Gadgetory

following up on the GN roadmap we published we're taking a very popular viewer request which is to update our testing methodology for cases and expand it even further so one of the things we've wanted to do for a long time is expand upon our already existing four years now noise normalize testing for coolers and add it to cases more on what that means in a moment and then you all wanted us to do a standard set of fans for testing all the cases that come through the lab so we've done that and today we'll be looking to see whether or not you all led us astray before that this video is brought to you by drop and their HD 6xx headphones the 6xx headphones offer high quality audio output with wide device compatibility mouth drops 6xx also includes a six-foot cable with 3.5 millimeter plug alongside a quarter inch adapter for those who might want to plug into an app the 6xx focuses its energy into balancing the sound leaning toward warmer and base your audio they're also easy to disassemble and replace individual parts making these headphones trivial to maintain for long-term use learn more at the link below the discussion here is actually really complicated genuinely it is it is a very complex topic to get into things of airflow in a case to begin with and then you start trying to normalize fans in cases with different air flow paths fans that are rare cases that are designed for different fans and you might be putting in there cases that do better with negative versus positive pressure setups and they do exist so it's it's genuinely a very difficult topic to tackle and the reason that we haven't done these tests despite getting requests from viewers for years is because we know there are a lot of flaws and doing them but there's absolutely validity to the requests of standardizing a set of fans and the Patrick's credit who does most of our testing for cases these days and the rights the build section it Patrick made a point to me which was well if a lot of people are requested it we really should explore it and I completely agree so we're gonna do that what we're going to do first is I'll talk through the scripted section here about the testing method in some brief detail there will be more in the article linked to the description below and then we'll talk about the result those tests and then please stick around for the conclusion because I'll talk through some special scenarios that this testing will not work for or that your next suggestion following this might not work for so stick around for that part of the video so that's I'm going to try and boil down years and years of case reviews we we have in our spreadsheet we have over 200 results like rows in our case review for one of the tables one of them and we published something like four or five charts before these so hundreds if not approached in a thousand rows of data on cases over the past several years and to try and compress this down is a great disservice to what we've done but I'm going to do my best to try and remember some of the key points that you need to keep in mind in case you haven't seen every case review we've ever done which is completely valid so first of all a few things problems with the standardized case fan placement testing we've listed a few of them today let me first start by saying it is still a valid idea we are glad that those of you who pitched it have pitched it and done so repeatedly and we're probably going to keep doing this kind of testing at least for the foreseeable future because it can teach us a few things for example in the fractal measure if I see it further demonstrates how the fans are sort of biased up towards the CPU which benefits the CPU greatly but doesn't help the GPU as much so that's something valuable that we can learn from that testing but there are a lot of things wrong with it too and the NZXT h 500 is just the tip of the iceberg on that we're in c b results you see a big improvement but in GPU results you see that stock was better and that's because this isn't that case but in that case you actually end up with the stock negative pressure set up here pushing there that way and this way it sucks in air through everywhere else in the case so you get a natural pull through the front because of the negative pressure set up but also and critically through places like the bottom so you pull some air up through the bottom that ends up in the video card you actually pull some air through the back through the PCIe slots and through the grills in that area that benefits the video card so the video card has effectively front panel access to air because that's as far as it cares that's what this is this is the back of the case this is a PCIe area but as far as the video card carries it might have whole be the front of the case with clean air that's cold and while relatively cold and that's because all the exhaust is pushing up and back and so the air will go in everywhere else it can as soon as you change that to a positive pressure setup what you're doing is you're forcing the hot air out the back and that means you're not pulling cool air through the back where it was originally so the results and the H 500 get worse now that's fine as long as we and this is our job in the review side of things as long as we clearly highlight what's happening we explain why it's happening and hopefully most of the people watching will will watch that part and catch you didn't understand what's going on now part of the challenge we face in terms of responsibility of reporting is that the viewer attention graph and it will be true for this video as well look something like this and as you drop to 30 40 % viewership towards that into the video potentially a lot of people are missing explanations and then those are the same people who often go around and post things on forums out of context and mislead everybody or provide misinformation so we have to be really careful about that and that's one of our greatest hesitations to doing the standardized fantastic because it's very likely at some point probably today when this video goes live people will share the information that they think they've learned without sharing the caveats which might include things like hey this standardized test setup is actually worse than stock for this specific scenario for GPU in this case of the H 500 so we still learn some really cool stuff it's a good test it should stay in the testing it's just that we at GN have to be very careful to explore why the results are what they are and then make the distinction in the reviews is it because the case is good that it happens this way is it because the case is bad that it happens this way or is it neutral and it's just because we've modified how the case was designed to perform a lot of cases don't really get that kind of design where they smartly try to do something like a negative pressure setup actually works or where they try to guide the airflow in t specific areas to cool specific components a lot of time you get a very simple case like this one the Versa j24 where their approach was let's put holes in the front let's put holes in the back and then let's put fans in the holes pretty simple front to back airflow so in these instances doing the standardized test can actually benefit us a lot because what we learn from this case is that switching to the standardized fans improves the performance what does that teach as well what it teaches us is that these fans and it sort of validates what we already thought we knew which it's always good to validate what you think you know it validates that these fans are not particularly good for static pressure performance and they're not particularly strong either and the CFM is a bit lower so you end up with the scenario where fewer better fans outperforms more low unfenced and that's valuable to learn from that testing a couple other things we've explained this a lot in the past but for those of you who may not have seen it there's an important item of note with the top of a case in a good amount of cases this is this kind of getting into a quagmire to it there's a lot of different like rabbit holes we can go down but in a lot of cases you can do things like put a fan on the top and if you put the fan on the top front and you do intake top front being the front most position of the top and you do intake it will often actually improve the CPU thermals a lot of people this is an old thing we've set on the channel for years now a lot of people like to do the whole well but hot air rises okay yes it absolutely does but the increase in molecular speed is about four percent for every 25 degrees Celsius you climb so the thing is when when hot air rises it sort of made irrelevant when you look at blasting the air with a 2,000 rpm fan or a 1400 or a thousand rpm fan it doesn't matter the point of the matter is that a fan moving out some rpm greater than zero is going to move more air than just hot air rising you can make hot air go down you can make hot air rise you can Hodder go vertically it doesn't matter if you have a fan blowing at a higher rpm than the air is naturally moving it's sort of irrelevant now it's not completely irrelevant so don't think that out of context of course it matters to some degree which direction air naturally moves and how physics dictates the movement but the point that we're getting to is that if you put a fan here in a lot of cases but not all of them and you do intake it will improve the CPU thermals and that has nothing to do with hot air rising it is entirely to do with the fact that up here is air that is colder than in here that's all it is ambient temperature outside is anywhere from five to maybe ten fifteen and the really bad cases like the and so but five to 15 degrees often five to ten degrees cooler than the inside ambient temperature of the case we've measured some of these cases in a room ambience of twenty one to be upwards of forty and that's for the really bad ones again like the ENSO but in general we see something like thirty so because this air is colder and because this fan is blowing in and because the intake for this fan is immediately adjacent to the intake for the CPU tower obviously the cool air will go straight through the tower and that's gonna be your coldest source of air so the points then this doesn't work on all cases and of course in a lot of cases to putting exhaust in the top would be better especially in larger or longer cases where you can do an exhaust behind the CPU cooler and push air up after it comes through the cooler that might be beneficial too so there's a lot of really screwy things you can do with cases once you start adding more and more fans and it's very possible that you make the performance worse and so the point of saying that is if we try to expand too much with a standardized fan tested we are going to make the performance words in a lot of cases or it can't be standardized it's one or the other so if we start picking and choosing where the fans go it's no longer exactly standardized so we pick the the positions that will be present in basically every case make the most sense in basically every case should produce the same performance in terms of how they scale against each other in basically every case and beyond that what we're going to do is keep testing these on isolated charts against only the case itself before getting to the comparative charts and look at individual test cases of what if we had one fan would have to remove the filter things like that so another if you do a lot of people have suggested this where when we said we don't want to do standard fan placement people have often said well why don't you just fill the case of fans every slot it'll probably it doesn't really teach us a lot for one but it's probably going to be worse and that's because you get again into situations of like noise versus thermals and then you further get into situations of if okay we put exhaust on the top now what's going to happen is the air will come in the front to this top fan and it will be pulled out right away by the top exhaust fan before after hitting the CPU so you can make things worse and I'm not saying that these things are true for every single case if you're posting an anecdote in the comments about how your case does something different that's fine it's probably true even although it's hard well but it might well be true and I guess that's the whole point isn't it is that this can change on a case to case basis applauded for that pun and so as a result of this it's very very difficult to properly standardized in a meaningful way we think we've done a pretty good job of it clearly the most important thing is to just explain the results and make sure we understand what's going on and communicate with viewers what's going on when they're making a purchasing decision because otherwise people can get confused and they might see a result with stock performs and go where I was that better than the better fans and it's because well maybe the case came with more fans came with better fans better fan positions for the most part these results we saw improvements in the CPU section so that's what we want to see because that says we're doing something that actually has some value we originally disagreed with the idea of standardizing a set of fans or case tests for reasons that will become partially clear in our GPU results section but it remains the single most requested test in every single case review we've posted we recognized the merits of potentially adding this test and ultimately we take great pride in expanding and improving by listening to community interests for this testing we're using to Noctua NFA 14 fans in the front with one nff 12 in the rear it is of critical importance that anyone attempting to do this controls where each individual fan is placed even though we're using two of the same model fan on the front we have marked which one goes in the top slot and which one goes in the or of the two front slots this is because fan rpm varies by plus or minus ten percent from unit to unit so blindly switching can drastically impact results in a way that creates wide error margins being blind further to these important differences in fan speed could lead a tester astray when analyzing the results you could potentially make claims that aren't true if you randomly swapped which was where we control where each fan is placed and always placed the 120-millimeter option in the rear slot we noted that almost every single ATX case we test has at least two 1:40 front positions and one 1:20 rear position unfortunately in some cases this means reducing the fan count but we can't test with more or we begin to enter scenarios that eliminate most cases from testing by way of insufficient fan mounts in those cases we must also note that just fill in a case with fans even if not at random although a lot of people seem to really do that when they build computers just filling a case with fans though even if you think you kind of have an idea that makes sense can actually make things way WAY worse and we'll talk about this more later in the on-camera segments at the end of the day it's important for us to continue doing one-off testing per case not comparatively just against itself like our panelists tests and our tests where we add one fan as a user might do when buying just one extra fan with the case still though these standardized results can teach us about some of the case designs it's not as simple as lowest temperature means best airflow in the case because the case might actually be better in its stock configuration we ask that you please exercise your brain and actually think about the results and why they are what they are we'll do our best to explain them as we go but don't just look at the chart and then post all over reddit about how X is better than Y because one chart is not sufficient to make such a claim especially since it exits the stock configuration we've had this problem in the past when people request tests anyway it's still useful and can teach us a lot so let's get into those results standardized fan results are up first we split these into two bars per case one representing these standardized results and one representing the full stock results a quick glance reveals one immediate apparent outlier although it's not one and that's the BitFenix end so getting this one out the way first note that both these standardized and the stock results were up against CPU throttling so this is actually an identical result between the two the CPU is throttling so we can't see the actual true temperature because the frequency is shifting around in a less predictable manner further we previously proved that BitFenix and it's and so will recirculate its own hot air when installing fans in the front which you can learn more about an hour and so review we're seeing some of that here so that's it's one of the reasons that this is among the worst cases we've ever reviewed moving on from this the NZXT age 500 sees the most immediate benefit from standardizing the fans and the cpu test but this will change later for the GPU is asked the aged 500 saw an improvement of about 9 degrees celsius for air forced straight through the CPU tower cooler the mesh device seeds and the best here and by looking at the verse of j24 is results we can learn part of why the mesh if I see and the versa j24 are the smallest of the cases on these charts and positioning the front fans closer to the components does benefit the cooling of those components so being smaller in this instance is beneficial to results the impact of performance is noticeable in some ways and the higher velocity air is dissipating heat more quickly off of that large tower cooler atop the CPU second to both cases have relatively punched through front panels allowing for more airflow than in some of the competition the s2 vision doesn't change a whole lot which isn't surprising when considering its four stock fan configuration and the NR 600 sees some improvement in results a Seuss's heliosphere mains nearly identical which is a testament to how much it was really not impressive in the cpu cooling department but the GPU department changes things the standardized fan results for GTO temperature are much more interesting they're more varied and take some additional thought to understand and you do have to look at the cases to look at the positioning of the fans but they do make sense once explained and explored let's start with an obvious one the NZXT age 500 sees worst GPU thermals in this configuration than in the original benchmark and posts the opposite scaling from what we saw with our noteworthy CPU thermal improvement previously this is for a few reasons that were detailed in our original age 500 review first to the HP 100 is configured in a negative pressure set up for out-of-the-box thermals testing which means that air finds its way through every single hole in the case this includes small perforations in the front of course but critically includes the rear PCIe slot covers and grilles are holes in the back of the case near the GPU the GPU fans are therefore able to actually pull air in through an immediate access to cool ambient external air or ambient temperature anyway right from the back of the case it's as close to the GPU as you can get closer than the front where the air is coming from by switching to a positive pressure setup we take this away that set of holes in the back now becomes exhaust and the H 500 loses its highlight and also highlights a potential shortcoming in this section of test methodology because the case exits an optimal configuration at least for this build it is perhaps unfair to test in this way we can see CPU performance uplift but GPU performance decay so it's important to try and keep these factors in mind so that you can judge the case fairly and not just look at one chart and make a conclusion from that regardless the NZXT age 500 does worse here despite actually holding a top 3 results when under its stock conditions noting that these results are very limited right now and that's represented by the red bar moving on the NR 600 does well in this test which is entirely result of the weird positioning in this case then our 600 claims it supports 140 millimeter fans in the front but it really doesn't properly do so as discussed in the review that said the upside for this case was that the fans ended up directly benefiting the GPU despite being very oddly placed because they don't fit in those holes those were 120 s but there's still screws for 140 s and if you mount multiple of them well you end up with the weirdness we saw on the review it all comes down to positioning though the J 24 in the Helio signed up in the same area though we should note that the Helios actually did very well overall for its geothermal results in the initial review the master for IC isn't as impressive here which is because the fans end up biased towards the CPU the GPU receives more air across its backplate than the front of the card which doesn't amount to much aside from better CPU cooling the s2 also sees worse thermal results with standardized testing as it includes for fans in the stock configuration and those aligned better to heavily direct airflow under the GPU shroud this set of tests illustrates again shortcomings in just standardizing fan placement in size even when you think about what to do and where to put them because some cases are simply built or equipped better out of the box and we can't have one standard set up for every single case that doesn't screw some cases over while making others look better so you're just going to have to do your best to bear with us through more in heart charts as we add three views we can't go too high in fan count because we then eliminate the ability to standardize for smaller cases making this difficult to do so and again at some point you do start making things worse for example if you start putting top exhaust fans in you very well may end up pulling air away from the CPU cooler before the cool air ever gets there thus making it much worse than if you just left those top slots unpopulated and allowed the air to move naturally so more discussion on this later let's move on to the noise normalized results we've been running noise normalized tests in our CPU cooler reviews for over two years now so there's certainly nothing new for gamers Nexus we introduced these tests when we realized that CPU coolers were lacking a critical component for the reviews and this was probably 2016 or around them that we made this this addition to the CPU testing so the difference the important thing was to illustrate performance when controlling for the variable of noise it's not very fair if a cooler performs the best on the charge by simultaneously being the loudest as you could have a cooler at 60 DBA that beats everyone else's 45 or 40 or 36 DBA coolers one measured at 20 inches away but that doesn't count for much because well it doesn't mean a lot when one is much louder what counts is how well our hypothetical 60 dB a cooler does in the situation one level does the same noise level as everyone else that shows the efficiency of cooling as opposed to just brute force cooling so we normalize for 40 dB at 20 inches in our cooler testing system this is in a room the same room is the case test system although they are tested in isolation one is off when the other is on and that's with the noise floor about 26 DBA for cases it's the same story if someone wins and thermals by being loudest but the two results are isolated on different charts it hard to quantify and illustrate the actual difference between cases with different strengths we first demonstrated this with cases in our be quiet review where we or the follow-up where we looked at it versus the are l06 and that we can show on the screen now as well we found in the chart for the RL 0 6 vs. to be quiet case that the better ventilated RL 0 6 was able to outperform to be quiet case in not just thermals but also acoustics the thing that it claims to do best this was because the RL 0 6 could run its fans slower and draw more air in at lower noise as a result of its match whereas the be quiet case had to run its fans faster to overcome its own noise damping material and the creating worst results for both for these tests today we're measuring at 36 DBA at 20 inches away from the front of the case the rest of the testing uses our standard torture test with controlled power consumptions by the CPU and the GPU for the whole test and the 20 inch measurement is one we chose ages ago because that's about the distance just from polling we did several years ago that we think the average user sits away from the front of their case so it's just it's a number chosen based on a an assumed realism for the scenario the noise normalized results are on the screen now for these the only outlier in the performance scaling is the Helios which breaks pattern with a stronger GP performance than the other cases this is due to the positioning of the fans as we learned before in the review and because of the quantity of the fans note that we're back down to the stock fan configurations just tuned for 36 DBA the Versa j24 doesn't make this chart as its front fans cannot be controlled and so it's noise levels can't be reasonably normalized for these results the cooler master and our 600 does the best thanks to its ultra fine mesh front CPU performance that leads all other cases that we've tested thus far and it is a limited list to be fair the master 5c comes in next followed by the s2 vision and it's full assortment of fan spun down to the targeted noise level the Helio sleeves every one in GPU performance but not in CPU and then so is doing what it does best throttling very very hard the takeaway then this was largely a viewer request we are going to continue doing these tests the noise normalize one is the most interesting or useful one going forward because it allows us to isolate the final variable between noise and temperature and pick one and then say okay how efficient are they at cooling at a given noise level and then we can really really start exploring things like acoustic focused cases versus airflow focused cases although you start getting into questions at that point of frequency spectrum and maybe that's a future topic so there's a lot more we can do here of course frequency spectrum included but hopefully this gives you a starting point and finally thank you for the suggestion sincerely we do like we do like to try and incorporate things like that and improve because if we don't change what we're doing we don't improve today then you start falling behind and becoming irrelevant and we obviously don't want to do that so we're very much open to your suggestions these were good ones just wanted to be very careful about pointing out some of the ways cases behave so that people don't have illusions about what to expect so that's it for this one thank you for watching that you were suggesting veins go to store documents access net if you'd like to support our efforts like by buying one of these anniversary edition shirts with a tear down logo or the mod mats where you go to patreon.com/scishow his exes to pick up the latest behind-the-scenes video subscribe for more we'll be in B at Computex in the coming week I'll see you all next time