups are different enough that you would change the airflow pattern in such a way to invalidate the holding hand behind test. 

 

Sure, the hand behind the case is no scientific test, hopefully I didn't give off the impression that it was.  But at the same time, I don't need CFD analysis to tell me that an automotive wind tunnel moves more air than my leaf blower.  Once the difference is large enough, human intuition does a pretty good job of figuring it out.  And the difference is easily large enough in the Torrent to rule out any fluid dynamics quirks leading to unexpected performance anomalies.  The 180s just move a lot more air than anything else you could mount on the front of the case, in any way you'd want to measure, at any noise level.

 

1 hour ago, Dogzilla07 said:

Subjective noise-normalization is ok with experience, but not if you're very tolerant to high dBA "clean" air gushing noise, and not if you're sensitive to certain frequencies. you'd be normalizing for them, not the barebones dBA vs dBA. Which is ok, but again too subjective and prone to error (You can figure out which fan has the least motor/bearing/resonance specific frequency noise for you, which is useful, you can experience why filling a case full with fans is gonna make more noise for not enough performance, etc, ... There's a ton of useful stuff u can find out, and comparisons to make, like u already did). But u'd to finish with it/do it after using a device to baseline non-subjective noise-normalization and then use ur ears to fine-tune which fan/motor is best for you.

 

Well again, past some point of certainty, the results from human intuition and scientific testing start to converge.  Most people are capable of discerning which random noise is louder than the other if you asked them.  And if you reach a point where you need a db tester to figure out the difference, then that difference no longer really matters.  For that reason I'm confident in saying (without any actual measurements) that the 180s move more air while being quieter than anything else I tried - the difference was blatantly obvious, and not academic. 

 

1 hour ago, Dogzilla07 said:

I'm confident any of the high quality LCP fans will not produce the frequency noise due to the bottom bracket (If the problem is what i think it is, too soft material making resonance noise with the material of the bracket, impellers flapping, motor not stabilized enough). If it's more than than, then, yeah u'd probably need standoffs.

No it's definitely the bracket causing an annoying blade pass frequency.  I even pulled the bracket out of the case and tested the fans on it without any other obstructions.  The mesh filter, its support structure, and everything else is fine and doesn't cause any issue.

 

1 hour ago, Dogzilla07 said:

Exactly, that's why i keep talking about weird frequency noise issues as separate from dBA comparison. Everybody is more sensitive (or less) to different things and it's quite subjective. But I think in your case, your experience applies to a very small % of people. If I'm correctly assuming you're fine with deafening amounts of noise as long as it's "clean" (one that is too much above the noise-floor of your living/work space) ?

No absolutely not, the opposite is true.  In fact I will sacrifice some "meaningless" performance differences (like allowing my GPU to run at 80c instead of 60c and losing 30mhz or whatever it is) just so my system can be as quiet as possible. 

 

But you have to be realistic about it, my GPU produces 450w of heat, so stuff is going to need to make some noise to be able to handle that, even if you do a deshroud/120mm mod.  You no longer have the luxury to be under the ambient noise floor with the power consumption we see these days.  So since you're going to be making noise regardless, you need to find a noise profile that works for you. And something like that can't be measured objectively.  You may prefer the sound profile of a fan that runs at 45dba compared to a different one that runs at 43dba, but you will never know that unless you try. 

 

1 hour ago, Dogzilla07 said:

Comparing T30 vs GP-18 how ?, at what speeds ?, outside of the case /inside of case ? instead of AF-14 ?, u've given to little details for this (for instance if it was outside of case and close to your ear, or in case and again very close to your ear, yes, 120mm will have the the higher pitch hum, but only cause you're really close, and the angle is wrong. A fan in the case would be a different angle and further apart, and there'd be metal between u and the fan. The specific hum from this would quickly disappear. You could always have very sensitive and good ears that it takes a longer distance that for an average person xD I'm assuming it was all in case , correct if i'm I'm wrong.

 

LCP fans produce less resonance and generally punch above their weight (speed and size wise). and i don't doubt at all that the GP-18 would have better specific noise at close range, compared to even t30 or a12x25 at high speeds. But when the GP-18 is like 600-700RPM and T30/a12x25 is 1000-1300RPM, that's what I'm skeptical about.

Yea I tested everything inside the case, at the normal spot on my desk.  I sit maybe 1 meter from it, and it's behind my monitor.  If you are doubtful of the GP-18 vs the T30 I would just encourage you to buy one of them, you might like it so much that you make a case mod to fit it, assuming you don't have a Torrent already.

 

1 hour ago, Dogzilla07 said:

If i'm correct in assuming u only tested in your case by fully populating all/almost all mounting spots with different fans and combination of fans.

 

It's the 140 vs 120 where I'm skeptical. Again please write out how exactly Phanteks t30-120 were mounted and how u tested them compared to the others.

Everything was fully populated (except the exhaust) but I turned the bottom fans off when I was testing the front fans, and vice versa.  Also tested with every fan turned on.  The different fans I tested were all mounted in the normal spot with the bracket that came with the case. 

 

1 hour ago, Dogzilla07 said:

I'm fully certain for your ears and what u tested, that what you say holds subjectively true. But i'm not certain how transferable that would be to someone else's setup / subjective experience.

Yep exactly, all I can offer is my own experience, hopefully someone finds it helpful because there isn't a lot of discussion on this case aside from the fan hub issue.  I think the few other builds I've seen people just run everything stock, or put an AIO, and say they are happy with it.  Which is good, if the case is so great that people don't feel the need to talk about tweaking it, then that says enough.

[Dogzilla07]

you're very wrong about subjective vs academic. They don't converge to be interchangeable (u can use intuition to save time and guide yourself to a path, but u must always double check and explore that with logic and/or math). They can work in sync, especially if u're a master in your field. but neither can substitute the other, u need both.

 

Human intuition is notoriously unreliable, even for someone very experienced with academically testing this exact stuff, can still make mistakes or not be able to discern stuff accurately. U need a meticulous methodology to rule out the human element. I'll always trust GamersNexus, VSG results over your experience/results, my own experience/results etc, ... Without details what a person did to eliminate error and stardardize there's no objective comparison. It's not to say subjective experience from amateurs is not useful, but it can only tell you so much, and specific things.

 

And it's not just in the realm of theoretical or too small difference, it translates into practice as well. GamersNexus has proven that in a simple way to the mases without a shadow of a doubt. And before that tens of thousands of posts of discussion / test/ results on overclock net forum.

2 hours ago, td. said:

Yea I tested everything inside the case, at the normal spot on my desk.  I sit maybe 1 meter from it, and it's behind my monitor.  If you are doubtful of the GP-18 vs the T30 I would just encourage you to buy one of them,

Being behind a monitor makes u an outlier for noise already. Me buying them would mean nothing, even knowing most of the  mistakes which to avoid, without using a good methodology, and decent equipment, nothing u do, or I do, will enable us to accurately find out which one of them is better. We need VSG(techpowerup) or GamersNexus to test for that. 

 

2 hours ago, td. said:

Everything was fully populated (except the exhaust) but I turned the bottom fans off when I was testing the front fans, and vice versa.  Also tested with every fan turned on.  The different fans I tested were all mounted in the normal spot with the bracket that came with the case. 

Well that's a problem, that partially explains your experience. Fully populating a case in this kind of comparison just invalidates everything. Your fan setup/airflow wasn't optimized properly for the various types of fans u tried. You've also not mentioned how you made sure the room temperature was the same if u did, where your temp probe was, and which programs u used to measure temps. There's so many variable that you probably didn't control, the results you observed could stem from any number of different reasons, and not actually which fan is better.

2 hours ago, td. said:

Which is good, if the case is so great that people don't feel the need to talk about tweaking it, then that says enough.

 

But there are, there's a post right here on the forum, case is new, it's very good, of course there's gonna be less than usual. It got recalled. There was only 500 sold, there's delays in shipping and manufacturing, give it time. And tweaking isn't done just because the stock isn't good enough. It's good enough for the vast majority of people. Tweaking could be just to get more, or get more quiet. 

 

For 140x120 LCP there's no doubt, it's been proven before, tested, results are there. It's only the GP-18 which is the joker. And the reason I'm skeptical is because we've been down this road before in the past. All 200mm fans, all 230mm fans were proven to be inferior to a proper 120/140mm fan. Even though Silverstone 180mm push a lot of air, all 180mm fans so far have been proven to be worse than a proper 120/140mm when noise-normalized. There's a precedent. I'd very much like to be proven wrong in my skepticism, matter of fact I eagerly await the possibility of it. But the physics of fluid-dynamics is not in favour of GP-18 surpassing t30-120 / a12x25. And this is not taking into account the much better quality and longevity of T30-120 and a12x25 vs GP-18.

[td.]

7 hours ago, Dogzilla07 said:

Me buying them would mean nothing, even knowing most of the  mistakes which to avoid, without using a good methodology, and decent equipment, nothing u do, or I do, will enable us to accurately find out which one of them is better. We need VSG(techpowerup) or GamersNexus to test for that. 

 

Nothing wrong with wanting a second opinion of course.  I only suggested you pick one up, because once you do, a 23 minute long youtube video with 7 different graphs comparing them will start to look pretty silly when you see the massive difference. 

 

8 hours ago, Dogzilla07 said:

Well that's a problem, that partially explains your experience. Fully populating a case in this kind of comparison just invalidates everything. Your fan setup/airflow wasn't optimized properly for the various types of fans u tried. You've also not mentioned how you made sure the room temperature was the same if u did, where your temp probe was, and which programs u used to measure temps. There's so many variable that you probably didn't control, the results you observed could stem from any number of different reasons, and not actually which fan is better.

 

If you are implying that fully populating the case is bad, I'm not sure what you mean.  In my experience, spreading the workload out over as many fans as possible always seems to produce to the least amount of noise.  I don't think that's a controversial take either, so maybe you are referring to something else. 

 

Also, maybe this will trigger you a little bit, but I actually don't care about temperatures on a granular level at all.  I think obsessing over temperatures is a waste of time.  There is only 1 temperature that really matters, the throttle temperature.  On my CPU it's 100c, and on my GPU it's 83c.  My goal is to keep those components under the throttle temperature, while producing as little noise as possible.  If my GPU is running at 60c under full load, that is a bad thing, it means my system is making more noise than it needs to be.  And that's how I test and set up all the PCs I've built in recent memory.  Give everything just enough airflow to prevent throttling, and not an ounce more.  And the Torrent can move enough air to accomplish this even with a 450w GPU and a 300w CPU at an extremely low noise level, because it moves the most air in general at any given noise level.  There may be some niche situation in where this scaling fails, but for most people, they will certainly find my observations to be the case for them. 

 

8 hours ago, Dogzilla07 said:

For 140x120 LCP there's no doubt, it's been proven before, tested, results are there. It's only the GP-18 which is the joker. And the reason I'm skeptical is because we've been down this road before in the past. All 200mm fans, all 230mm fans were proven to be inferior to a proper 120/140mm fan. Even though Silverstone 180mm push a lot of air, all 180mm fans so far have been proven to be worse than a proper 120/140mm when noise-normalized. There's a precedent. I'd very much like to be proven wrong in my skepticism, matter of fact I eagerly await the possibility of it. But the physics of fluid-dynamics is not in favour of GP-18 surpassing t30-120 / a12x25. And this is not taking into account the much better quality and longevity of T30-120 and a12x25 vs GP-18.

 

You seem pretty interested in this stuff, you really should just buy a GP-18 and see for yourself.  It will make all of this scientific minutia seem pretty silly, I promise.  And as far as reliability, I don't think we can really make any comments about that when these fans have been on the market for 1 month.  If after a year the bearings wear out and they start to wobble then that would suck, but only time will tell.  

[Dogzilla07]

22 hours ago, td. said:

If you are implying that fully populating the case is bad, I'm not sure what you mean.  In my experience, spreading the workload out over as many fans as possible always seems to produce to the least amount of noise.  I don't think that's a controversial take either, so maybe you are referring to something else. 

Fully populating a case has the opposite effect, it doesn't boost airflow, it makes it worse, and just produces more noise than 2-4 fans. No air cooled setup in any case requires more than 2-4 fans (good static pressure), possible a few tweaks to the some cases, for the best airflow, and the best performance you'll gonna get from that case. This has been discussed to death on overclock forums. And GamersNexus standardized testing proves this. 2x140mm front, or 3x120mm front is all a pc needs. Most of the time no exhaust. And in specific cases for specific cases xD, maybe 1 bottom intake, and an exhaust and that's it.

 

The golden rule of thumb is, the least amount of fans for sufficient airflow for your components nets you the best noise profile. Having 6, 7, 8 fans just messes with  the airflow so much, and the noise profile as well. Spreading the workload over as many fans is an oxymoron, no matter the speed they amplify the collective noise profile, I've talked to, read more than 150 posts of people having the opposite experience as to to the number fans and noise. And that's not counting knowledgeable people explaining and showing why that is, nor IRL experiences.

 

22 hours ago, td. said:

Also, maybe this will trigger you a little bit, but I actually don't care about temperatures on a granular level at all.  I think obsessing over temperatures is a waste of time.  There is only 1 temperature that really matters, the throttle temperature.  On my CPU it's 100c, and on my GPU it's 83c.  My goal is to keep those components under the throttle temperature,

You're correct on ryzen processor's obsessing over temperatures is waste of time, they are fine at their max temps (though this wasn't the case for previous CPUs). I personally don't care that the temp is low, I care about max squeezed performance out of a part for an acceptable noise level, and slightly under throttle/best-operating temp. Either fixed OC, or PBO, or similar.

And yeah torrent is absolutely capable of being good enough (especially to a person undervolting, or similar). However not everyone has the same hearing as you, nor wants to lower performance as much, so having those mediocre 140mm GP fans on the bot is an issue.

22 hours ago, td. said:

ou seem pretty interested in this stuff, you really should just buy a GP-18 and see for yourself.  It will make all of this scientific minutia seem pretty silly, I promise.  And as far as reliability

Why would I Want to buy it before I see a respectable reviewer review it ? It's been half a decade+ since I've outgrown impulse buying and figured out how poor a subjective/intuitive experience is. I either have the tools capability and time to test them, I watch/read/learn from someone who does, or I don't buy.  Also how can you promise reliability when it's been out so little, and it has inferior bearings ? If it was sunon magnetic levitation, or Noctua SSO2, or Be Quiet! SW3 motor, there's a precedent, but these GP-18 need to prove themselves. 

 

Your experience and other subjective experiences are also very helpful, but mostly related to noise profile experience and similar. But they're just that, subjective, and on their own without more experiences from different people to cross-check, and a scientific method to produce results, they're not nearly enough, they're just one piece of the puzzle.

 

I don't have the patience, the time, the will, nor do i want to spend any more money, on products which aren't the best in the world (I want max fps within my needs, I want the max viable frequency on all components to do that, I want temps that don't damage those components, and I want it to be quiet enough to not annoy me).

 

And if my research/knowledge/experience in this quest, can help other people for their needs which are fall somewhere in between anything's good use and mine, why not help them ? Especially if it's something big like LCP (Liquid-Crystal Polymers), and their game-changing impact on normalized noise-performance and subjective noise tolerance.

[thrasher_565]

the only change i would do is take the bottom front fan and move it to the back done. even then its probably not needed.

 

only real reason to change the fans if you want rgb (but they do offer that too) or you mounting rads.

 

its not the best case for aios ya you can put it in the front but it wont be as good as other case were its in the top. and if you got 2 aios then it could get interesting...

 

the case is mostly an air flow focest case with an tower cooler. people have been asking for that and that's what it dose.

Edited October 13, 2021 by thrasher_565

[Dogzilla07]

11 minutes ago, thrasher_565 said:

the case is mostly an air flow focest case with an tower cooler. people have been asking for that and that's what it dose.

This, the case is incredible, I love, it. Fractal did an incredible job. I commend them on the new tooling, the risk-taking, the innovation everything. It's the first case to fully implement a wider exhaust area, with more holes, and a more optimized back fan mounting. It's the proof no Exhaust fan is needed at the back ( even though it technically it has an exhaust in the form of the PSU exhaust fan).

 

They're using 2 decades old knowledge that u don't need exhaust fans to have a sufficiently effective exhaust. It's just not been practical/economical to have the tooling for it. Too risky. Same like with meshfilters.

 

But just because that's the case, I'm not gonna overlook mediocre fans in the bottom, and let my emotions cloud my judgement. I'm also not gonna obsess over a 180x38mm (I really like the form factor), without adequate reviews of it. I've been burned before.

[thrasher_565]

1 minute ago, Dogzilla07 said:

This, the case is incredible, I love, it. It's the first case to fully implement a wider exhaust area, with more holes, and a more optimized back fan mounting. It's the proof no Exhaust fan is needed at the back ( even though it technically has an exhaust in the form of the PSU exhaust fan).

ya i was looking at case and there was an fractal case that also has mostly open in the back too. not as much but it was open. people wanted it close off because of dust but with 30xx gpus an open back, fans on bottom, and a back fan is needed. the mods people do like remove the pci slots and add a fan there to help the hot pocket the gpu crates. no idea if thats the reason for the open back or they just thought to do it. server have a mostly open back.

 

ya even slow moving fans in the fron should push the air out instead into the psu. as far as i can tell the psu needs 55 degree in the psu to use its fan.

 

one thing the case is not good at thow is sound if you got say hdd in it or coile wine.

[td.]

2 hours ago, Dogzilla07 said:

Fully populating a case has the opposite effect, it doesn't boost airflow, it makes it worse, and just produces more noise than 2-4 fans. No air cooled setup in any case requires more than 2-4 fans (good static pressure), possible a few tweaks to the some cases, for the best airflow, and the best performance you'll gonna get from that case.

If this is what you believe, then I'm going to have to leave the conversation there then.  Cheers. 

[Dogzilla07]

9 minutes ago, td. said:

If this is what you believe, then I'm going to have to leave the conversation there then.  Cheers. 

i didn't "believe it" for years, or think that it works like that, so it's less what I believe goes on physics wise, and more what's been proven to me by knowledgeable/experienced people, and what I've come to understand over the years after an an enormous amount of discussions, research, examples, tests, etc, ...

 

 

[thrasher_565]

8 minutes ago, Dogzilla07 said:

i didn't "believe it" for years, or think that it works like that, so it's less what I believe goes on physics wise, and more what's been proven to me by knowledgeable/experienced people, and what I've come to understand over the years after an an enormous amount of discussions, research, examples, tests, etc, ...

 

 

back in the day we use to put as much fans as posable to cool and it works but not as good as having an air flow. so were i read that each fan added adds about 3dbs to the total number. so people ask why not use 4x 40mm fans vs 1 80mm fan and  well the air flow about the same the nose is not not to metchen probably cost more, more cables and posably more voltage. in servers they dont care much about voltage or noise and its about putting a much in a small case so stacking fans was a thing and air flow from front to back was optimized . over the years people clame to make fans the make the airflow more state less of an angle then others fans and with 3d printing we can make may different fan booster that do just that. but we are talk exstream performance here. most people that want that will go wc any way.

 

even my antec 1200 case even thoe i had 4x 120m fans in the front 200mm on top 2x120mm on the back i even add an 120mm on the bottom. an exstream amount of air throw the case. then there the azza 1000 case that i got for $40 with 2x200, 2x 140mm in front 1x120mm in back aslo  alot of air.

 

 

[td.]

removed