I have overclocking questions
I have overclocking questions
My specifications are reflected in my signature. The auto volts set by the bios is 1.04. At around 72 degrees with about 30 minutes on prime95, it reaches 100 degrees when increased to 1.25 volts and 4.5 ghz. Lowering it to 1.20 volts brings it to 92 degrees. I’m curious about the general connection between voltage and core clock speed. I aim for at least 4.2 and want to confirm if 1.1 volts is sufficient. How can I tell if it meets the required speed? I currently have a Hyper 212 EVO, but I’m hoping for a water cooler in the future. Thanks.
Typically, the connection between voltage (v), temperature © and speed (ghz) follows an exponential pattern as speed increases. In simpler terms, more voltage is needed to achieve smaller speed gains. This process generates heat as a result. There are various techniques to enhance speed performance, with maintaining cooler chip temperatures being a common approach. Honestly, the 212 Evo offers a strong cooling solution, particularly considering its price point. Improved water cooling could further boost cooling efficiency and potentially lead to more stable speeds, though sometimes success depends on other factors.
When increasing performance, you typically start by fine-tuning the core clock and checking for stability. If the system holds up during the stress test, the voltage settings seem appropriate at that speed. Continue raising the core clock until instability appears, then adjust the Vcore accordingly. This process also depends on your ability to manage heat output. Since each CPU behaves differently, solutions that suit one model may not apply to another. Therefore, follow this method to find what suits your specific CPU. Also keep in mind that Prime 95 can indicate high temperatures well, but it doesn't always reflect overall stability. I've observed stable Prime 95 experiences during gaming or browsing with Chrome, even when idle. Test with Prime 95 and then run the computer for a while doing regular tasks. If it runs smoothly, you can safely increase further. Avoid rushing the overclocking process.
I wanted to find out if it's stable before moving on to the next core speed.
There are several programs you can push to test (OCCT Prime 95, Hyper PI, etc) using various workloads. ASUS has developed RealBench, offering a range of different types of tasks. I support this method. It’s not just one workload type, which helps avoid unrealistic expectations. You wouldn’t be surprised if someone insisted they could run Prime 95 for days without issues, only to experience a BSOD after playing BF4 for ten minutes. They would point to the game as the cause and dismiss any overclocking concerns.
By stressing it with different loads, you become more confident in its stability. It’s also wise to stay just above the voltage where you find stability. For example, if you chose a final overclock of 4.4GHz and found 1.250V to be stable, any lower voltage wouldn’t work, so you’d adjust to 1.255 or 1.260V for consistent performance.
Typically, the connection between voltage (v), temperature © and speed (ghz) follows an exponential pattern. In simpler words, more voltage is needed to achieve smaller speed gains. As a result, heat is produced.
There are various techniques to enhance speed performance, with cooling being the most common approach. Honestly, the 212 Evo provides a very effective cooler, particularly considering its price point. Using water cooling would boost cooling efficiency, likely improving consistent speed output, though sometimes success depends on other factors.
The standard overclocking process involves these steps:
1) Lower the voltage and disable turbo mode.
2) Increase the multiplier setting.
3) Perform a stability check and keep temperatures under control; if successful, proceed to 2).
4) If stability fails, increment voltage slightly and repeat step 3, or adjust accordingly.
5) Reduce voltage and multiplier by one each time.
6) Conduct another stability test to confirm performance.
This method helps identify the optimal speed with a stable average boost. Settings to avoid include:
1) Voltage – ideally capped at around 1.30v for sustained operation.
2) Temperatures – maintaining between 70C and 80C is recommended.
There are several programs you can challenge with (OCCT Prime 95, Hyper PI, etc) that simulate various workloads. ASUS has developed RealBench, which offers a range of different task types. I support this method. It’s not just one workload type, which helps avoid unrealistic expectations. People often claim they can run Prime 95 for days without issues, but they experience a BSOD after playing BF4 in just 10 minutes. They point to the game as the cause and can’t understand how their overclock might be involved. This makes it hard for them to accept that stability could come from something else. The more you push the system with different loads, the more confident you’ll feel about its reliability. Also, aim for just a bit above the voltage where you see stability. For example, if you found 4.4GHz stable at 1.250V, lowering it to 1.255 or 1.260V would be better for consistent performance.
Is it possible to find links for some of those tests you mentioned?
Typically, the connection between voltage, temperature, and speed shows an exponential trend as speed increases. In simpler words, more voltage is needed to achieve smaller speed gains. The result is heat production. There are various techniques to boost performance, but cooling the chip remains the most common approach. The 212 Evo offers a reliable cooler for its cost. Enhanced water cooling could improve cooling efficiency and potentially stabilize speed, though sometimes success depends on the hardware. The standard overclocking procedure involves adjusting settings step by step while monitoring temperatures. If stability is achieved without exceeding 80°C, proceed; otherwise, adjust accordingly. Be cautious with voltage and temperature limits to avoid issues like BSOD or crashes.
There are several programs you can push to test their limits (OCCT Prime 95, Hyper PI, etc.) that work with various workloads. ASUS has developed RealBench, which offers a range of different types of tasks. I support this method. It’s not just one workload type, which helps avoid unrealistic expectations. People often claim they can run Prime 95 for days without issues, but they end up with a BSOD after just 10 minutes playing BF4. They point the finger at the game and can’t understand how their overclock might be the cause. Even if you convince them to go back to stock and test, they usually won’t accept the results and think something else is responsible.
By stressing the system with different loads, you become more confident in its stability. Also, aim for just a bit above the voltage where you find stability. For example, if you chose 4.4GHz as your final overclock and found stability at 1.250V, lowering it to 1.255 or 1.260V would be ideal for consistent performance.
Can you share links to some of those tests you mentioned?
RealBench
ICCT
LinX
Intel® Extreme Tuning Utility
Here are a few options to begin with. For stability checks, anything that deviates from normal behavior indicates instability. This includes software crashes, BSODs, unexpected restarts, etc. In my experience, one of the most frequent BSODs linked to overclocking usually points to a need for increased Vcore. The WHEA BSOD is particularly common in this context.