A nice introduction to a more in depth investigation of real world fan performance. I hope to see many more articles like this one with comparisons between brands where possible. Great work.
Agreed, this was an interesting and informative read. Looking forward to some comparisons, especially high-end brands like Noctua, to see what value you're actually getting from the extra outlay.
Thank you for setting the record straight in regards to the LED models. At first look anyone could be fooled that these are the standard models with some led's strapped in.
With that in mind, is there a reason the LED versions are built differently? That is, the article mentions the difference in the fan blades, but I'm wondering why there is a difference.
I actually asked (well, kinda angrily ranted) at a Corsair product manager on IRC about that once. Basically, people buying LED fans don't care about high-end performance as much. There's no market for high-end LED fans.
I think it's dumb, but they probably have market research to back it up.
Corsair George here - this is pretty close to right, or at least it was at the time of the product dev 2-3 years ago. When we started doing fans in 2012 we had no data that proved that "high cost" LED fans sold at all. Basically, if it had LEDs in it, nobody paid a premium. At the time the best fans were Noctua or Scythe, neither of which used LEDs.
So when we decided to do performance LED fans, we focused on making them perform very well for their price. Note that the LED fans do not include the swappable color rings or the rubber corners or the hydraulic bearings - all cost saving measures because LED fans had to cost significantly less.
Nowadays, with a few more higher-cost LED options, we are somewhat rethinking that.
I think there's room for a truly high performance LED fan out there - if it doesn't look and feel cheap.
Also, it's been almost 4 years since we released the SP / AF performance fans - I think we're about due for something new.
It's nice to see a fan review again. But I think that the presentation is awful. You should add scale lines for the different cooler types instead of the color gradients. Then it would actually be useful. And the word impedance should be limited to electrical resistance. I am studying mechanical engineering and I haven't heard impedance once (in fluid dynamics). It is usually refered as resistance (or drag for bodies in free flow).
The word impedance is by no means limited to electrical resistance. Actually, electrical resistance and electrical impedance are different things to begin with.
I can only assume that you are a new student if you have never heard of mechanical impedance. You definitely need to read more about it.
Interesting. I have a BS in Mechanical Engineering from Rutgers and never heard the term impedance used to describe resistance to fluid flow. Not even in Fluid Mechanics where we spent about 5 weeks deriving the Navier-Stokes equations, which as you know completely describe all fluid behavior. Of course there isn't a closed form solution to these equations so complicated (real life) situations much be modeled via numerical computation.
I have a feeling mechanical impedance is an older term that is rarely used anymore.
I think a much more descriptive term for this article would be to use the term "head" to quantify and describe the resistance to airflow through various computer systems as it relates to cooling. ie maintaining a certain rate of airflow though a case with a filter would have greater resistance (head) than one without a filter and thus require a different pump curve. Sizing pumps, which is what we are really talking about here is all about matching pump curves.
But I think what most of us really want to know here is how much air a certain fan can move in a high head (resistance) situation and how much noise is makes, as well as in relatively low head situations. I think perhaps testing through a case with no filter would be a low head situation and though a filter would be a high head case. Perhaps testing though a relatively constrictive radiator would also be useful unless results were similar to the filter case. But once tested we'd know if both tests would be useful in the future.
Anyway, how much air flow (cooling)? How quiet? How long will it last? That's the bottom line.
Also I think most readers would better understand the term "head" for fluid resistance because it can be quantified as the height of a certain air column. ie pushing air case "x" with a filter is akin to pumping air an air column of "y" height, whereas removing the filter reduces the head (air column) height by z.
Well, I can only assume that it depends on the school or the professor. Mechanical impedance is a very widely used term. You could run a Google Scholar search and you'd find thousands of papers published only this year.
I think that I mentioned that already, but resistance and impedance are different figures. Resistance includes only Real parts. Impedance accounts for the Reactive parts as well.
Head is not resistance. Head is a measure of performance that is based on pressure. Practically, it can be converted from the left axis directly. Head = Pressure/404.3 for air, where head is in ft and pressure is in mmH20. It should generate a similar graph with a different Y axis.
Head is a measure of performance based on pressure and performance is exactly what you are trying to determine here. That's why I said head may be a better parameter rather than impedance. If you've worked in industry you've probably noticed no one (at least on the US East coast when it comes to pipelines or flow) talks about impedance but rather refer to "head loss." It's a working term, not a school/theory term.
I too have mechanical engineering degree and have never seen the term impedance used like how you use it in the fan curve graphs. Whenever I hear impedance used in a fluids context, it's normally electrical people trying to explain the resistance of a flow due to pressure. In the text books I've studied, what you describe as impedance in the curves is what would have normally described as the system curve or the system head or something along those lines.
On a side note I find it pretty ridiculous that you guys have to actually do testing in order to get the fan curves instead of manufacturers providing a datasheet that has the fan curves on it (like they should be doing). I guess that shows the state of PC industry. I appreciate you guys doing real fan testing though and providing this data.
I've heard of impedance being used in fluid dynamics and, like the used mmH2O instead of Pascal, it's an older term used mainly by the 'old guard'. I would however recommend Anandtech to next time use Pa. as a term as you'll not find high tech fan-curve sheets using mmH2O (anymore).
On a side-note, if fan reviews are going to be used.. I would like to suggest a push-pull efficiency review on liquid coolers.
In the industry we almost always use pull configurations through heat exchangers as underpressure draws more evenly through an exchanger than otherwise possible with the momentum of the air. I don't have enough liquid coolers systems to be able to make a review, but my own pc is a degree lower. Considering some people might like to know these kind of things and I enjoyed to read something this close to my actual work, who knows.
Ohw, and just for bs argumentation and to be part of this thread: MSc.
I want to echo [Freaky Angelus]'s comments; mmH2O (millimeter of water) is a dated unit of measurement that isn't the preferred unit in any context as far as I know. It was essentially a metrication variation of inches of Mercury (inHg) that should be left to the history books. Pascals (Pa) would be the unit used for scientific literature in US and abroad. I'm not familiar with US engineering / commercial preferences for fluid dynamics and air flow/fan measurement to be sure, but I suspect they would prefer Pascals over mmH2O as well.
I haven't even thought of getting any kind of degree on anything, yet I still use word impedance to describe the flow resistance of helium mixture flowing inside mechanical parts and I do it daily at work. In physics it is quite normal thing to use word impedance, because you can't just say flow resistance since there are so many things affecting the flow resistance and as you can see it's much longer term than impedance.
Probably it would be fine to use the term "head", but I think that is largely used in technical papers and not in review texts meant for common people and I think that the graphs with impedance are nice and clear.
And E. Fylladitakis is not being rude at all, he is just explaining how things are done in this review.
I don't come to Anandtech for this passive aggressive condescending horseshit.
Before today, I was able to say that I've never seen an Anandtech writer snap at a commenter regardless of how coarse they were. You just broke that streak for me. I'm pretty disappointed and a little upset.
Let's be 100% clear in that the content of your post can be completely correct (it seems to be) and the post is still unacceptably rude.
I come to Anandtech to learn new things and I absolutely love to see its writers present new ideas. I don't want that to ever change. However, I don't come to Anandtech to see its writers be appear rude to readers even if those readers are rude "first".
It's really hard to accomplish that because it's so easy to be accidentally appear ruder than you intended - that's just the nature of the limitations of written communication. You really have to go overboard to make sure that your comments can't be misinterpreted.
I'm just concerned because Anandtech has historically been so awesome at that and I don't want that to change.
I am sorry if you, or anyone else, is offended in any way. I certainly did not mean to offend anyone. It just appear to me as common sense that someone who is studying mechanical engineering would not have heard of the term mechanical impedance, except if he/she was a first or second year student. That is why I took my time to explain the difference anyway.
Holy crap! Your post wasn't rude at all!!! It was informative and your informed opinion. That's what we're supposed to do here isn't it. Are so we politically correct that we can't even politely disagree. And your post was TOTALLY polite. As you can see from my post I also have a different opinion, but it's just that an opinion. There was no mocking or personal attacks here until the person who was so "offended" by your response posted.
Please, please, please, don't apologize when there is no need. We need to have just a little free speech don't we? Can't we discuss and...yes, God forbid sometimes disagree, and in the end learning something? I learned that fluid resistance is also impedance. That's cool and that's why I'm here.
Let's be active and engaged, let's argue sometimes, let's laugh sometimes, but please let's not be afraid to freely honestly express ideas and opinions!!!
Where was the author rude in his reply? Direct maybe, rude no!
And as far as I can see the author is from europe - maybe both of them misunderstood each other completly when it comes to the term for resistance/ impedance. I get the feeling that american and european schools/universities might be using different terms and methods to describe the same phenomenon, that wouldn't be first time (X-Rays vs "Röntgenstrahlung").
And in general europeans are a little bit more direct and you re lucky that the author isnt from germany, we re often even more direct ;-)
As a reader for many many years, I'm admittedly a mega-fanboy of Anandtech and I've previously paraded them around as being nigh-perfect compared to other tech blogs in many areas - including their writers' attitudes with readers (which is often lacking as technical-minded people can often be).
It really comes down to no longer being able to say almost-absurd comments like, "I've never seen a comment even remotely rude/insulting from an Anandtech writer" in good conscience. I like to think that if I'm going to act like a shameless fanboy around others, I have to be more than justified in doing so. I just can't do that anymore.
I fully realize that I probably put Anandtech on an unrealistically high pedestal. They are good, but the site is ultimately run by human beings and that's probably ok.
If you want to get offended on someone else's behalf, Tumblr is thataway --->
Although in all seriousness, I think you must have read that comment from an *entirely* different perspective to me; I thought it was two people familiar with the science discussing their different interpretations of terminology in a fairly chilled out way. I mean, if you think someone has got something completely wrong and are trying to explain it the way you understand it, it's hard to *not* come across as passive aggressive, but I'm not getting that vibe at all.
According to the Wikipedia article you link, "mechanical impedance" is the ratio of force to velocity in response to a harmonic force. It's analogous to electrical impedance. It follows that your use of the term "impedance" is incorrect for two reasons. First, you are measuring a constant airflow, not one that alternates directions. For "impedance" to be relevant, you would need a force that alternates directions, which would generate an alternating air flow. Second, air resistance is not proportional to air velocity, so there is no fixed ratio between the two. To talk about "impedance," you need a linear system.
Your second link does show that your usage is not unique. But your usage doesn't seem to be correct.
Salsa, obviously. Or hot sauce. Or any other number of sauces unrelated to that horrid sugary concoction known as ketchup that should be avoided at all costs. ;)
That's why my go fast stripes are black. And led lights? pfffff, all the extra wiring, boards etc add weight. Weight is the enemy of speed. No LEDS for me...
Great review of fans. I love to see performance curces rather than the regular two point tests that are more conventient to do. It would be great to see sound dB info too if you get the equipment in the future.
Also, I agree that the graphs need to look a little more consistent with the Anandtech style.
A suggestion for when you've got data on more fans. Since the scales on the graphs aren't constant; it would be really helpful for comparison if Bench was able to take the datapoints used to generate the performance curves and use them to create a single combined chart for the user selected set of fans/speeds.
A max and min speed chart also needs to be in there as user selected fan speeds ignore the fact that some fans can spin quicker/slower than others are capable of spinning.
My Corsair 380T came with the LED version of the fans and I was not happy with them at all. I ended up changing them out for the Enermax "Batwing" LED fans (which have removable blades for cleaning) which are much quieter but also have better airflow.
Very well written write up with lots of good information.
Curious, how did you go about increasing/decreasing the air resistance levels for the fans? As an mechanical engineer, I've been reading up on fan curves for work and I believe an adjustable duct type thing is common. I've always been curious to how this part is done.
Also, for anyone interested, the AMCA 210 Standard has lots of good info as well. A quick google search will bring you to the .pdf.
Hey, nice to see a more in depth review of the matter. Just two suggestions:
First, why are you plotting pressure over volume? Iwould have done it the other way around.
Second, could you measure the impedance (ballpark figures) of some typical obstructions and post photos of them? For example finger guards, one or two tower coolers, typical filter (ideally clean and dirty, would be interessting to see the effect). Just to get a feel for the dimensions involved.
I think a necessary corollary of AT doing such a solid job of reviewing and explaining fan performance, is that they will also (have to) start measuring and publishing flow curves of the things they review that fans get hooked up to. I was mainly thinking of cases and coolers, but you are right, filters and guards qualify as well. Maybe someday that kind of data will be part of the specs that manufacturers publish.
@simonpschmitt: "First, why are you plotting pressure over volume? Iwould have done it the other way around."
Typically the controlled variable is X and the measured variable is Y. They are controlling pressure and measuring airflow so typically it would be represented as you state. That said, the data is all there and the author is technically free and correct to use R and W for variables if he chooses as long as the axis' are labeled. Though in the end, I'd just as soon have the author use the typical approach to avoid confusion.
@simonpschmitt: "Second, could you measure the impedance (ballpark figures) of some typical obstructions and post photos of them? For example finger guards, one or two tower coolers, typical filter (ideally clean and dirty, would be interessting to see the effect). Just to get a feel for the dimensions involved."
This. Very this. Also, since the author mentioned "dense radiators", measurements and photos of dense vs sparse radiators.
"one Pascal is the pressure required to accelerate 1 kg of mass at a speed of 1 meters per second squared"
Looks like you mixed up the definitions of a Newton and a Pascal. One Pascal is the pressure required to exert the force of one Newton per square meter. The square meter disappeared in your definition. I guess you could say "one Pascal is the pressure required to exert on a surface of one square meter the force required to accelerate 1 kg of mass at a speed of 1 meters per second squared".
Sorry, that was my bad. I edited a few things and got mixed up. The definition now reads 'one Pascal is the pressure required to accelerate 1 kg of mass at a speed of 1 meters per second squared per square meter'. Admittedly it's a bit clunky, but it's the best way (from my perspective) to visualize it.
It kind of says that 1 pascal is still the pressure required to accelerate 1 kg of mass over an area, i would just keep it at 1 pascal is equal to 1 newton per square meter. the gravity acting on 1 kg is 9,81 newton.
If you're starting on fans, check out the old X-Bit Labs site. They had very thorough reviews which helped a lot in balancing noise vs cooling performance for different applications. Here's a link to that section:
It might help if there was a way to summarize the results in a more condensed manner. I don't have an ideal solution, but perhaps overlay more results in a single graph.
It would help if the voltage was specified in the graphs, and not only the RPM.
Some basic details are missing on the fans: startup voltage, "cruise" voltage, official specs (would be interesting to see even if somewhat arbitrary).
It would be nice to see more detailed noise-vs-RPM/voltage figures.
BTW, parts of some of the graph images are blurry. Anandtech's become much less consistent in the visual looks of graphs, and with occasional hiccups in quality. Also the frequent low quality JPEGs in the homepage make it less polished than it used to be.
Great technical analysis as always on this site. I'd just recommend one addition, a real world test. It would be very interesting to take a standardized test bench, let's say an overclocked i7, and have a 240mm AIO solution and an air cooler. It would be great to be able to see the actual effect on the temps by just changing fans, and would give a great data point for user value analysis. Being able to see how a fan statistically performs is great, but the real world effect vs. cost is what I would LOVE to see.
Whilst i love all the technical stuff all the graphs don't really help me decide which fans i should be buying for my case and radiator. What I need is some specifics such as fan a will need to run at x000 rpm to cool a 200w processor to 60C and produce xdb at that point. That would be a repeatable real world test that had some meaning. For case fans you need some form of test for rpm and db for each fan to provide a specific airflow that would be need for something like the above load plus a Graphics card.
A separate article on the differences a case design can make to noise and fan performance would be good, especially the difference a grill would have over a perforated sheet metal.
One thing that most people don't take into account are the frequencies emitted from fans. I know it's a lot to ask but it could be nice to see the spectrum for fans that are tested. Typically, more decibels in lower frequencies is more pleasant, and less likely to aggravate tinnitus, than more in shriller ones.
The graphs are a bit confusing, especially with the colours. And did there have to be so many? Impedance in this case is not the correct term, because that applies to the impedance it has to changing air flow, in this case it is resistance. (like in electricity, resistance vs impedance, where impedance is used for AC)
I would really rather use Pascal instead of mmH2O and m3/h instead of cfm but i guess that is your choice, having 1 imperial and 1 metric measurement is a bit strange though. 1mmH2O or 9,81 pascal is equal to about 1 meter of air column head. The left legend also should say static pressure instead of just pressure, because the pressure is a compination of static and velocity.
The graphs are called Pressure/flow, using volume is the incorrect term, as it is not talking about the volume but the actual displacement of air.
I was thinking of making a test setup myself for testing fans and their pressure/flow curves. How did you setup your apperatus to restrict air flow and test at different static pressures?
It would be really nice to have a simple module on the website, that puts the graphs of fans and graphs of different components like radiators and cpu coolers over each other to see what the best performance would be.
All in all, an excellent article with very informative information. I am also not use to seeing performance/fan curves presented this way but I am able to understand what the author was trying to convey.
Typically in the US, airflow is measured by CFM (Cubic Feet Minute) and In H2O (Inches Of Water Column/Gauge) or M/S (Cubic Meters per Second) and Pascals.
I do agree mixing imperial and SI units is not typical, but there are plenty of applications in the real world that this is done.
I do believe the author tried to perform a controlled test based upon his other articles and methodology. Without knowing the test setup and techniques used, there is always questions that could potentially affect the results. I for one would not have used a pitot tube, most likely a thermal anemometer. A very accurate test setup for airflow is a lot of money for a tech review site that would most likely never recoup the cost.
End result was a great amount of time by the author testing multiple units with many potential variables and presenting the data in a way that they felt was appropriate. For that I am greatly appreciative of, it was an excellently presented piece taking their perspective into account.
I have no college education, although I do have almost 10 years of Biotech and Semi Conductor Cleanroom Certification and Balancing.
I also learned something a long time ago about engineers summed up to a saying I saw "Arguing with an engineer is like wrestling a pig, after awhile your realize they enjoy it"
He actually did use a anemometer for the airflow test i think. The pitot tube is just used for static pressure readings. I'd say the measuring setup is pretty accurate for what he is doing here. All in all its a great way to measure actual fan performance instead of going after just a flow number and a static pressure number.
From my work as an engineer i'm more used to seeing actual graphs of pressure/flow and power to make a good selection. Papst for example does supply them. The only computer fan i've seen else that show these are ek waterblocks
Thank you, I meant to write vane anemometer instead of a pitot tube. As long a the same equipment is to be used these fan comparisons are effective.
Just some points to make about the equipment used. Not to say the article is not accurate, but the equipment used is "inexpensive" by trade standards. Such as some of my equipment that I have for those once in a while measurements. The temp/RH, sound, and vibration equipment I own is "cheap" compared to people who specialize in it the field. I wouldn't expect Anand to be using $2,000+ USD thermal anemometers (TSI 9565P temp and barometric compensation) and $2,000+ USD (Shortridge ADM-880C) digital manometers.
The Extech AN200 description does say a thermal anemometer, but the picture shown is a vane anemometer with a FPM accuracy of 3% reading + 40 FPM.
Assuming a 140mm diameter @ 60 CFM = approximately 600 FPM velocity. The AN200 unit is +/- 10% at this reading.
The HD350 accuracy is +/- 0.3% of full scale @ 510 mm h2o = +/- 1.5 mm h2o. Most the readings taken are within the tolerance of the meters accuracy.
I do enjoy reading the author's articles. I have never posted before to Anand, I just hope a few people will hopefully understand before attacking/leaving harsh criticism of the article.
Great article, super technical as always, and as usual WAY WAY over my head ha.
What I think would be useful is a real world scenario that dumdums like me could look at and see the benefits of. Say, an overclocked CPU with a certain cooler, and the temperature taken with the normal fan, then taken with the custom fan, so we can see the benefit we care about - how much cooler it makes something.
If this isn't possible, due to the fans not fitting, maybe another scenario, a system instead of a CPU cooler, with it's standard fans, then these fans.
I know this probably won't be a popular idea as AnandTech is more for the tech savvy crowd, I just feel like some real world results would help
This article is mostly about how much flow the fans actually create in certain situations. It can be coupled with how much cooling you have at a certain flow, altough that really depends on the cooler/aio
Noctua fans, always and forever. The best combination of static pressure and airflow for a given noise level. On top of that, they last close to forever. Of course, one has to say that they are from beautiful, but that's minor.
Sure they do. Their industrial line is black/dark brown. I have 6 of them on my system, on a push-pull 3x120mm radiator. Nothing short of superb. They come at price, though, and not because of their color, but because of their extra specs. Anyway, as I said before, color is minor.
Too bad I couldn't find the non-led versions around here in time and budget; I'm sick and tired of that blue glow emanating from a case when you need some sleep. I got a case with thick foam filters (Tt Armor), so SP fans for me all the way (3rd rebuild in this case). Fewer, thicker blades is what I am looking for. No care for loud fans, since the graphic card makes the point moot.
Yes, I got a similar case, and was given significant spousal input that the extra light had to go. A little bit of time with a wire cutter extinguished the LEDs, but am still left with the inferior fans.
I thought the same thing about the wire cutters, but I'd need to take the whole thing apart and search the proper wire to cut on the led ones I got. Not gonna happen, can't be ars... bothered.
Very good article, but performance/noise charts and analysis are missing. I.e. in this case I'm not even sure that SP perf and quiet versions have same noise at let's say 1200rpm.
Cool stuff, guys. I appreciate the detail, and I appreciate even more that you didn't couple it to a self-deprecating "if your eyes haven't glazed over yet" type comment, all too common in the popular press when the going gets technical. AT continues to differentiate itself, *and* increase its credibility.
what Jeffery said I am impressed that some one able to make $8960 in one month on the computer . you could try this out............................... earni8
I hate how you change the scale for every single chart. Why leave so much blank space in the CFM scale for some charts but not others? You make the scale all the way up to 120CFM but the max a fan got in the chart was 70 CFM. You could have lowered all the charts to 80CFM scale and it would have been easier to read and compare. I thought they teach these things in math classes in high school.
We’ve updated our terms. By continuing to use the site and/or by logging into your account, you agree to the Site’s updated Terms of Use and Privacy Policy.
Facebook
Twitter
RSS
92 Comments
Back to Article
RdVi - Wednesday, November 25, 2015 - link
A nice introduction to a more in depth investigation of real world fan performance. I hope to see many more articles like this one with comparisons between brands where possible. Great work.Valantar - Thursday, November 26, 2015 - link
Agreed, this was an interesting and informative read. Looking forward to some comparisons, especially high-end brands like Noctua, to see what value you're actually getting from the extra outlay.maximumGPU - Wednesday, November 25, 2015 - link
Thank you for setting the record straight in regards to the LED models. At first look anyone could be fooled that these are the standard models with some led's strapped in.inighthawki - Wednesday, November 25, 2015 - link
With that in mind, is there a reason the LED versions are built differently? That is, the article mentions the difference in the fan blades, but I'm wondering why there is a difference.ninjaquick - Wednesday, November 25, 2015 - link
Ask Corsair.tipoo - Wednesday, November 25, 2015 - link
Just a misanthropic guess, but saving pennies to make up for the LED?TheKiwi - Thursday, November 26, 2015 - link
I actually asked (well, kinda angrily ranted) at a Corsair product manager on IRC about that once. Basically, people buying LED fans don't care about high-end performance as much. There's no market for high-end LED fans.I think it's dumb, but they probably have market research to back it up.
Barbarossa - Thursday, December 3, 2015 - link
Corsair George here - this is pretty close to right, or at least it was at the time of the product dev 2-3 years ago. When we started doing fans in 2012 we had no data that proved that "high cost" LED fans sold at all. Basically, if it had LEDs in it, nobody paid a premium. At the time the best fans were Noctua or Scythe, neither of which used LEDs.So when we decided to do performance LED fans, we focused on making them perform very well for their price. Note that the LED fans do not include the swappable color rings or the rubber corners or the hydraulic bearings - all cost saving measures because LED fans had to cost significantly less.
Nowadays, with a few more higher-cost LED options, we are somewhat rethinking that.
I think there's room for a truly high performance LED fan out there - if it doesn't look and feel cheap.
Also, it's been almost 4 years since we released the SP / AF performance fans - I think we're about due for something new.
Phreedom1 - Sunday, December 6, 2015 - link
Awesome explanation George. Hope Corsair does come out with a new fan product.rscsrAT - Wednesday, November 25, 2015 - link
It's nice to see a fan review again.But I think that the presentation is awful. You should add scale lines for the different cooler types instead of the color gradients. Then it would actually be useful.
And the word impedance should be limited to electrical resistance. I am studying mechanical engineering and I haven't heard impedance once (in fluid dynamics). It is usually refered as resistance (or drag for bodies in free flow).
E.Fyll - Wednesday, November 25, 2015 - link
The word impedance is by no means limited to electrical resistance. Actually, electrical resistance and electrical impedance are different things to begin with.I can only assume that you are a new student if you have never heard of mechanical impedance. You definitely need to read more about it.
https://en.wikipedia.org/wiki/Mechanical_impedance
The resistance of a system to airflow is called impedance. Drag is the force that opposes airflow, generated by this impedance.
https://www.google.gr/url?sa=t&rct=j&q=&am...
Hulk - Wednesday, November 25, 2015 - link
Interesting. I have a BS in Mechanical Engineering from Rutgers and never heard the term impedance used to describe resistance to fluid flow. Not even in Fluid Mechanics where we spent about 5 weeks deriving the Navier-Stokes equations, which as you know completely describe all fluid behavior. Of course there isn't a closed form solution to these equations so complicated (real life) situations much be modeled via numerical computation.I have a feeling mechanical impedance is an older term that is rarely used anymore.
I think a much more descriptive term for this article would be to use the term "head" to quantify and describe the resistance to airflow through various computer systems as it relates to cooling. ie maintaining a certain rate of airflow though a case with a filter would have greater resistance (head) than one without a filter and thus require a different pump curve. Sizing pumps, which is what we are really talking about here is all about matching pump curves.
http://www.pricepump.com/pumpschool/psles2.html
But I think what most of us really want to know here is how much air a certain fan can move in a high head (resistance) situation and how much noise is makes, as well as in relatively low head situations. I think perhaps testing through a case with no filter would be a low head situation and though a filter would be a high head case. Perhaps testing though a relatively constrictive radiator would also be useful unless results were similar to the filter case. But once tested we'd know if both tests would be useful in the future.
Anyway, how much air flow (cooling)? How quiet? How long will it last? That's the bottom line.
Hulk - Wednesday, November 25, 2015 - link
Also I think most readers would better understand the term "head" for fluid resistance because it can be quantified as the height of a certain air column. ie pushing air case "x" with a filter is akin to pumping air an air column of "y" height, whereas removing the filter reduces the head (air column) height by z.E.Fyll - Wednesday, November 25, 2015 - link
Well, I can only assume that it depends on the school or the professor. Mechanical impedance is a very widely used term. You could run a Google Scholar search and you'd find thousands of papers published only this year.I think that I mentioned that already, but resistance and impedance are different figures. Resistance includes only Real parts. Impedance accounts for the Reactive parts as well.
Head is not resistance. Head is a measure of performance that is based on pressure. Practically, it can be converted from the left axis directly. Head = Pressure/404.3 for air, where head is in ft and pressure is in mmH20. It should generate a similar graph with a different Y axis.
Hulk - Wednesday, November 25, 2015 - link
Head is a measure of performance based on pressure and performance is exactly what you are trying to determine here. That's why I said head may be a better parameter rather than impedance. If you've worked in industry you've probably noticed no one (at least on the US East coast when it comes to pipelines or flow) talks about impedance but rather refer to "head loss." It's a working term, not a school/theory term.Oxford Guy - Thursday, November 26, 2015 - link
"It's a working term, not a school/theory term." That doesn't make it better. It just makes it common.Cold Fussion - Thursday, November 26, 2015 - link
I too have mechanical engineering degree and have never seen the term impedance used like how you use it in the fan curve graphs. Whenever I hear impedance used in a fluids context, it's normally electrical people trying to explain the resistance of a flow due to pressure. In the text books I've studied, what you describe as impedance in the curves is what would have normally described as the system curve or the system head or something along those lines.On a side note I find it pretty ridiculous that you guys have to actually do testing in order to get the fan curves instead of manufacturers providing a datasheet that has the fan curves on it (like they should be doing). I guess that shows the state of PC industry. I appreciate you guys doing real fan testing though and providing this data.
Erazor51 - Thursday, November 26, 2015 - link
E=mc2Freaky_Angelus - Friday, November 27, 2015 - link
LoL, let's throw titles as arguments...I've heard of impedance being used in fluid dynamics and, like the used mmH2O instead of Pascal, it's an older term used mainly by the 'old guard'. I would however recommend Anandtech to next time use Pa. as a term as you'll not find high tech fan-curve sheets using mmH2O (anymore).
On a side-note, if fan reviews are going to be used.. I would like to suggest a push-pull efficiency review on liquid coolers.
In the industry we almost always use pull configurations through heat exchangers as underpressure draws more evenly through an exchanger than otherwise possible with the momentum of the air. I don't have enough liquid coolers systems to be able to make a review, but my own pc is a degree lower. Considering some people might like to know these kind of things and I enjoyed to read something this close to my actual work, who knows.
Ohw, and just for bs argumentation and to be part of this thread: MSc.
mctylr - Wednesday, December 23, 2015 - link
I want to echo [Freaky Angelus]'s comments; mmH2O (millimeter of water) is a dated unit of measurement that isn't the preferred unit in any context as far as I know. It was essentially a metrication variation of inches of Mercury (inHg) that should be left to the history books. Pascals (Pa) would be the unit used for scientific literature in US and abroad. I'm not familiar with US engineering / commercial preferences for fluid dynamics and air flow/fan measurement to be sure, but I suspect they would prefer Pascals over mmH2O as well.toxinate - Thursday, December 3, 2015 - link
BS in Computer Engineering from Rutgers, represent!I just hope you're right, lmao
AntiLumo - Sunday, December 6, 2015 - link
Hey Hulk,I haven't even thought of getting any kind of degree on anything, yet I still use word impedance to describe the flow resistance of helium mixture flowing inside mechanical parts and I do it daily at work. In physics it is quite normal thing to use word impedance, because you can't just say flow resistance since there are so many things affecting the flow resistance and as you can see it's much longer term than impedance.
Probably it would be fine to use the term "head", but I think that is largely used in technical papers and not in review texts meant for common people and I think that the graphs with impedance are nice and clear.
And E. Fylladitakis is not being rude at all, he is just explaining how things are done in this review.
ImSpartacus - Wednesday, November 25, 2015 - link
I don't come to Anandtech for this passive aggressive condescending horseshit.Before today, I was able to say that I've never seen an Anandtech writer snap at a commenter regardless of how coarse they were. You just broke that streak for me. I'm pretty disappointed and a little upset.
Let's be 100% clear in that the content of your post can be completely correct (it seems to be) and the post is still unacceptably rude.
I come to Anandtech to learn new things and I absolutely love to see its writers present new ideas. I don't want that to ever change. However, I don't come to Anandtech to see its writers be appear rude to readers even if those readers are rude "first".
It's really hard to accomplish that because it's so easy to be accidentally appear ruder than you intended - that's just the nature of the limitations of written communication. You really have to go overboard to make sure that your comments can't be misinterpreted.
I'm just concerned because Anandtech has historically been so awesome at that and I don't want that to change.
E.Fyll - Wednesday, November 25, 2015 - link
I am sorry if you, or anyone else, is offended in any way. I certainly did not mean to offend anyone. It just appear to me as common sense that someone who is studying mechanical engineering would not have heard of the term mechanical impedance, except if he/she was a first or second year student. That is why I took my time to explain the difference anyway.Hulk - Wednesday, November 25, 2015 - link
E. Fyll,Holy crap! Your post wasn't rude at all!!! It was informative and your informed opinion. That's what we're supposed to do here isn't it. Are so we politically correct that we can't even politely disagree. And your post was TOTALLY polite. As you can see from my post I also have a different opinion, but it's just that an opinion. There was no mocking or personal attacks here until the person who was so "offended" by your response posted.
Please, please, please, don't apologize when there is no need. We need to have just a little free speech don't we? Can't we discuss and...yes, God forbid sometimes disagree, and in the end learning something? I learned that fluid resistance is also impedance. That's cool and that's why I'm here.
Let's be active and engaged, let's argue sometimes, let's laugh sometimes, but please let's not be afraid to freely honestly express ideas and opinions!!!
Manch - Wednesday, November 25, 2015 - link
Oh hell... here we go... there will be protests, a demand for E. Fylladitakis firing, and a creation of a "safe zone" comments section...That poster must be from that college in Missouri...
Kutark - Thursday, November 26, 2015 - link
Seriously. People literally INVENT things to be offended by these days.tim851 - Wednesday, November 25, 2015 - link
No worries, dude. Your reply was fine.ninjaquick - Wednesday, November 25, 2015 - link
I don't come to anandtech to read shitpost comments by butthurt intertards.Kutark - Thursday, November 26, 2015 - link
<3Folterknecht - Wednesday, November 25, 2015 - link
Where was the author rude in his reply? Direct maybe, rude no!And as far as I can see the author is from europe - maybe both of them misunderstood each other completly when it comes to the term for resistance/ impedance. I get the feeling that american and european schools/universities might be using different terms and methods to describe the same phenomenon, that wouldn't be first time (X-Rays vs "Röntgenstrahlung").
And in general europeans are a little bit more direct and you re lucky that the author isnt from germany, we re often even more direct ;-)
No reason to get all upset here imo.
ImSpartacus - Wednesday, November 25, 2015 - link
As a reader for many many years, I'm admittedly a mega-fanboy of Anandtech and I've previously paraded them around as being nigh-perfect compared to other tech blogs in many areas - including their writers' attitudes with readers (which is often lacking as technical-minded people can often be).It really comes down to no longer being able to say almost-absurd comments like, "I've never seen a comment even remotely rude/insulting from an Anandtech writer" in good conscience. I like to think that if I'm going to act like a shameless fanboy around others, I have to be more than justified in doing so. I just can't do that anymore.
I fully realize that I probably put Anandtech on an unrealistically high pedestal. They are good, but the site is ultimately run by human beings and that's probably ok.
Beany2013 - Thursday, November 26, 2015 - link
If you want to get offended on someone else's behalf, Tumblr is thataway --->Although in all seriousness, I think you must have read that comment from an *entirely* different perspective to me; I thought it was two people familiar with the science discussing their different interpretations of terminology in a fairly chilled out way. I mean, if you think someone has got something completely wrong and are trying to explain it the way you understand it, it's hard to *not* come across as passive aggressive, but I'm not getting that vibe at all.
Kutark - Thursday, November 26, 2015 - link
I'll just leave this here for you, since clearly you ran out...http://www.amazon.com/Playtex-Tampons-Multipack-Un...
Oxford Guy - Thursday, November 26, 2015 - link
Save the sexism for 4chan where it can, at least, pose as wit.Phreedom1 - Sunday, December 6, 2015 - link
He wasn't rude in anyway. Man you are thin skinned. So many people these days getting upset over little or nothing.KAlmquist - Wednesday, November 25, 2015 - link
According to the Wikipedia article you link, "mechanical impedance" is the ratio of force to velocity in response to a harmonic force. It's analogous to electrical impedance. It follows that your use of the term "impedance" is incorrect for two reasons. First, you are measuring a constant airflow, not one that alternates directions. For "impedance" to be relevant, you would need a force that alternates directions, which would generate an alternating air flow. Second, air resistance is not proportional to air velocity, so there is no fixed ratio between the two. To talk about "impedance," you need a linear system.Your second link does show that your usage is not unique. But your usage doesn't seem to be correct.
jann5s - Thursday, November 26, 2015 - link
+1ninjaquick - Wednesday, November 25, 2015 - link
Impedance is the capacity or nature of impeding. Which is to say prevent motion or flow or activity.Saying it is limited to Electrical Engineering is like saying Tomatoes are limited to Ketchup.
wolfemane - Wednesday, November 25, 2015 - link
Why... What... What else would they be used for?????phoenix_rizzen - Wednesday, November 25, 2015 - link
Salsa, obviously. Or hot sauce. Or any other number of sauces unrelated to that horrid sugary concoction known as ketchup that should be avoided at all costs. ;)Strife2211 - Saturday, March 12, 2016 - link
I see it all the time. Instead it is noise caused by resistance. Filters. Grills. Etc.tipoo - Wednesday, November 25, 2015 - link
LED...Lower performing? I thought LEDs made everything in PCs faster, everyone knows that! My world is shattered.Devo2007 - Wednesday, November 25, 2015 - link
Actually, black makes a PC faster, so having LEDs will just slow things down.Devo2007 - Wednesday, November 25, 2015 - link
.... I read that online from some guy named Dan a long time ago... He had a lot of Data about all kinds of things.(obviously the post in question was sarcasm at its finest, but it seemed relevant to tipoo's post.
Flunk - Wednesday, November 25, 2015 - link
Na-uh, everyone knows stripes are the fasterest!Chapbass - Wednesday, November 25, 2015 - link
That dude that shows up to the LAN with his case all decked in like...camo...yeah, watch out for that guy.Manch - Wednesday, November 25, 2015 - link
These rules apply to cars as well.That's why my go fast stripes are black. And led lights? pfffff, all the extra wiring, boards etc add weight. Weight is the enemy of speed. No LEDS for me...
BurntMyBacon - Tuesday, December 1, 2015 - link
@Devo2007: "Actually, black makes a PC faster, so having LEDs will just slow things down."That's racist. </joke>
Also, it didn't make my bacon faster, but I guess that's not a PC.
Caravaggios Fist - Wednesday, November 25, 2015 - link
Great review of fans. I love to see performance curces rather than the regular two point tests that are more conventient to do. It would be great to see sound dB info too if you get the equipment in the future.Also, I agree that the graphs need to look a little more consistent with the Anandtech style.
DanNeely - Wednesday, November 25, 2015 - link
A suggestion for when you've got data on more fans. Since the scales on the graphs aren't constant; it would be really helpful for comparison if Bench was able to take the datapoints used to generate the performance curves and use them to create a single combined chart for the user selected set of fans/speeds.BurntMyBacon - Tuesday, December 1, 2015 - link
A max and min speed chart also needs to be in there as user selected fan speeds ignore the fact that some fans can spin quicker/slower than others are capable of spinning.Stuka87 - Wednesday, November 25, 2015 - link
My Corsair 380T came with the LED version of the fans and I was not happy with them at all. I ended up changing them out for the Enermax "Batwing" LED fans (which have removable blades for cleaning) which are much quieter but also have better airflow.supastar1568 - Wednesday, November 25, 2015 - link
Very well written write up with lots of good information.Curious, how did you go about increasing/decreasing the air resistance levels for the fans? As an mechanical engineer, I've been reading up on fan curves for work and I believe an adjustable duct type thing is common. I've always been curious to how this part is done.
Also, for anyone interested, the AMCA 210 Standard has lots of good info as well. A quick google search will bring you to the .pdf.
simonpschmitt - Wednesday, November 25, 2015 - link
Hey, nice to see a more in depth review of the matter.Just two suggestions:
First, why are you plotting pressure over volume? Iwould have done it the other way around.
Second, could you measure the impedance (ballpark figures) of some typical obstructions and post photos of them? For example finger guards, one or two tower coolers, typical filter (ideally clean and dirty, would be interessting to see the effect). Just to get a feel for the dimensions involved.
jann5s - Thursday, November 26, 2015 - link
+1EddyKilowatt - Thursday, December 3, 2015 - link
I think a necessary corollary of AT doing such a solid job of reviewing and explaining fan performance, is that they will also (have to) start measuring and publishing flow curves of the things they review that fans get hooked up to. I was mainly thinking of cases and coolers, but you are right, filters and guards qualify as well. Maybe someday that kind of data will be part of the specs that manufacturers publish.rocktober13 - Monday, November 30, 2015 - link
+1 to both suggestionsBurntMyBacon - Tuesday, December 1, 2015 - link
@simonpschmitt: "First, why are you plotting pressure over volume? Iwould have done it the other way around."Typically the controlled variable is X and the measured variable is Y. They are controlling pressure and measuring airflow so typically it would be represented as you state. That said, the data is all there and the author is technically free and correct to use R and W for variables if he chooses as long as the axis' are labeled. Though in the end, I'd just as soon have the author use the typical approach to avoid confusion.
@simonpschmitt: "Second, could you measure the impedance (ballpark figures) of some typical obstructions and post photos of them? For example finger guards, one or two tower coolers, typical filter (ideally clean and dirty, would be interessting to see the effect). Just to get a feel for the dimensions involved."
This. Very this. Also, since the author mentioned "dense radiators", measurements and photos of dense vs sparse radiators.
JanW1 - Wednesday, November 25, 2015 - link
"one Pascal is the pressure required to accelerate 1 kg of mass at a speed of 1 meters per second squared"Looks like you mixed up the definitions of a Newton and a Pascal. One Pascal is the pressure required to exert the force of one Newton per square meter. The square meter disappeared in your definition. I guess you could say "one Pascal is the pressure required to exert on a surface of one square meter the force required to accelerate 1 kg of mass at a speed of 1 meters per second squared".
Ian Cutress - Wednesday, November 25, 2015 - link
Sorry, that was my bad. I edited a few things and got mixed up. The definition now reads 'one Pascal is the pressure required to accelerate 1 kg of mass at a speed of 1 meters per second squared per square meter'. Admittedly it's a bit clunky, but it's the best way (from my perspective) to visualize it.nutral - Thursday, November 26, 2015 - link
It kind of says that 1 pascal is still the pressure required to accelerate 1 kg of mass over an area, i would just keep it at 1 pascal is equal to 1 newton per square meter. the gravity acting on 1 kg is 9,81 newton.Arbie - Wednesday, November 25, 2015 - link
If you're starting on fans, check out the old X-Bit Labs site. They had very thorough reviews which helped a lot in balancing noise vs cooling performance for different applications. Here's a link to that section:http://www.xbitlabs.com/articles/coolers/
They haven't published in a year, but fortunately the material is still available. Sad to see them quit...
sheh - Wednesday, November 25, 2015 - link
Thanks for the interesting review.It might help if there was a way to summarize the results in a more condensed manner. I don't have an ideal solution, but perhaps overlay more results in a single graph.
It would help if the voltage was specified in the graphs, and not only the RPM.
Some basic details are missing on the fans: startup voltage, "cruise" voltage, official specs (would be interesting to see even if somewhat arbitrary).
It would be nice to see more detailed noise-vs-RPM/voltage figures.
BTW, parts of some of the graph images are blurry. Anandtech's become much less consistent in the visual looks of graphs, and with occasional hiccups in quality. Also the frequent low quality JPEGs in the homepage make it less polished than it used to be.
britjh22 - Wednesday, November 25, 2015 - link
Great technical analysis as always on this site. I'd just recommend one addition, a real world test. It would be very interesting to take a standardized test bench, let's say an overclocked i7, and have a 240mm AIO solution and an air cooler. It would be great to be able to see the actual effect on the temps by just changing fans, and would give a great data point for user value analysis. Being able to see how a fan statistically performs is great, but the real world effect vs. cost is what I would LOVE to see.lorribot - Wednesday, November 25, 2015 - link
Whilst i love all the technical stuff all the graphs don't really help me decide which fans i should be buying for my case and radiator. What I need is some specifics such as fan a will need to run at x000 rpm to cool a 200w processor to 60C and produce xdb at that point. That would be a repeatable real world test that had some meaning. For case fans you need some form of test for rpm and db for each fan to provide a specific airflow that would be need for something like the above load plus a Graphics card.A separate article on the differences a case design can make to noise and fan performance would be good, especially the difference a grill would have over a perforated sheet metal.
Oxford Guy - Thursday, November 26, 2015 - link
"all the graphs don't really help me decide which fans i should be buying for my case and radiator"What you meant to say is that it doesn't tell you everything you want to know to make informed choices.
Oxford Guy - Thursday, November 26, 2015 - link
One thing that most people don't take into account are the frequencies emitted from fans. I know it's a lot to ask but it could be nice to see the spectrum for fans that are tested. Typically, more decibels in lower frequencies is more pleasant, and less likely to aggravate tinnitus, than more in shriller ones.nutral - Thursday, November 26, 2015 - link
The graphs are a bit confusing, especially with the colours. And did there have to be so many? Impedance in this case is not the correct term, because that applies to the impedance it has to changing air flow, in this case it is resistance. (like in electricity, resistance vs impedance, where impedance is used for AC)I would really rather use Pascal instead of mmH2O and m3/h instead of cfm but i guess that is your choice, having 1 imperial and 1 metric measurement is a bit strange though. 1mmH2O or 9,81 pascal is equal to about 1 meter of air column head. The left legend also should say static pressure instead of just pressure, because the pressure is a compination of static and velocity.
The graphs are called Pressure/flow, using volume is the incorrect term, as it is not talking about the volume but the actual displacement of air.
I was thinking of making a test setup myself for testing fans and their pressure/flow curves. How did you setup your apperatus to restrict air flow and test at different static pressures?
It would be really nice to have a simple module on the website, that puts the graphs of fans and graphs of different components like radiators and cpu coolers over each other to see what the best performance would be.
kalboston - Thursday, November 26, 2015 - link
All in all, an excellent article with very informative information. I am also not use to seeing performance/fan curves presented this way but I am able to understand what the author was trying to convey.Typically in the US, airflow is measured by CFM (Cubic Feet Minute) and In H2O (Inches Of Water Column/Gauge) or M/S (Cubic Meters per Second) and Pascals.
I do agree mixing imperial and SI units is not typical, but there are plenty of applications in the real world that this is done.
I do believe the author tried to perform a controlled test based upon his other articles and methodology. Without knowing the test setup and techniques used, there is always questions that could potentially affect the results. I for one would not have used a pitot tube, most likely a thermal anemometer. A very accurate test setup for airflow is a lot of money for a tech review site that would most likely never recoup the cost.
End result was a great amount of time by the author testing multiple units with many potential variables and presenting the data in a way that they felt was appropriate. For that I am greatly appreciative of, it was an excellently presented piece taking their perspective into account.
I have no college education, although I do have almost 10 years of Biotech and Semi Conductor Cleanroom Certification and Balancing.
I also learned something a long time ago about engineers summed up to a saying I saw "Arguing with an engineer is like wrestling a pig, after awhile your realize they enjoy it"
nutral - Thursday, November 26, 2015 - link
He actually did use a anemometer for the airflow test i think. The pitot tube is just used for static pressure readings. I'd say the measuring setup is pretty accurate for what he is doing here. All in all its a great way to measure actual fan performance instead of going after just a flow number and a static pressure number.From my work as an engineer i'm more used to seeing actual graphs of pressure/flow and power to make a good selection. Papst for example does supply them. The only computer fan i've seen else that show these are ek waterblocks
http://img.ebmpapst.com/products/grafik/7438-GRAFI...
https://shop.ekwb.com/media/catalog/product/cache/...
kalboston - Thursday, November 26, 2015 - link
Thank you, I meant to write vane anemometer instead of a pitot tube. As long a the same equipment is to be used these fan comparisons are effective.Just some points to make about the equipment used. Not to say the article is not accurate, but the equipment used is "inexpensive" by trade standards. Such as some of my equipment that I have for those once in a while measurements. The temp/RH, sound, and vibration equipment I own is "cheap" compared to people who specialize in it the field. I wouldn't expect Anand to be using $2,000+ USD thermal anemometers (TSI 9565P temp and barometric compensation) and $2,000+ USD (Shortridge ADM-880C) digital manometers.
The Extech AN200 description does say a thermal anemometer, but the picture shown is a vane anemometer with a FPM accuracy of 3% reading + 40 FPM.
Assuming a 140mm diameter @ 60 CFM = approximately 600 FPM velocity. The AN200 unit is +/- 10% at this reading.
The HD350 accuracy is +/- 0.3% of full scale @ 510 mm h2o = +/- 1.5 mm h2o. Most the readings taken are within the tolerance of the meters accuracy.
I do enjoy reading the author's articles. I have never posted before to Anand, I just hope a few people will hopefully understand before attacking/leaving harsh criticism of the article.
jann5s - Thursday, November 26, 2015 - link
Love the quote in the end, and I'm an engineerTunnah - Thursday, November 26, 2015 - link
Great article, super technical as always, and as usual WAY WAY over my head ha.What I think would be useful is a real world scenario that dumdums like me could look at and see the benefits of. Say, an overclocked CPU with a certain cooler, and the temperature taken with the normal fan, then taken with the custom fan, so we can see the benefit we care about - how much cooler it makes something.
If this isn't possible, due to the fans not fitting, maybe another scenario, a system instead of a CPU cooler, with it's standard fans, then these fans.
I know this probably won't be a popular idea as AnandTech is more for the tech savvy crowd, I just feel like some real world results would help
nutral - Friday, November 27, 2015 - link
This article is mostly about how much flow the fans actually create in certain situations.It can be coupled with how much cooling you have at a certain flow, altough that really depends on the cooler/aio
galta - Thursday, November 26, 2015 - link
Noctua fans, always and forever.The best combination of static pressure and airflow for a given noise level. On top of that, they last close to forever.
Of course, one has to say that they are from beautiful, but that's minor.
Oxford Guy - Thursday, November 26, 2015 - link
They make boring black fans, too.galta - Friday, November 27, 2015 - link
Sure they do. Their industrial line is black/dark brown.I have 6 of them on my system, on a push-pull 3x120mm radiator. Nothing short of superb.
They come at price, though, and not because of their color, but because of their extra specs.
Anyway, as I said before, color is minor.
FriendlyUser - Friday, November 27, 2015 - link
Great, rigorous review!! This is exactly the kind of fan review that I want. Please do a roundup of premium fans soon.Omega215D - Friday, November 27, 2015 - link
Time to compare it to the H.E.C Cougar and Noctua fans.Gonemad - Friday, November 27, 2015 - link
Too bad I couldn't find the non-led versions around here in time and budget; I'm sick and tired of that blue glow emanating from a case when you need some sleep. I got a case with thick foam filters (Tt Armor), so SP fans for me all the way (3rd rebuild in this case). Fewer, thicker blades is what I am looking for. No care for loud fans, since the graphic card makes the point moot.jhh - Sunday, November 29, 2015 - link
Yes, I got a similar case, and was given significant spousal input that the extra light had to go. A little bit of time with a wire cutter extinguished the LEDs, but am still left with the inferior fans.Gonemad - Monday, November 30, 2015 - link
I thought the same thing about the wire cutters, but I'd need to take the whole thing apart and search the proper wire to cut on the led ones I got. Not gonna happen, can't be ars... bothered.TelstarTOS - Saturday, November 28, 2015 - link
Very good article, but performance/noise charts and analysis are missing. I.e. in this case I'm not even sure that SP perf and quiet versions have same noise at let's say 1200rpm.EddyKilowatt - Thursday, December 3, 2015 - link
Cool stuff, guys. I appreciate the detail, and I appreciate even more that you didn't couple it to a self-deprecating "if your eyes haven't glazed over yet" type comment, all too common in the popular press when the going gets technical. AT continues to differentiate itself, *and* increase its credibility.juliabrown943 - Thursday, December 3, 2015 - link
what Jeffery said I am impressed that some one able to make $8960 in one month on the computer . you could try this out...............................juliabrown943 - Thursday, December 3, 2015 - link
what Jeffery said I am impressed that some one able to make $8960 in one month on the computer . you could try this out............................... earni8samsp99 - Monday, December 7, 2015 - link
To PWM or not to PWM?Does undervolting a fan to 7v have the same effect as a PWM fan set to 70% ? Is the static pressure and noise the same when flow is somewhat impeded?
TheRealAnalogkid - Monday, December 7, 2015 - link
I am greatly offended that I didn't something blah blah my rights blah blah my Dad's an astronaut bippy-boppity thar tekken ar jerbs.Oh, and thanks for starting fan reviews.
kaesden - Monday, January 4, 2016 - link
do these fans have PWM connectors or just the 3 wire connectors?Gasaraki88 - Tuesday, January 5, 2016 - link
I hate how you change the scale for every single chart. Why leave so much blank space in the CFM scale for some charts but not others? You make the scale all the way up to 120CFM but the max a fan got in the chart was 70 CFM. You could have lowered all the charts to 80CFM scale and it would have been easier to read and compare. I thought they teach these things in math classes in high school.Gasaraki88 - Tuesday, January 5, 2016 - link
But thanks for your hard work with the tests.