Jump to content

rYzYr

Member
  • Posts

    16
  • Joined

  • Last visited

Awards

This user doesn't have any awards

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

  1. “It's more like I just read the room wrong.” **Trust me, bro.** You’re still doing it.
  2. https://electronics.stackexchange.com/questions/3343/how-to-become-an-embedded-software-developer
  3. Why not do both? Companies like Amazon hire software development engineers for low level development that involves working closely with hardware. This can be devices like the Amazon Echo or servers and networking gear used by AWS. Here's one job posting from Amazon Go that combines both interests: https://www.amazon.jobs/en/jobs/651770/senior-embedded-software-engineer
  4. I tested the Cryorig M9i and Cooler Master Hyper 212 Evo. Unfortunately, the Cryorig M9i caused low RPM errors with an RPM of less than or equal to 500. While a sensor or some part of the motherboard may be at fault, this result contradicts the specified range of 600 - 2200 RPM +/- 10%. The Cooler Master Hyper 212 Evo performed significantly better with an initial reading of 600 RPM, a top end of 1000 RPM, and a max CPU temperature of 61 degrees C when stressed using the following test command: sudo stress --cpu 8 --timeout 100000 The Cooler Master Hyper 212 Evo at 100% CPU utilization is quieter than the stock cooler at idle (980 RPM). FWIW, here’s a review of the M9i that compares it to many other popular coolers: https://www.techpowerup.com/reviews/CRYORIG/M9i/
  5. Based on my testing, I'm going to stick with the GP-12. As for the CPU cooler, I wasn't able to find the low end RPM for the be quiet! Pure Rock Slim. The Intel E97378-003 runs at 2300 RPM at 100% CPU. I tested the Noctua NH-U12S and it wanted to stay at the minimum of 300 RPM, so I assume whatever tower cooler I might use would operate at the low end at 100% CPU. A switch to something with a 600 RPM minimum could net a 50% reduction in noise. Based on that, I could come out way ahead with something like the Cooler Master Hyper 212 Evo (http://www.coolermaster.com/cooling/cpu-air-cooler/hyper-212-evo/).
  6. The minimum RPM alarm causes the fan to ramp to 100%. It will then lower as the temperature is measured and eventually go too low once more. It cycles the fan RPM. See this note from Noctua on fan issues with Supermicro: https://noctua.at/en/im-experiencing-fan-speed-issues-with-my-motherboard-from-supermicro-what-can-i-do. The error does not prevent the system from posting or otherwise functioning, but it adds noise to an error log that could mask real fan issues.
  7. The NF-P12 1700 RPM has a minimum of 450 RPM according to the link provided, so it's still too low. Remember, even if the fan has a stated minimum of 550, there's a margin of error typically around 10%, so a realistic minimum that won't cause alarms is more like 600 RPM. Thanks for all the comparative information. I'll spend some time and read through it. It would be nice to see a comparison filtered on the minimum RPM being 600+ as most (all?) of the highly rated fans will not be compatible with Supermicro motherboards. The question isn't whether the Nanoxia's are among the best, it's whether they are quieter than the Fractal Design Dynamic X2 GP-12 or Supermicro FAN-0124L4. Regardless, my testing is showing the CPU cooler fan to be the bigger problem. And, even if you could eliminate a CPU fan, my PSU (Seasonic S12II 620W - https://seasonic.com/s12ii) is arguably noisier than the GP-12. The question then becomes fixed RPM versus the potential energy savings of a PWM. At a cost of around $1.28/yr (https://www.maximintegrated.com/en/design/tools/calculators/general-engineering/energy-cost.cfm), spending $20 - $30 on a PWM fan is not a financially good decision. Whether an aftermarket CPU cooler like the Cryorig M9i (min RPM of 600 +/- 10%) is less noisy than the stock cooler and worth the expense is the question that really needs to be answered. Have you guys done any comparisons of aftermarket coolers to the stock coolers from Intel and AMD?
  8. The provided links are to a third party website and not to the manufacturer. NF-P12 PWM has a minimum of 300 RPM. I’ve witnessed the RPM bounce that is typical of Noctua fans with Supermicro due to the low RPM. https://noctua.at/en/nf-p12-pwm/specification I did some digging and here’s what I’ve found: Nanoxia Deep Silence 120mm PWM 650 - 1500 RPM, 16.9 dB at max 67.3 CFM durability: 80,000 hours http://nanoxia-world.com/en/products/fans/deep-silence-pwm/247/deep-silence-120-mm-pwm?c=52 Nanoxia Deep Silence 140mm PWM 700 - 1400 RPM, 16.2 dB at max 76.5 CFM durability: 80,000 hours http://nanoxia-world.com/en/products/fans/deep-silence-pwm/248/deep-silence-140-mm-pwm?c=50 Both cost around half the Supermicro FAN-0124L4, have higher CFM ratings, and provide superior acoustics. Supermicro doesn’t specify the durability.
  9. Thank you for the suggestions. The NF-P12 has a 3-pin connector, but would be interesting with the low noise adapter--bringing the RPM down to 900. be quiet! hasn't published the low end RPM (https://www.bequiet.com/en/casefans/449), so it's not clear if it exceeds the 500 RPM alarm level.
  10. Yes. I was able to remove the front fan with no noticeable increase in temps. I'm currently using the stock Intel cooler that comes with the E3 v6 CPUs. The stock case fan is decent (after removing the second fan), but I would prefer a quieter option. I'm also looking at passively cooling the CPU. In fact, I was able to passively cool a E3-1270v6 with the stock cooler heat sink (sans fan) for light use (temps around 70). I suspect switching to a tower heatsink will make running at full load with the single case fan a possibility.
  11. The stock case fans from Fractal Design (Dynamic X2 GP-12 PWM) have 3-pin connectors and the Supermicro boards have 4-pin PWM headers. While I could stick with the stock fan, I'm hoping a PWM fan would run at less than 1200 RPM at full load, the max for the stock fan. It's unclear whether I can set a lower fixed RPM with the Supermicro firmware, so I should also look into that.
  12. I believe the 400 RPM minimum will cause errors with Supermicro (seems to require at least 500 RPM).
  13. I have a Supermicro X11SSH-LN4F housed in a Fractal Design Define Mini C. The cases comes with two Dynamic X2 G-12 3-pin fans. I removed the front fan. I’d now like to replace the rear fan with a compatible PWM fan but I’m concerned the FAN-0124L4 may be too loud Are there other more quiet 120mm options?
  14. I was wrong. The Noctua NH-U12S CPU cooler is not compatible with the Supermicro motherboard. Intel motherboards only support the stock cooler and alternatives from Supermicro appear to provide no advantages over the stock cooler in large form factor cases.
×