Impressively Bad: Corsair A500 CPU Cooler Review & Benchmark
video description
Date: 2020-05-06
Related videos
Comments and reviews: 10
Royameadow
It's very fortunate that the lot of us were already aware about the troubles that the A500 has, when compared to the NH-D15 and Assassin III, even before it launched: The design concept, while interesting and allows for a bit more ease of access and customization, still turns out to be a Brick of a cooler; the NH-D15, Assassin III, and Be Quiet! 's Dark Rock Pro 04 all manage to stay within or below 07 inches in X Axis Width and don't have much of an issue on the Z Axis, which when compared to the A500, which makes Noctua, Deepcool, and Be Quiet! look more suitable for an Air Cooled HEDT system in a Micro ATX setup, so it's clear that Corsair made their cooler with ATX, Extended ATX, and Xtra Large ATX in mind, even in spite of how audible it still is no matter the case one uses. Personally, I was waiting for your insight on these Air Coolers for a while, Steve; as I've mentioned on plentiful occasions, I'm still in the middle of turning my eMachines ET1331G-07w into a Ryzen 3950X/Radeon VII Sleeper PC (presently, I still have to save up at least 2000 USD to get the second half of the parts, which includes the 3950X, ECC Memory, CPU Cooler, and ASRock X570M Pro4; the remainder will go to extra fans and storage, and the biggest issue that I have had to worry about is actually the CPU Cooler, because I wasn't absolutely sure on which one would assure optimal temperatures and stability as the cooler ages; in this case (pardon the pun, it seems as if the NH-D15 (especially the Chromax Black variant, which interestingly is not presented here for this review, though I'm sure it'll show in future benchmarks) is the ideal choice and likely will remain to be for the foreseeable future, the ET1331G case by eMachines doesn't have a lot of clearance and airflow as is for Vertical/Tower Coolers and it's good to know that my judgement on going for the NH-D15: Chromax Black will not be for naught; surely, the other companies will catch up to it, but it is very clear that Noctua will be the Go To until a more compact solution can compete against the NH-D15 with similar thermals and acoustics, only time will tell. (:
reply
It's very fortunate that the lot of us were already aware about the troubles that the A500 has, when compared to the NH-D15 and Assassin III, even before it launched: The design concept, while interesting and allows for a bit more ease of access and customization, still turns out to be a Brick of a cooler; the NH-D15, Assassin III, and Be Quiet! 's Dark Rock Pro 04 all manage to stay within or below 07 inches in X Axis Width and don't have much of an issue on the Z Axis, which when compared to the A500, which makes Noctua, Deepcool, and Be Quiet! look more suitable for an Air Cooled HEDT system in a Micro ATX setup, so it's clear that Corsair made their cooler with ATX, Extended ATX, and Xtra Large ATX in mind, even in spite of how audible it still is no matter the case one uses. Personally, I was waiting for your insight on these Air Coolers for a while, Steve; as I've mentioned on plentiful occasions, I'm still in the middle of turning my eMachines ET1331G-07w into a Ryzen 3950X/Radeon VII Sleeper PC (presently, I still have to save up at least 2000 USD to get the second half of the parts, which includes the 3950X, ECC Memory, CPU Cooler, and ASRock X570M Pro4; the remainder will go to extra fans and storage, and the biggest issue that I have had to worry about is actually the CPU Cooler, because I wasn't absolutely sure on which one would assure optimal temperatures and stability as the cooler ages; in this case (pardon the pun, it seems as if the NH-D15 (especially the Chromax Black variant, which interestingly is not presented here for this review, though I'm sure it'll show in future benchmarks) is the ideal choice and likely will remain to be for the foreseeable future, the ET1331G case by eMachines doesn't have a lot of clearance and airflow as is for Vertical/Tower Coolers and it's good to know that my judgement on going for the NH-D15: Chromax Black will not be for naught; surely, the other companies will catch up to it, but it is very clear that Noctua will be the Go To until a more compact solution can compete against the NH-D15 with similar thermals and acoustics, only time will tell. (:
reply
Djhg2000
Once again we're stumbling on the limit of pseudo science. That surface levelness scatter plot is misleading as it only shows absolute delta values, making the points impossible to compare between each other. A smooth concave surface would be impossible to distinguish from a convex surface, or even a brushed surface. You need to use actual delta values instead of absolute deltas. Then there's the testing methodology for the sound pressure level. While it's better than what most reviewers do you're only measuring noise as a scalar value in an uncontrolled environment. You need to use either a sound dampening chamber emulating infinite distances to nearby walls or an echo chamber where you're effectively measuring the echo. You should also present the full spectrum instead of the single dB(A) scalar, because while the dB(A) scale does a good job at summarizing how loud the noise is it will tell you nothing about how annoying it is. A calibrated measurement microphone and a decent audio interface isn't that expensive (a basic good enough quality setup would be a few hundred USD, we're not building a professional audio recording studio, and in combination with some simple free software like REW and Audacity you get a very reliable and comprehensive spectrum plot. This will allow you and your viewers to quickly identify the character of the noise, and avoid pitfalls such as how low amplitude pink noise and a single high amplitude sine wave can look identical when you're looking at the dB(A) scalar alone. Remember to show the spectrum of ambient noise as well, you can calibrate away the other stuff but we need to know what the ambient noise looks like so that we can tell what the cooler actually contributes. This is particularly important if you present the spectrum as dB(A) instead of SPL. The other charts look good to me though. dB(A) is fine after you've established what the spectrum looks like. Nice job on those and thank you for making an effort to improve the quality of cooler reviews.
reply
Once again we're stumbling on the limit of pseudo science. That surface levelness scatter plot is misleading as it only shows absolute delta values, making the points impossible to compare between each other. A smooth concave surface would be impossible to distinguish from a convex surface, or even a brushed surface. You need to use actual delta values instead of absolute deltas. Then there's the testing methodology for the sound pressure level. While it's better than what most reviewers do you're only measuring noise as a scalar value in an uncontrolled environment. You need to use either a sound dampening chamber emulating infinite distances to nearby walls or an echo chamber where you're effectively measuring the echo. You should also present the full spectrum instead of the single dB(A) scalar, because while the dB(A) scale does a good job at summarizing how loud the noise is it will tell you nothing about how annoying it is. A calibrated measurement microphone and a decent audio interface isn't that expensive (a basic good enough quality setup would be a few hundred USD, we're not building a professional audio recording studio, and in combination with some simple free software like REW and Audacity you get a very reliable and comprehensive spectrum plot. This will allow you and your viewers to quickly identify the character of the noise, and avoid pitfalls such as how low amplitude pink noise and a single high amplitude sine wave can look identical when you're looking at the dB(A) scalar alone. Remember to show the spectrum of ambient noise as well, you can calibrate away the other stuff but we need to know what the ambient noise looks like so that we can tell what the cooler actually contributes. This is particularly important if you present the spectrum as dB(A) instead of SPL. The other charts look good to me though. dB(A) is fine after you've established what the spectrum looks like. Nice job on those and thank you for making an effort to improve the quality of cooler reviews.
reply
AlphaLion
I really love your in-depth reviews about everything, not just some quick thermals like most other content creators, which in comparison seem a lot less credible/reliable. You're doing a lot of great work for the community sorting out terrible and good products that might not seem bad at first or without taking it apart. I wonder, will you test Arctic's Freezer 34 air cooler? I have the 33 wit dual fans and it's surprisingly great for a very low price. I can't do noise measurements, but it's quiet enough for my taste, it has good cooling capability as far as i can tell and it looks good as well with great memory clearance. I was kind of disappointed that almost no big name youtubers reviewed it when it launched a few years ago with the exception of Bitwit. Now the 34 is here and user reviews say it's pretty good, so i'm really interested to see what you can find on it. So far it looks like a great cooler for a ridiculous price, like 35? The 33 was about 50 (including tax) when i got it 2 years ago, even that is a great price in my opinion
reply
I really love your in-depth reviews about everything, not just some quick thermals like most other content creators, which in comparison seem a lot less credible/reliable. You're doing a lot of great work for the community sorting out terrible and good products that might not seem bad at first or without taking it apart. I wonder, will you test Arctic's Freezer 34 air cooler? I have the 33 wit dual fans and it's surprisingly great for a very low price. I can't do noise measurements, but it's quiet enough for my taste, it has good cooling capability as far as i can tell and it looks good as well with great memory clearance. I was kind of disappointed that almost no big name youtubers reviewed it when it launched a few years ago with the exception of Bitwit. Now the 34 is here and user reviews say it's pretty good, so i'm really interested to see what you can find on it. So far it looks like a great cooler for a ridiculous price, like 35? The 33 was about 50 (including tax) when i got it 2 years ago, even that is a great price in my opinion
reply
JpHDev
Lmfao. 10 grand! PLUS power supplies. You are bloody crazy. You were robbed. A total wasre of money. I told you before, 400 dollars would be a stretch for that rig. It is a freeking stabilised hot plate (10 dollars made from an ihs and load resistors and reflective undercasing which is the easiest thing to make and a few 5 dollar sensors. The psu needed is about 50 dollars max. It can be any rough arsed switched mode box or two as external simple circuitry will control what reaches the plate. Voltage sensing can be done by any reasonable adc such as a 12 bit bit on ths DUE. Current sensing is a piece of piss with active senso units such as alegro. The only thing needed is to control the voltage and current which is also a piece of piss to do externally frm the cpu. The control logic is less than 15 dollars. You basically have no idea what you are doing. This brings into question any of your results. You were conned. scammed.
reply
Lmfao. 10 grand! PLUS power supplies. You are bloody crazy. You were robbed. A total wasre of money. I told you before, 400 dollars would be a stretch for that rig. It is a freeking stabilised hot plate (10 dollars made from an ihs and load resistors and reflective undercasing which is the easiest thing to make and a few 5 dollar sensors. The psu needed is about 50 dollars max. It can be any rough arsed switched mode box or two as external simple circuitry will control what reaches the plate. Voltage sensing can be done by any reasonable adc such as a 12 bit bit on ths DUE. Current sensing is a piece of piss with active senso units such as alegro. The only thing needed is to control the voltage and current which is also a piece of piss to do externally frm the cpu. The control logic is less than 15 dollars. You basically have no idea what you are doing. This brings into question any of your results. You were conned. scammed.
reply
Dylan
Just to increase the pedantry to breaking point: Is levelness the correct/useful term here? Wouldn't flatness be more relevant? If it was off level (slanted) it would likely even out by keeling over till it was level with the IHS. Flatness on the other hand being how much the bottom deviates on a large scale from a plane- tilted or otherwise seems like the relevant measure? Roughness being almost irrelevant (small scale deviations from a plane) While the meter can almost certainly measure levelness in addition to flatness, I don't think it's necessarily relevant here. Pedantry over: p
reply
Just to increase the pedantry to breaking point: Is levelness the correct/useful term here? Wouldn't flatness be more relevant? If it was off level (slanted) it would likely even out by keeling over till it was level with the IHS. Flatness on the other hand being how much the bottom deviates on a large scale from a plane- tilted or otherwise seems like the relevant measure? Roughness being almost irrelevant (small scale deviations from a plane) While the meter can almost certainly measure levelness in addition to flatness, I don't think it's necessarily relevant here. Pedantry over: p
reply
consummateVssss
The noise normalized comparisons is the best methodology in fan reviews so far. While dBA is weighted towards human hearing, it is still just an average noise power, which doesn't account for weird tonal noises that might occur at specific peaks. One fan at 35 dBA could have a smooth frequency spectrum while another at 35 dBA has a sharper, higher peak noise level that is actually perceived as a louder noise. Is there anyway you could also include normalization of frequency spectrum, such as using Noise Criteria (NC) Curves, to better account for perceived noise.
reply
The noise normalized comparisons is the best methodology in fan reviews so far. While dBA is weighted towards human hearing, it is still just an average noise power, which doesn't account for weird tonal noises that might occur at specific peaks. One fan at 35 dBA could have a smooth frequency spectrum while another at 35 dBA has a sharper, higher peak noise level that is actually perceived as a louder noise. Is there anyway you could also include normalization of frequency spectrum, such as using Noise Criteria (NC) Curves, to better account for perceived noise.
reply
Andrew
Gamers Nexus, 1) Maybe a x-y-z 3D scatter plot of the surface heights instead of just a scatter? Extra points for making your x-y points actual locations of the measurements on the cold plate. 2) Would it make sense to do a second round of testing using a reference fan? I agree dBA controlled numbers with stock fans are useful, but then again so would numbers using a good reference fan. SPCR was doing this sort of testing years ago, though they were far more focused on sound, down to building a sound chamber with a 10dBA floor (i think.
reply
Gamers Nexus, 1) Maybe a x-y-z 3D scatter plot of the surface heights instead of just a scatter? Extra points for making your x-y points actual locations of the measurements on the cold plate. 2) Would it make sense to do a second round of testing using a reference fan? I agree dBA controlled numbers with stock fans are useful, but then again so would numbers using a good reference fan. SPCR was doing this sort of testing years ago, though they were far more focused on sound, down to building a sound chamber with a 10dBA floor (i think.
reply
ShroudedWolf51
Another concern that I had since I've heard about this cooler is just how durable are these rails that the fans use to slide up and down the cooler? With a metal clip, if one breaks or warps or whatever, you can replace it easily enough. With these, how long am I going to be out of action if I (hypothetically) buy the A500 and then the fans stop attaching. say, two years down the line? A heatsink with no fan is not a great experience for how I use my desktop.
reply
Another concern that I had since I've heard about this cooler is just how durable are these rails that the fans use to slide up and down the cooler? With a metal clip, if one breaks or warps or whatever, you can replace it easily enough. With these, how long am I going to be out of action if I (hypothetically) buy the A500 and then the fans stop attaching. say, two years down the line? A heatsink with no fan is not a great experience for how I use my desktop.
reply
bobiseverywhere
I think you need to try out Thermaltake' old line of CPU coolers that really make a case for jet engine. i think the last one i had was the volcano 11, put that thing at 100% fan and i am pretty sure it was much higher dba than these. Would be funny to see if you could rig up some old cooling solutions with some jenky setup just for a fun piece and see how they would fair on modern hardware if anyone out there still has one to give you for testing.
reply
I think you need to try out Thermaltake' old line of CPU coolers that really make a case for jet engine. i think the last one i had was the volcano 11, put that thing at 100% fan and i am pretty sure it was much higher dba than these. Would be funny to see if you could rig up some old cooling solutions with some jenky setup just for a fun piece and see how they would fair on modern hardware if anyone out there still has one to give you for testing.
reply
videos,
I really hope you'll review the Thermalright Le Grand Macho RT, I've been using it because it's supposed to be comparable to the D15 and looked better/was more compatible. It's interesting to me because it's not a dual tower but a f off huge stack of aluminum with a fan attached I guess it's harder to test in a relevant way because in a lot of cases it would benefit from the exhaust fan, but it's also viable without a fan which could be interesting.
reply
I really hope you'll review the Thermalright Le Grand Macho RT, I've been using it because it's supposed to be comparable to the D15 and looked better/was more compatible. It's interesting to me because it's not a dual tower but a f off huge stack of aluminum with a fan attached I guess it's harder to test in a relevant way because in a lot of cases it would benefit from the exhaust fan, but it's also viable without a fan which could be interesting.
reply
Add a review, comment
Other channel videos
