If you have been considering adding an 80mm or 92mm fan to your computer case, this article should provide some insight. I compared fans from SilenX, Scythe, and Noctua to see whether the numbers on the boxes are accurate. The competition consisted of two parts: the first part determined the best fan for cooling, while the second part factored in the level of noise. The overall winners blow a decent amount of air without being too noisy.
Testing Methodology
For the first part, I set up a simple structure with a piece of paper hanging in the middle. The fans would face each other and play out a game of “tug of war” – minus the rope, any actual pulling, and that guy who ties the rope to a tree. The fan that blows the paper across the center point and towards the opposing fan wins. If the paper didn’t move, the paper was rotated to avoid paper bias. If that didn’t solve the stalemate, I moved the fans closer. Finally, if that did not result in a clear winner, I went with the one that was quieter. The winner of the paper blowing competition was designated as “Best Blower”.
Testing Rig at Rest
Just in case you were wondering, that is indeed a scrapped printer test page wedged between two hard drive trays, sitting on top of 4 dual SATA hard drive hot-swap boxes.
The fans were powered by a computer power supply from a spare computer. The fans are connected with Molex adapters that came with the fans.
The second part should have involved a decibel reader to determine the noise at different distances. However, I decided to go with a low-tech solution. I compared fans by holding one to each ear at varying distances. By no means is this a scientific approach, but when it comes to annoying noises, a decibel reader can’t judge high- pitched whining.
Terminology
RPM (Revolutions per Minute): The rate at which the fan spins. The higher the number, the more air the fan pushes, and the louder the noise.
dBA (A-weighted decibels): Represents how loud a sound is. The higher the value, the louder the sound.
CFM (Cubic Feet per Minute): Represents the volume of a fluid or gas that flows in a minute. In this case, the higher the value, the more air the fan is pushing out. Cubic Meters per Minute is the metric alternative, but since CFM is more common, I chose it as the standard measurement.
Technical Specifications
80mm Fans
| Brand | Model | RPM | dBA | CFM |
| Scythe | S-Flex SFF80D | 2800 | 32.4 | 40.2 |
| Verax | S0802512LD-3M | 2050 | 21 | 28? |
| Noctua | NF-R8 | 1800 | 17 | 31 |
| SilenX | IXP-52-14 | 1800 | 14 | 24 |
| SilenX | IXP-54-14G | 1700 | 14 | 32 |
| SilenX | IXP-52-11 | 1600 | 11 | 18 |
92mm Fans
| Brand | Model | RPM | dBA | CFM |
| Scythe | S-Flex SFF92D | 2500 | 33.3 | 54.0 |
| Scythe | S-Flex SFF92C | 2000 | 25.3 | 43.2 |
| Noctua | NF-B9 | 1600 | 17.6 | 37.8 |
| SilenX | IXP-64-14 R | 1600 | 14.0 | 42.0 |
| Scythe | S-Flex SFF92B | 1500 | 17.4 | 32.4 |
Noise Levels
Personally, I find that any noise level approaching 25 dBA is too loud for a case fan. I don’t want to be able to hear the system from a few feet away. In an ideal system, I would have to look for status lights.
Paper Blowout Results
The original setup for the tournament-style matchup was symmetrical and elegant, but I had to change things as more fans were added to my test. In addition, I missed out on some key brawls that I knew would have been good matches. To have a comprehensive tournament, I added a bonus round to fill in the gaps.
Bonus Round Results
Thanks to these bonus rounds, I now have a clear idea of what the top 5 fans are in terms of sheer blowing/cooling power.
Final score board
| Rank | Brand | Model | Defeated | Defeated by |
| 1 | Scythe | S-Flex SFF80D | 2: NF-R85: IXP-52-14
4: IXP-54-14G |
N/A |
| 2 | Noctua | NF-R8 | 4: IXP-54-14G3: S0802512LD-3M | 1: S-Flex SFF80D |
| 3 | Verax | S0802512LD-3M | 4: IXP-54-14G5: IXP-52-14 | 2: NF-R8 |
| 4 | SilenX | IXP-54-14G | 5: IXP-52-146: IXP-52-11 | 3: S0802512LD-3M 1: S-Flex SFF80D |
Detailed Results
Round 1
| Fans | Estimated Win Margin | |
| Left: WINNER | Right: LOSER | |
| IXP-52-14CFM: 24 | IXP-52-11 CFM: 18 |
20mmCFM Diff: 6 |
No surprise here
Round 2
| Fans | Estimated Win Margin | |
| Left: LOSER | Right: WINNER | |
| IXP-54-14GCFM: 32 | SFF80D CFM: 40.2 |
87mm CFM Diff: 8.2 |
Literally blew away the other fan
Round 3
| Fans | Estimated Win Margin | |
| Left: LOSER | Right: WINNER | |
| IXP-52-11CFM: 18 | IXP-54-14G CFM: 32 |
22mmCFM Diff: 14 |
An unimpressive win.
Round 4
| Fans | Estimated Win Margin | |
| Left: LOSER | Right: WINNER | |
| S0802512LD-3MCFM: 28(estimate) | NF-R8 CFM: 31 |
150mm: 8mm75mm: 9mm CFM Diff: 3 |
After 4 repeats, removing paper bias, and changing the distance between the fans, I was convinced that these results were accurate.
Round 5
| Fans | Estimated Win Margin | |
| Left: LOSER | Right: WINNER | |
| IXP-54-14GCFM: 32 | NF-R8 CFM: 31 |
150mm: 2mm75mm: 27mm CFM Diff: -1 |
Again, the 54-14G disappoints. I expected a much closer match, but after 4 repeats I still got the same conclusion. It is possible that the CFM is inaccurate, or I got a defective fan.
Round 6
| Fans | Estimated Win Margin | |
| Left: LOSER | Right: WINNER | |
| IXP-54-14GCFM: 32 | SFF80D CFM: 40.2 |
88mm CFM Diff: 8.2 |
A decisive victory for the 80D.
Round 7
| Fans | Estimated Win Margin | |
| Left: WINNER | Right: LOSER | |
| SFF80D CFM: 40.2 |
NF-R8 CFM: 31 |
58mm CFM Diff: 9.2 |
Predictable results here, but I was happy to see the Noctua put up more of a fight than the SilenX IXP-54-14G.
Bonus Round 1
| Fans | Estimated Win Margin | |
| Left: LOSER | Right: WINNER | |
| IXP-52-14CFM: 24 | IXP-54-14G CFM: 32 |
150mm: 20mm75mm: 21mm CFM Diff: 8 |
The 54-14G sought redemption against a fellow SilenX fan. It won both on paper, and against the paper. I was expecting a more convincing win, though.
Bonus Round 2
| Fans | Estimated Win Margin | |
| Left: WINNER | Right: LOSER | |
| S0802512LD-3MCFM: 28(estimate) | IXP-52-14CFM: 24 | 150mm: 23mm75mm: 21mm CFM Diff: 4 |
The 52-14 predictably lost again, but the win margin is greater than the first bonus fight even though the CFM difference is smaller.
Bonus Round 3
| Fans | Estimated Win Margin | |
| Left: WINNER | Right: LOSER | |
| I S0802512LD-3MCFM: 28(estimate) | IXP-54-14GCFM: 32 | 150mm: 10mm75mm: 20mm CFM Diff: -4 |
The IXP-54-14G underperformed yet again, despite having a larger CFM value.
Conclusion
Overall the stats on the boxes are roughly accurate. Although this isn’t a very scientific test, the simplicity of a face-off makes the results pretty clear. If this tournament were to be repeated, I would measure dBA with a decibel reader. While Scythe was clearly the best performer in terms of cooling power, Noctua was the quietest.
