Did your analysis miss Corsair's success compared to Swiftech?
Did your analysis miss Corsair's success compared to Swiftech?
Hey everyone,
My friend and I developed a system together.
He used an i7 5820k with Corsair H105, pull/exhaust setup.
I chose a 5930k with Swiftech H-240x, also pull/exhaust.
He's clocking 4.7ghz @ 30c, while I'm at 4.4ghz @ 33c.
What's going on? Did I make a mistake? Am I overheating more than expected because of the chipset? Should I consider adding another fan for partial push/pull (Swiftech only supports three fans)?
Thanks.
Did you experience a crash, or was the system shutdown due to Occt?
Usually, the shutdown temperature is around 85°C.
It's typical for cores to reach varying temperatures.
If the CPU failed, such as showing a blue screen, then your Oc parameters might not have functioned correctly.
Finding idle is tricky; he might start with fewer apps running than you do.
https://www.techpowerup.com/reviews/Swif...0-X/6.html
The Swiftech and H105 perform well when overclocked, but at normal settings the Swiftech often leads.
haven't performed any stress testing yet (not sure what to use TBH).
i'm just weighing whether to keep or return the swift tech. the open loop isn't something i'm particularly interested in; though it seems everyone was recommending it as the better option compared to Corsair and NZXT.
i'm puzzled about how he's running more OC'd and cooler. i attempted to push mine up to 4.7 and the system shut down... uhm? honestly, i'm not sure what I'm doing. i'm depending on the Asus auto tuning and then just increasing the clock speed. i think this might be incorrect. he did the same, ran the auto-tuning and it reached 4.7; whereas mine went to 4.5. odd.
i have a single 200mm intake, a 120mm exhaust and dual 140mm exhausts on the rad. i could add another 140mm intake and swap the 200mm intake for dual 140's. is it worth making any of those changes?
Don't auto tune at high clocks, that's your initial error.
Autotune will push the system as much as it can, but the outcomes won't be uniform. Various chips demand different amounts of power to hit specific clock speeds. Some may need 4.7 at 1.25v while others require 1.30 or higher. Autotune doesn't know if a chip needs more or less power. It only sets the voltage based on its assumptions.
It's not as problematic as it seems. Asus has extensively experimented with overclocking and applied statistical methods to create boards that function optimally at each clock speed. This approach doesn't guarantee top speeds, but anything above 4.5 is a good chance. 4.7 is achievable on many chips, though some will need significantly more voltage.
Are those idle temperatures normal? What are the stress test temperatures with occt? Do you both use the same case? Cooling mainly depends on how much clean air you can intake. By the way, I think your friend got a golden chip if he achieves 4.7; only around 11% can reach that. I don’t have any data for the 5930k.
We both started from scratch and are learning online, which isn't always simple. It's strange I upgraded to version 4.4 while he went to 4.7, and my chip seems a bit better now.
Idle times indicate whether your cooler is installed correctly. Temperatures above 10-15°C from ambient are normal.
You're doing well.
The single 200mm fan for intake is preferred—it delivers strong airflow at lower speeds and operates more quietly.
For stress testing, I prefer OCCT. It follows more typical everyday commands and stops the test at 85°C. At 100°C, the CPU will slow down or shut down to prevent heat-related damage. Don’t worry about that.
While testing, focus on monitoring vcore. It should stay below 1.35v for safe continuous use.