The T33 is almost 3dB quieter than the Noctua? 😲 That's astonishing. At a slightly lower RPM too. If you could bump the RPM 100 I wonder if it would beat the cooling and still be quieter. That fan needs to be further explored!
u r absolutly right - lower noise and loser speed - almost as good in cooling and this without the same RPM i think, with the same RPM as the NOCTUA, it would perform better than the NOCTUA - this should be tested !!!!!!
Bear in mind that the Noctua A12x25 is a high static pressure fan, not a high airflow fan. It does both well of course, but airflow is just not its primary purpose and it would absolutely torch the T33 for static pressure.
@@FanPhys yes and no. When you place the fan on the case it needs to overcome some pressure build up in the case. After all you want positive pressure case
@@FanPhys I guess you've never put a fan against a radiator, or a dense heatsink. Think of it like power washing vs a standard hose. The hose may have a larger flow, but without pressure, you aren't cleaning jack shit. Dunning Kruger strikes again!
Wow. I don't know if I missed you saying it, but the T33 coming in with the lowest dBA (even lower than the noctua stock fan!) by a NOT INSIGNIFICANT MARGIN while being only .2 behind the noctua and .4 behind the cheater in deltas... That's insane. An injection-molded version of that should be quieter and more performant too.. I suspect you've got another fan manufacturer reaching out to get you in contact with that designer...
If you could make the centre mass (& motor) small as possible (thus larger blade area), it should perform better. There should be testing with/without wing tips. If fact the channel could take a few designs/styles and just keep tweaking until it can perform no higher.
The T33 does need more investigation. The noctua needs to be retested to be put on the board and we need a 3d printed noctua mockup as reference on the board.
I was just thinking I want to see something that plays either a very ominous chord, or some famously pleasant chord. I unfortunately know very little toward music theory or the aerodynamics and aeroacoustics required for such a feat.
I would love it if you brought back the #1 fans from each season and retested.. So that the #1 fans are always part of each seasons list so we could see what fans are the best over all. Maybe even the top 2 fans
I liked the slight change in video style, with commentary over the smoke/airflow tests. You did a good job of highlighting the differences and points of interest. You should keep it up for future fan show down vids
Great episode! A quick tought; could you try filming the smoke test with a black cloth/sheet/whatever behind the fan? With some backlighing of the smoke it should come out much better on video!
I'd love to see a professionally made version of the T-33 fan, that thing is amazing. Wonder how hard it would be to make a fan sound like the F-104 Starfighter, or even the XF-84H Thunderstreak.
If i assume you meant the XF-84H "Thunderscreech" and not the Thunderstreak, that plane knocked people over with sound waves. Have fun with supersonic fan blades in you PC 🤣
Fan designs have a direction. The "shredder" was spinning backwards. I see that a lot in your testing. Another thing that will make a huge difference is the surface. A quick sanding with sandpaper will give the fans a massive boost. It not only increases laminar flow, but also lowers resistance/friction. Back when airplanes were riveted, switching to countersunk rivets actually caused planes to break world records by a wide margin due to the sudden performance increase.
But these are desk fans, not planes. He has a video where he tests the hypothesis that the print artefacts reduce performance, and it turns out it negligible (below incertitude). Also I believe he tests the shredder in the proper direction, if he was spinning it backwards there would be close to no flow (and in the opposite direction). Maybe the illusion that it spins backwards comes from undersampling.
The T33 looks like a great design. Those wingtips are added on to wings for a reason and it seems like they're helping a lot here. Print one out of ASA, and then maybe smooth the surface with 15 minutes or something in acetone vapor and it would probably outperform the Noctua (if it was smooth, that is). Although - that's not necessarily the case, come to think of it. A golf ball has a bunch of small indentations on it and it's incredibly aerodynamic. Mythbusters even tried scaling that effect up to a car and - it worked. So maybe the layer lines and such from 3D printing isn't hurting here, but helping? Would be interesting to test.
I think he did a test once where he smoothed the fans. there were no real difference, if i recall correctly. But, that may vary on a fan to fan basis, so it would be interesting if he took the best fans of the season and compared them with and without smoothing.
You should try and do a collab with Gamers Nexus once they get their fan tester set up. Take all the winners from the seasons and see how they stack up...
So many good suggestions in the comments, and this is one of them. I also like the idea to test a printed version of the noctua to test the effect of surface finish, and the idea to have a series dedicated to refining designs.
Your editing has improved so much. Funny as shit. If you've hired someone, keep them, if you're doing it yourself, keep spending the extra time. Keep on keeping on, right on!
agreed. the quick joke interludes consistently have me cracking up - not sure if it is just dad/pun humor or what but whatever the reason it doesn't matter keep them coming
Having a mix of freqs makes it more like white noise. I loved falling asleep to a fan blowing in my face in the summer where we had no AC back in the late 90s early 00s as a teeniebopper.
It would be super interesting to see a fan that was designed to have sound peaks arranged in a chord / harmonic. Something that accepts it will make noise, and strives to make that noise as pleasant to hear as possible, rather than technically quieter.
What you're describing is tonal noise, with spectral energy piled up at one or more frequencies related to bladepass. In the field of noise masking, atonal noise (with energy spread evenly across a spectrum) is deliberately chosen because it masks tonal sound to some degree without being as objectionable. The least objectionable noise of all is atonal noise (white noise, pink noise, brown noise) with reduced high frequency content (anything above 1kHz).
been a while since i checked in on fan showdown and im amazed by how thorough you are. You record and compare everything you could ever want to know about how the fans perform! I can't imagine how much work this takes. great job!
Wow the frequency analysis is fascinating! I feel like you can almost see where each frequency comes from in the mask fan design. Maybe I'm crazy though.
I don't even own a computer but I've been a fan of this series. It's interesting to see everyone's creativity and the effects that they have on fan designs and performance. It's also nice how you present everything and test the different designs
7:35 I haven’t been keeping up with the series, but the last episode I watched, you wanted a spectrum db meter thing. I tune in today, and well, there it is. Super cool dude. Awesome series!
I really like the airflow analysis comments. I never knew what I was looking at (really the significance), other than it looked cool. It'd be neat to see more of this (more flow patterns, why specific fans make those patterns, etc)
T33 looks a lot like the Gelid Wing fans (which were a great performing fan in their day). looks a little bit like convergent design with excellent results! hats off.
I'm so glad you have the fans pictured in the sound test again! The 3d models work great for this! I feel heard. Thanks so much for another awesome Fan Showdown where you listen to your fans!! 🥁
5:00 - Here in Florida attic ventilation can save you big bucks, couple passive cooling fans with insulation it can save hundreds a year in cooling costs.
the t33 is actually fan sized too most of the fans that stack up to it are monstrous beasts that are large hard to print and basically would never work in real world use but the t33 will absolutely function as a fan in the real world in some ones case
Holy cow! The T33 may beat noctua in performance if noise normalised. Please retest with a more restricting heatsink! There noctua should win, but cant be sure without testing.
I am curious to see the performance of the t33 mk3 with the cheater velocity stack and defuser. Why not test the a12 with it too. What is the advantages of the defuser in numbers?
I agree as well. A note from an engineering student here though: It isn't technically a diffuser, a flow straightener would be more accurate. Diffusers reduce airspeed by increasing the area the flow goes through. The benefit here should be that the flow will be more laminar(less turbulent) entering the cooler. That is also why velocity stacks are helpful. It is acting as a nozzle and the flow is accelerated as it approaches the fan blades. Then the fan isn't accelerating near stationary air and has a head start to increase the velocity. Just finished my fluids final so I though this might be an interesting bit lol. cheers.
Lookin major good in the video buddy! I like that look you've chosen! As always, nice video and i personally like that you toned the memes done a smidge.
We naturally tune out single constant pitch tones. A good example is you can be next to a case fan at a constant speed and stop hearing it. But if it changes tone, like one starting to go bad you will notice it almost immediately. Same goes for hard drives. If you have been around comps long enough you have probably already heard it.
I feel like the fans should be split into 2 groups between those that fit within the original dimenions and those that do not, since extra space does allow for much more room to manipulate the air
@@z0quetee in major hardware's own test between a12x25 and competitors, specically the t30, it was able to massively outperform the a12x25 at a slightly slower speed likely due to the extra 5mm of depth, since it has extremely similar blade design. Also, it's fairly telling that the best fan is one that massively exceeds the dimensions of the original fan, even removing the shroud.
I've been saying this for a while and he's apparently got no interest in that. I'm assembling a test rig and will be running a showdown as well, no fluff, just showcasing the designs and publishing data. No meme fans, just viable product.
t33 seems to be absolutely awesome. 1) very efficient; 2) very quiet; 3) doesn't exceed standard size; 4) easy to produce. it ticks all the important boxes and doesn't lose anywhere
You can double the throughput of the paths by making them loops and putting sparks on the right sides of the inside and outside of the path, a neat logistical trick.
The fan showdown usually just tests static pressure, but high airflow could affect cooling in a more passive way. I wonder if there is a better way to test this as well. I do believe if the airflow were on its own board i think that would be a bit of a different outcome than having the contest being focused on Static Pressure only.
I absolutely love being able to see the audio spectrum, it shows how the tri-mask is peaking on harmonics with the fundamental being one revolution at ~42Hz Edit: the harmonics are the relatively least "annoying" overtones given how they don't rub harshly against each other Edit2: comparing the peaks to each other will lead to some interesting data. For example: the T33 MK3 peaks at ~30Hz and ~220Hz, which is 7 times the fundamental, meaning that the noise peaks at each full revolution and as each blade passes the microphone. The fact that they all more or less peak at the same frequency as the revolution speed indicates that it may be caused by your setup and could potentially be ignored for the most part.
wonder if there is a good way to show static pressure. maybe a glass tube in between the fan and a radiator. so we can see how the smoke gets slowed by the rad. Would be interesting to see if some fans work better in airflow or static pressure scenarios.
The T33 with the way it moves air out in a tight beam makes it a good fan for headsinks. The ones that spreads the air out behind them are better as case fans where they distribute the cool air better.
If you want to make a fan for heatsink performance then angle the blade steeply from front to back with space inbetween them. For quite fan add more blades and let them almost overlap when viewed.
Tyler, if you see this, I think the monstrosity needs the angle to be more parallel to the direction of intended air travel. In other words, rotate the blades to have a steep angle of attack, so they push more air back at high speeds, even though they don't push as much air back at low speeds as a compromise. You use what? 10 degrees to 30 degrees of angle? Make that 60 degrees to 80 degrees of angle, or at least 45 degrees to 60 degrees of angle, and I think it will work a lot better. Finding the right angle is the difficult part, and you can probably use simulation software to find the right angle for the blades.
The sound spectrum is 10/10. Also I would like to point out, according to your testing, it's only 0.2 degrees more than the Noctua, but 2.8 db quieter!
I think you ran the shredder in the wrong direction, maybe due to how it was designed but a roof vent is designed to pull air out of a attic like you said. The shredder was pulling air in, maybe if it was spinning reversed or has the blades reversed to pull air/heat away from the radiator instead of push air across it might effect how it works, it may also need to be in an semi enclosed case with openings to draw cold air across the radiator and push it out. From what I learned in a civil class, they are to create a vacuum which and forces cold air to rush in, using a vacuum fan in reverse seems to kinda defeat its purpose in design so I can believe it would cause issues.
Have you done a 3d printed A12x25 fan (or a very close approximation of one)? Would be fun to see how 3d printed parts compare with the same shape to the original material composition. Or maybe a few different tests with different printed materials. There are some lightweight filaments that foam up when extruded at certain temperatures and it would be cool to see that compare to maybe a copper filled one that is super dense.
The performance is great - but what makes this a real standout is the noise level. The performance-to-noise ratio is IMO the primary measure of a fan, and when it can beat the original noctua by a significant level that is incredible. I really hope a company can step in and do further testing with a proper prototype.
Great episode! Would love to see a colab with you and Gamers Nexus. They will soon have an actual professional fan testing unit set up in the new studio.
That T33 needs to have more testing. Changing the pitch of the blades would be VERY interesting! How much of a difference would a steeper or shallower pitch effect noise and cooling performance? Also, I'd love to see it printed on a Resin printer. I know you've done tests in the past showing no real difference, but you haven't tried it in a while, and not on such a highly performing fan. You should try to get ahold of Gamers Nexus. See if they are interested in testing these fans. They are setting up a fan testing machine right now. I don't know if they have a 3D printer that they could print them themselves, which if they do, maybe you can get Gamers Nexus in contact with the creator of some of these fans. And if they don't, Maybe you can send them your fans ( or print new ones for them ) with the blessing of the fan creator. Would love to see these ( or at least some of them ) tested on a machine that's used by the industry to validate their own numbers. Also, maybe such a machine would be able to detect the differences in FDM vs Resin prints.
This is the first video from this channel I have watched in a while, so I have been a bit out of the loop, but I just have to say, James is rocking going full bald! It honestly looks great paired with the beard.
T33 lookin' hella fine! Bringing A12x25-like performance to the table, all while being even more silent? I'm sold! Hopefully a company, like Arctic will adapt the design (working together with the original designer of course) and create a new noctua-killer for third of the price.
7:00 I think it would depend on if the multiple frequencies were harmonious or dissonant, as well as where the overall pitch lies. In all cases, I'd think higher pitches would be more annoying, and then starting with most annoying: dissonant frequencies, single frequency, harmonious frequencies. Though, also, I always find myself humming to harmonize with steady background sounds, so I'd probably be fine with single or harmonious frequencies of a moderate-to-low pitch.
Just starting this video, I want to say I love this orange color for the fans. I missed the last couple videos so not sure if you've been using it before now.
I just want to see the room/locker/drawer/whatever , where you store all the fans you printed , like it would look amazing , a room having walls filled with different fans, like a fan art gallery lol
When you see multiple peaks at like 40 80 and 160 mhz these are considered multipliers of the same frequency. They sound won't be easily distinguishable. It's basicly like listening to the same chord on a different octave.
Turbulent air is better for heat dissipation than laminar air flow across a heatsink for example. Just thought I should point this out! Could you get a vane anemometer and build a box to seal between it and the fan to calculate the flow rate (volume flow in M3/s = velocity in m/s * cross sectional area in m2)
Is there an easy way to compare the area under the DB curve? After listening to way too much of The Jay Situation, my understanding is that’s more relevant than peak DB when discussing perceived “loudness”. It’d be interesting to see if it follows with your perception at least
I'd create 0.5 degree brackets and sort within them by noise. A temp difference of +/- 0.25 degrees is trivial but +3dB is double the sound energy (+10db seems twice as loud). Sound character matters too, of course, but 10.4 dB is *huge*, and probably should matter. Also, on that table the stock fan is listed as S1E3. Have you updated the results with your new test methodology? If not I'd be curious to see how it does with your current gear and techniques as things have aged or been upgraded.
do noise-cancelling headphones work better with a monstrosity-type spectrum, or a shredder-type spectrum? or is it more about temporal variability in spectrum?.....
The T33 is almost 3dB quieter than the Noctua? 😲 That's astonishing. At a slightly lower RPM too. If you could bump the RPM 100 I wonder if it would beat the cooling and still be quieter. That fan needs to be further explored!
u r absolutly right - lower noise and loser speed - almost as good in cooling and this without the same RPM
i think, with the same RPM as the NOCTUA, it would perform better than the NOCTUA - this should be tested !!!!!!
Bear in mind that the Noctua A12x25 is a high static pressure fan, not a high airflow fan. It does both well of course, but airflow is just not its primary purpose and it would absolutely torch the T33 for static pressure.
@@catgirldoll airflow is the primary purpose of every fan... that's what fans do lol
@@FanPhys yes and no. When you place the fan on the case it needs to overcome some pressure build up in the case. After all you want positive pressure case
@@FanPhys I guess you've never put a fan against a radiator, or a dense heatsink. Think of it like power washing vs a standard hose. The hose may have a larger flow, but without pressure, you aren't cleaning jack shit. Dunning Kruger strikes again!
Wow. I don't know if I missed you saying it, but the T33 coming in with the lowest dBA (even lower than the noctua stock fan!) by a NOT INSIGNIFICANT MARGIN while being only .2 behind the noctua and .4 behind the cheater in deltas... That's insane. An injection-molded version of that should be quieter and more performant too.. I suspect you've got another fan manufacturer reaching out to get you in contact with that designer...
Yeah & unlike the Cheater is contained in the 120x25mm Frame, so it is a real Noctua Contender
If you could make the centre mass (& motor) small as possible (thus larger blade area), it should perform better. There should be testing with/without wing tips. If fact the channel could take a few designs/styles and just keep tweaking until it can perform no higher.
@@fredfinks wrong
The T33 does need more investigation. The noctua needs to be retested to be put on the board and we need a 3d printed noctua mockup as reference on the board.
Also needs a side by side comparison with the same rpm.
Like the clean cut my dude! Nice to see a fan showdown been a minute 🍻
Bread 🍞 👍
Eyyy! This was a good one.
...
Yeah I didn’t really notice before but he really is quite attractive…
Now that we get to see the sound spectrum, someone needs to send him another siren.
Someone make Jericho trumpet
I was just thinking I want to see something that plays either a very ominous chord, or some famously pleasant chord. I unfortunately know very little toward music theory or the aerodynamics and aeroacoustics required for such a feat.
@@freescape08 maker’s muse made a video on how to make a whistling top. That could be interesting to see
I would love it if you brought back the #1 fans from each season and retested.. So that the #1 fans are always part of each seasons list so we could see what fans are the best over all. Maybe even the top 2 fans
Greatest hits in resin!
Yes!
top3 even, if there's room on the wall.
Make an All Stars leage for the best 2 or 3 fans of each season.
@@bkupron absolutely this. I'd love to see how all the top contenders perform against one another.
I wonder how well the T33 would perform if it was resin printed...?
I liked the slight change in video style, with commentary over the smoke/airflow tests. You did a good job of highlighting the differences and points of interest. You should keep it up for future fan show down vids
I miss being able to watch the fans running while listening to the sound they make.
it will take a bit more editing but it's doable to have both.
Great episode!
A quick tought; could you try filming the smoke test with a black cloth/sheet/whatever behind the fan? With some backlighing of the smoke it should come out much better on video!
I'd love to see a professionally made version of the T-33 fan, that thing is amazing.
Wonder how hard it would be to make a fan sound like the F-104 Starfighter, or even the XF-84H Thunderstreak.
Good god…. For that we need supersonic tips of the fan haha
@@nick_steele9790 And a spare turboprop engine. hmmmmmm
If i assume you meant the XF-84H "Thunderscreech" and not the Thunderstreak, that plane knocked people over with sound waves.
Have fun with supersonic fan blades in you PC 🤣
@@EnderMalcolm indeed
@@rasta77-x7o come on now, it only makes those within a 50 meter radius vomit, it's not THAT bad XD
Fan designs have a direction. The "shredder" was spinning backwards. I see that a lot in your testing. Another thing that will make a huge difference is the surface. A quick sanding with sandpaper will give the fans a massive boost. It not only increases laminar flow, but also lowers resistance/friction. Back when airplanes were riveted, switching to countersunk rivets actually caused planes to break world records by a wide margin due to the sudden performance increase.
But these are desk fans, not planes. He has a video where he tests the hypothesis that the print artefacts reduce performance, and it turns out it negligible (below incertitude).
Also I believe he tests the shredder in the proper direction, if he was spinning it backwards there would be close to no flow (and in the opposite direction). Maybe the illusion that it spins backwards comes from undersampling.
Our boi looks tough AF full bald and with a mighty beard.
The t33 Mk 3 was 10.4 Db lower and only .4 Delta higher I would say that the T33 Mk 3 should be in first place.
Especially since sound is less than half.
They place by temp. As an all-around great fan, though, you're right .
It was only 2.8dB quieter though?
@@802Garage Which is true, but those 3ish dB will make a big difference when you got a bunch in a case all stacking their sound signature together.
@@802Garage i was referring to the cheater fan on his list vs this fan. It was a 10db difference, which is half volume.
The T33 looks like a great design. Those wingtips are added on to wings for a reason and it seems like they're helping a lot here. Print one out of ASA, and then maybe smooth the surface with 15 minutes or something in acetone vapor and it would probably outperform the Noctua (if it was smooth, that is). Although - that's not necessarily the case, come to think of it. A golf ball has a bunch of small indentations on it and it's incredibly aerodynamic. Mythbusters even tried scaling that effect up to a car and - it worked. So maybe the layer lines and such from 3D printing isn't hurting here, but helping? Would be interesting to test.
I think he did a test once where he smoothed the fans. there were no real difference, if i recall correctly.
But, that may vary on a fan to fan basis, so it would be interesting if he took the best fans of the season and compared them with and without smoothing.
Take look at the be quiet fans they all have ripples on the blades and they very quiet and efficient
There's a difference between having a rough surface finish to begin with and having a smooth surface finish which is dimpled. Does that make sense?
In addition to giving up less than a half-degree in cooling performance, the T33 is the quietest fan on the list. That's solid right there.
Ayyyyy loving the bald look my man. You pull it off very well
You should try and do a collab with Gamers Nexus once they get their fan tester set up. Take all the winners from the seasons and see how they stack up...
So many good suggestions in the comments, and this is one of them. I also like the idea to test a printed version of the noctua to test the effect of surface finish, and the idea to have a series dedicated to refining designs.
Your editing has improved so much. Funny as shit.
If you've hired someone, keep them, if you're doing it yourself, keep spending the extra time.
Keep on keeping on, right on!
agreed. the quick joke interludes consistently have me cracking up - not sure if it is just dad/pun humor or what but whatever the reason it doesn't matter keep them coming
Having a mix of freqs makes it more like white noise. I loved falling asleep to a fan blowing in my face in the summer where we had no AC back in the late 90s early 00s as a teeniebopper.
It would be super interesting to see a fan that was designed to have sound peaks arranged in a chord / harmonic. Something that accepts it will make noise, and strives to make that noise as pleasant to hear as possible, rather than technically quieter.
What you're describing is tonal noise, with spectral energy piled up at one or more frequencies related to bladepass. In the field of noise masking, atonal noise (with energy spread evenly across a spectrum) is deliberately chosen because it masks tonal sound to some degree without being as objectionable. The least objectionable noise of all is atonal noise (white noise, pink noise, brown noise) with reduced high frequency content (anything above 1kHz).
been a while since i checked in on fan showdown and im amazed by how thorough you are. You record and compare everything you could ever want to know about how the fans perform! I can't imagine how much work this takes. great job!
I'm loving these edited clips, my guy. Your skills with video editing has come a long ways! You had me slappin my knee with the Dynamite ref haha
Wow the frequency analysis is fascinating! I feel like you can almost see where each frequency comes from in the mask fan design. Maybe I'm crazy though.
I don't even own a computer but I've been a fan of this series. It's interesting to see everyone's creativity and the effects that they have on fan designs and performance. It's also nice how you present everything and test the different designs
7:35 I haven’t been keeping up with the series, but the last episode I watched, you wanted a spectrum db meter thing. I tune in today, and well, there it is. Super cool dude. Awesome series!
Wow looking fresh! Glad to see you back!
I really like the airflow analysis comments. I never knew what I was looking at (really the significance), other than it looked cool. It'd be neat to see more of this (more flow patterns, why specific fans make those patterns, etc)
T33 looks a lot like the Gelid Wing fans (which were a great performing fan in their day). looks a little bit like convergent design with excellent results! hats off.
I'm so glad you have the fans pictured in the sound test again! The 3d models work great for this! I feel heard. Thanks so much for another awesome Fan Showdown where you listen to your fans!! 🥁
the hair moved from one side of the head to the other
5:00 - Here in Florida attic ventilation can save you big bucks, couple passive cooling fans with insulation it can save hundreds a year in cooling costs.
the t33 is actually fan sized too most of the fans that stack up to it are monstrous beasts that are large hard to print and basically would never work in real world use but the t33 will absolutely function as a fan in the real world in some ones case
Holy cow!
The T33 may beat noctua in performance if noise normalised. Please retest with a more restricting heatsink! There noctua should win, but cant be sure without testing.
@Green Mamba Games Smart!
I am curious to see the performance of the t33 mk3 with the cheater velocity stack and defuser. Why not test the a12 with it too. What is the advantages of the defuser in numbers?
Absolutely agree!
I agree as well. A note from an engineering student here though: It isn't technically a diffuser, a flow straightener would be more accurate. Diffusers reduce airspeed by increasing the area the flow goes through. The benefit here should be that the flow will be more laminar(less turbulent) entering the cooler. That is also why velocity stacks are helpful. It is acting as a nozzle and the flow is accelerated as it approaches the fan blades. Then the fan isn't accelerating near stationary air and has a head start to increase the velocity. Just finished my fluids final so I though this might be an interesting bit lol. cheers.
@@tristanhanley8741 thanks for the info. The cheater fan designer called it a defuser so I used the same name.
@@norad64 totally fair! I didn't want to seem rude or anything just though I would input a bit of extra haha
It would be interesting to see the T33 with different number of fan blades. One or two less and more, just to see how it turns out.
Lookin major good in the video buddy! I like that look you've chosen! As always, nice video and i personally like that you toned the memes done a smidge.
We naturally tune out single constant pitch tones. A good example is you can be next to a case fan at a constant speed and stop hearing it. But if it changes tone, like one starting to go bad you will notice it almost immediately. Same goes for hard drives. If you have been around comps long enough you have probably already heard it.
I'm falling in love with your editor. Great sense of humor!
I feel like the fans should be split into 2 groups between those that fit within the original dimenions and those that do not, since extra space does allow for much more room to manipulate the air
Ah, the Creative and "Just for Fun" groups.
Yes, but because of their larger diameter and weigth, they spin slower
@@z0quetee in major hardware's own test between a12x25 and competitors, specically the t30, it was able to massively outperform the a12x25 at a slightly slower speed likely due to the extra 5mm of depth, since it has extremely similar blade design. Also, it's fairly telling that the best fan is one that massively exceeds the dimensions of the original fan, even removing the shroud.
I've been saying this for a while and he's apparently got no interest in that. I'm assembling a test rig and will be running a showdown as well, no fluff, just showcasing the designs and publishing data. No meme fans, just viable product.
@@SwervingLemon You've got my subscription. I'll keep an eye out for it.
Impressive performance by the T33. Also that haircut looks absolutely boss!
Your flow analysis was way better than usual! Good job
this show just keepps getting better and better
Loved the movie clip joke edits throughout the video, chuckled at every one of those
t33 seems to be absolutely awesome. 1) very efficient; 2) very quiet; 3) doesn't exceed standard size; 4) easy to produce. it ticks all the important boxes and doesn't lose anywhere
You can double the throughput of the paths by making them loops and putting sparks on the right sides of the inside and outside of the path, a neat logistical trick.
The fan showdown usually just tests static pressure, but high airflow could affect cooling in a more passive way. I wonder if there is a better way to test this as well.
I do believe if the airflow were on its own board i think that would be a bit of a different outcome than having the contest being focused on Static Pressure only.
loving the hair dude, strong work
I absolutely love being able to see the audio spectrum, it shows how the tri-mask is peaking on harmonics with the fundamental being one revolution at ~42Hz
Edit: the harmonics are the relatively least "annoying" overtones given how they don't rub harshly against each other
Edit2: comparing the peaks to each other will lead to some interesting data. For example: the T33 MK3 peaks at ~30Hz and ~220Hz, which is 7 times the fundamental, meaning that the noise peaks at each full revolution and as each blade passes the microphone. The fact that they all more or less peak at the same frequency as the revolution speed indicates that it may be caused by your setup and could potentially be ignored for the most part.
woow your production quality really picked up. I like the memes, I like that you narrate the smoke test instead of music. Keep it up.
The clean shave looks really good!
wonder if there is a good way to show static pressure. maybe a glass tube in between the fan and a radiator. so we can see how the smoke gets slowed by the rad. Would be interesting to see if some fans work better in airflow or static pressure scenarios.
The T33 with the way it moves air out in a tight beam makes it a good fan for headsinks. The ones that spreads the air out behind them are better as case fans where they distribute the cool air better.
If you want to make a fan for heatsink performance then angle the blade steeply from front to back with space inbetween them. For quite fan add more blades and let them almost overlap when viewed.
The comedy bits are great, keep throwing some in.
Oooh T33 should be placed into Cheater's ducting :-)
What do you think Jason?
Look its Mr. Clean. LOL
I think the t33 mkIII should be in first from its combination of temp delta and low dB. And it's a proper fan size
Favourite series on UA-cam
Tyler, if you see this, I think the monstrosity needs the angle to be more parallel to the direction of intended air travel. In other words, rotate the blades to have a steep angle of attack, so they push more air back at high speeds, even though they don't push as much air back at low speeds as a compromise. You use what? 10 degrees to 30 degrees of angle? Make that 60 degrees to 80 degrees of angle, or at least 45 degrees to 60 degrees of angle, and I think it will work a lot better. Finding the right angle is the difficult part, and you can probably use simulation software to find the right angle for the blades.
The sound spectrum is 10/10. Also I would like to point out, according to your testing, it's only 0.2 degrees more than the Noctua, but 2.8 db quieter!
Also the T38 has a vortex coming off the back, which definitely aides in keeping the exhaust stream coherent.
The hard 30 messed with my head. I like the sound of the other. The last fan sounded awesome too.
That T33 is dope, I want one.
YES on the spectrum!
You might have 191k subs but now you also have a few fans! Good job! :D
Something to note - hearing is not linear, so it is useful to focus more on higher sound frequencies and compare relative to the other fans.
that t33 is fantastic! i wonder if it is that quiet because those blade tips reduces vortices or if ts down to the blade shape...
I think you ran the shredder in the wrong direction, maybe due to how it was designed but a roof vent is designed to pull air out of a attic like you said. The shredder was pulling air in, maybe if it was spinning reversed or has the blades reversed to pull air/heat away from the radiator instead of push air across it might effect how it works, it may also need to be in an semi enclosed case with openings to draw cold air across the radiator and push it out. From what I learned in a civil class, they are to create a vacuum which and forces cold air to rush in, using a vacuum fan in reverse seems to kinda defeat its purpose in design so I can believe it would cause issues.
Love the haircut, looks really clean
More performance with airodynamic design
Have you done a 3d printed A12x25 fan (or a very close approximation of one)? Would be fun to see how 3d printed parts compare with the same shape to the original material composition.
Or maybe a few different tests with different printed materials. There are some lightweight filaments that foam up when extruded at certain temperatures and it would be cool to see that compare to maybe a copper filled one that is super dense.
Oh my god yes. This is a great idea.
I believe he already did that
The performance is great - but what makes this a real standout is the noise level. The performance-to-noise ratio is IMO the primary measure of a fan, and when it can beat the original noctua by a significant level that is incredible. I really hope a company can step in and do further testing with a proper prototype.
Looking fresh with the clean shave sir!
i feel like you should put the 3d printed version of the a12x25 on the board for a more fair reference
This is true.
Love the new look! Keep It up man!
Great episode! Would love to see a colab with you and Gamers Nexus. They will soon have an actual professional fan testing unit set up in the new studio.
I don't know that many makers that can have 0 dislikes on their videos. Welldone!!!!
With 4,8k likes on the otherside...extremely well done!
That T33 needs to have more testing. Changing the pitch of the blades would be VERY interesting! How much of a difference would a steeper or shallower pitch effect noise and cooling performance? Also, I'd love to see it printed on a Resin printer. I know you've done tests in the past showing no real difference, but you haven't tried it in a while, and not on such a highly performing fan.
You should try to get ahold of Gamers Nexus. See if they are interested in testing these fans. They are setting up a fan testing machine right now. I don't know if they have a 3D printer that they could print them themselves, which if they do, maybe you can get Gamers Nexus in contact with the creator of some of these fans. And if they don't, Maybe you can send them your fans ( or print new ones for them ) with the blessing of the fan creator. Would love to see these ( or at least some of them ) tested on a machine that's used by the industry to validate their own numbers. Also, maybe such a machine would be able to detect the differences in FDM vs Resin prints.
I like the commentary for the fan air performance
The spectrograph is most definitely a welcome addition to the show
This is the first video from this channel I have watched in a while, so I have been a bit out of the loop, but I just have to say, James is rocking going full bald! It honestly looks great paired with the beard.
Love the series! Not to be greedy but something I’d love to see added into future videos would be playing all fan sounds back to back.
I hate to make it sound like the content wasn't awesome, it is and that T33 is sexy, but your editing was really on point, bravo to you good Sir.
T33 lookin' hella fine! Bringing A12x25-like performance to the table, all while being even more silent? I'm sold! Hopefully a company, like Arctic will adapt the design (working together with the original designer of course) and create a new noctua-killer for third of the price.
7:00 I think it would depend on if the multiple frequencies were harmonious or dissonant, as well as where the overall pitch lies. In all cases, I'd think higher pitches would be more annoying, and then starting with most annoying: dissonant frequencies, single frequency, harmonious frequencies.
Though, also, I always find myself humming to harmonize with steady background sounds, so I'd probably be fine with single or harmonious frequencies of a moderate-to-low pitch.
The shredder is the sort of design I would choose of I was going to make design one
Factoring sound floor is such a good idea, probably one of the biggest missing variables!
Just starting this video, I want to say I love this orange color for the fans. I missed the last couple videos so not sure if you've been using it before now.
I just want to see the room/locker/drawer/whatever , where you store all the fans you printed , like it would look amazing , a room having walls filled with different fans, like a fan art gallery lol
When you see multiple peaks at like 40 80 and 160 mhz these are considered multipliers of the same frequency. They sound won't be easily distinguishable. It's basicly like listening to the same chord on a different octave.
Where are the .stl files? I would love to download and print a few of these.
Oh come on!!! I wasn’t expecting that napoleon clip, and I just died laughing 😂 thanks for making me waking up the whole house. Man wtf 😂
Turbulent air is better for heat dissipation than laminar air flow across a heatsink for example. Just thought I should point this out!
Could you get a vane anemometer and build a box to seal between it and the fan to calculate the flow rate (volume flow in M3/s = velocity in m/s * cross sectional area in m2)
Is there an easy way to compare the area under the DB curve? After listening to way too much of The Jay Situation, my understanding is that’s more relevant than peak DB when discussing perceived “loudness”. It’d be interesting to see if it follows with your perception at least
You're looking slimmer than last time I saw a video. Keep at it because its paying off
Damn new style got you looking badass
2:30 should be good for air pressure if nothing else?
why wasnt the tri-mask called the FANtom of the opera
It'd be good to see some iterations of the T33, such as broader/narrower blades and different angle of attack etc.
I'd create 0.5 degree brackets and sort within them by noise. A temp difference of +/- 0.25 degrees is trivial but +3dB is double the sound energy (+10db seems twice as loud). Sound character matters too, of course, but 10.4 dB is *huge*, and probably should matter. Also, on that table the stock fan is listed as S1E3. Have you updated the results with your new test methodology? If not I'd be curious to see how it does with your current gear and techniques as things have aged or been upgraded.
Good point.
do noise-cancelling headphones work better with a monstrosity-type spectrum, or a shredder-type spectrum? or is it more about temporal variability in spectrum?.....
Another FANtastic episode ;)
Awesome video! Maybe you could put a black backdrop on the smoke test so there is more contrast and its easier to see.