Overclocking i5 4670k concern
Overclocking i5 4670k concern
I chose to experiment with my CPU's overclocking. As mentioned, it's an Intel i5 4670K paired with a Cooler Master Hyper 212 Evo cooler.
I adjusted the multiplier to 44 (which equals 4.4Ghz) and set the voltage to 1.250, then performed a stability test on Prime95 using the blend preset.
Initially everything worked fine, with CPU temperatures around 55-60°C and power consumption at 134W. However, after about 30 minutes, some unusual behavior appeared.
Is this typical? (see image: https://snag.gy/5ECdqW.jpg)
Thank you ahead of time.
Andaja,
with respect, you're using the incorrect Prime95 version and likely the wrong testing method (Blend), which is why your core temperatures initially stabilized but then rose sharply, along with high Vcore and Core Watts readings. You seem to be running the most recent release (29.4), though the recommended setup is 26.6, and the appropriate test should be Small FFT's. The issue lies here:
2nd through 8th generation i3, i5, and i7 CPUs support AVX instruction sets. Prime95 versions older than 26.6 execute AVX code on the CPU's floating-point unit (FPU), leading to temperatures up to 20°C higher. Other tools might show similar behavior.
andaja :
I chose to experiment with my CPU's overclocking. Following the guidelines, I used an Intel i5 4670k paired with a Cooler Master Hyper 212 Evo cooler. I adjusted the multiplier to 44 (which gives 4.4Ghz) and set the voltage to 1.250, then performed a stability test on Prime 95 using the blend preset. Initially, everything worked fine—CPU temperatures stayed between 55-60°C and power consumption around 134W. However, after about 30 minutes, some unusual behavior appeared.
Is this typical? (see image: https://snag.gy/5ECdqW.jpg)
Thank you ahead of time.
It's been a while since I worked with a Haswell chipset, but unless there have been any updates, Haswell models seem to struggle with Prime95, possibly due to a code bug. The temperatures are much higher than expected, so double-check your readings with another stress test.
Also, the 212 cooler is on the lower side, and a multiplier of 4.4 might be too high for that heat sink. Some processors can handle it better. Have you turned off adaptive voltage? That could cause damage. Did you increase both the uncore multiplier and voltage? That would help stabilize the overclock.
andaja, Respectfully, you're running the wrong Prime95 version, and probably the wrong torture test (Blend), which is why your Core temperatures started OK then went high, and why your Vcore and Core Watts are also high. You're probably running the latest version (29.4), however, the correct version to run is 26.6, and the correct torture test to run is Small FFT's. Here's the problem: 2nd through 8th Generation i3, i5 and i7 CPU's have AVX (Advanced Vector Extension) Instruction Sets. Prime95 versions later than 26.6 run AVX code on the CPU's Floating Point Unit (FPU) which causes unrealistic temperatures up to 20°C higher. Certain other utilities may have similar results. • Prime95 v26.6 - http://www.mersenneforum.org/showthread.php?t=15504 AVX can be disabled in Prime95 versions later than 26.6 by inserting "CpuSupportsAVX=0" into the "local.txt" file in Prime95's folder. However, since Core temperatures will be the same as 26.6, it's easier to just use 26.6. AVX doesn't affect Core i 1st Generation, Core 2, Pentium or Celeron processors since they don't have AVX Instruction Sets. If you’re overclocked and run AVX apps such as for rendering or transcoding, you may need to reduce Vcore and Core speed or upgrade your cooler and case fans so Core temperatures don’t reach 85°C. Many 6th, 7th and 8th Generation motherboards address the AVX problem by providing offset adjustments in BIOS. An offset of -2 or -3 (200 or 300 MHz) is usually sufficient. 4th Generation Haswell processors have no AVX offset adjustments in BIOS. Asus RealBench runs a realistic AVX workload typically within +/- a few % of TDP, and is an excellent utility for testing overall system stability, whether you're overclocked or not. RealBench is also the utility that Silicon Lottery uses for testing CPU stability. • Asus RealBench - http://rog.asus.com/rog-pro/realbench-v2-leaderboard/ Here’s a sample of thermal and stability test utilities shown according to % of TDP at Default BIOS settings: TDP … Thermal Test - Steady Workload 129% … Prime95 v27.7 through v29.4 - Small FFT’s (AVX, No Offset) 101 % <-- Prime95 v26.6 - Small FFT’s 89% … HeavyLoad v3.4.0.234 - Stress CPU 87% … FurMark v1.19.1.0 - CPU Burner 78% … CPU-Z v1.82.0 - Bench - Stress CPU 66% … AIDA64 v5.95.4500 - System Stability Test - Stress CPU 55% … Intel Processor Diagnostic Tool v4.0 - CPU Load TDP … Stability Test - Fluctuating Workload (Peak) 123% … OCCT v4.5.1 - CPU: OCCT (AVX, No Offset) 118% … LinX v0.6.5 - Default 116% … IntelBurn Test v2.54 - High 113% … OCCT v4.5.1 - CPU: Linpack (AVX, No Offset) 110% … AIDA64 v5.95.4500 - System Stability Test - Stress FPU 99 % <-- Asus RealBench v2.56 - Stress Test (AVX, No Offset) 94% … Sandra 2017.09.24.41 - Burn in - Processor Tests 92% … CineBench v15.0 - CPU - Render Test 79% … Intel Extreme Tuning Utility v6.4.1.15 - CPU Stress Test All tests will show 100% CPU Utilization in Windows Task Manager, regardless of actual Workload. Higher TDP tests produce higher Core temperatures. Results will vary according to Microarchitecture, Core count, Cache, Core speed, Turbo Boost, Core voltage, Hyperthreading, Instruction Sets, BIOS versions and CPU microcode. Intel tests their processors on an open bench, without a case, under carefully controlled conditions at 100% TDP. So when performing your own thermal test, the goal is to run a utility that will get you as close as possible to 100% TDP, which for your 4670K is 84 Watts. Prime95 version 26.6 Small FFT's is ideal for CPU thermal testing, because it's a steady 100% workload with steady Core temperatures that typically runs Core i variants with Hyperthreading and Core 2 processors within +/- a few % of TDP. No other utility so closely replicates Intel's proprietary test conditions. This is also the utility that Real Temp uses to test Core temperature sensors. Prime95's default test, Blend, is a fluctuating workload for testing memory stability, and Large FFT's combines CPU and memory tests. As such, Blend and Large FFT's both have fluctuating workloads which aren’t well suited for CPU thermal testing. Other stability tests such as OCCT have cycles that exceed 120% workload, which again aren’t well suited for CPU thermal testing. However, OCCT will by default, terminate the CPU tests at 85°C. The "Charts" in SpeedFan span 13 minutes, and show how each test creates distinct thermal signatures. Shown above from left to right: Small FFT's, Blend, Linpack and Intel Burn Test. Note the steady thermal signature of Small FFT's, which allows accurate measurements of Core temperatures. A steady 100% workload is crucial for thermal testing, as the CPU, socket, motherboard, VRM's and power delivery components can heat soak and stabilize. Shown above from left to right: Small FFT's, Intel Extreme Tuning Utility CPU Test, and AIDA64 CPU Test. Intel Extreme Tuning Utility is also a fluctuating workload. Although AIDA64's CPU test is a steady workload, it's far below TDP, which is insufficient for thermal testing. All other AIDA64 CPU test combinations are fluctuating workloads, which again aren't well suited for thermal testing. Guys, if you'd like to get yourself up to speed on this topic, then check out this Sticky near the top of the CPU's Forum: Intel Temperature Guide CT
I chose to experiment with my CPU's overclocking. As mentioned, it's an Intel i5 4670K paired with a Cooler Master Hyper 212 Evo cooler. I adjusted the multiplier to 44 (which equals 4.4Ghz) and set the voltage to 1.250, then performed a stability test using Prime95 (blend preset). Initially, everything worked fine—CPU temperatures hovered between 55-60°C and power consumption around 134W. However, after about 30 minutes, some unusual behavior appeared. Is this typical? (see image: https://snag.gy/5ECdqW.jpg)
Thank you ahead of time.
It's been a while since I worked with a Haswell chip, but unless there have been any updates, Haswell models seem to struggle with Prime95, possibly due to a code bug. The temperatures are much higher than expected. Please verify your readings with another stress test.
Also, the 212 cooler is on the lower side, and a multiplier of 4.4 is quite high for that heat sink. Some processors can handle it better. Did you forget to turn off adaptive voltage? That could cause damage. Also, remember to boost your core multiplier and voltage—this helps stabilize the overclock.
Yes, adaptive voltage was disabled.
I wouldn't say 212 is a weak cooler or that 4.4 is excessive. I had good luck with a chip, and my default voltage is nearly 1V. I actually tested overclocking back in 2014 and achieved stable 4.4 at 1.16V. Plus, I don’t run any rendering programs, so even during games I rarely exceed 50-55°C, even when overclocked.
Appreciate your response.
This was quite unusual, jumping from 55-60 to 99c. Did it fail? I tried using different tools like MSI Afterburner to check CPU temperature, and even ran another stress test with Aida64. It didn’t crash after about 30 seconds. After two weeks on this setup, it hasn’t stopped and temperatures stayed below 60°C, so I’m not concerned. 😀
CompuTronix: andaja, Respectfully, you're running the wrong Prime95 version, and probably the wrong torture test (Blend), which is why your Core temperatures started OK then went high, and why your Vcore and Core Watts are also high. You're probably running the latest version (29.4), however, the correct version to run is 26.6, and the correct torture test to run is Small FFT's. Here's the problem: 2nd through 8th Generation i3, i5 and i7 CPU's have AVX (Advanced Vector Extension) Instruction Sets. Prime95 versions later than 26.6 run AVX code on the CPU's Floating Point Unit (FPU) which causes unrealistic temperatures up to 20°C higher. Certain other utilities may have similar results. • Prime95 v26.6 - http://www.mersenneforum.org/showthread.php?t=15504 AVX can be disabled in Prime95 versions later than 26.6 by inserting "CpuSupportsAVX=0" into the "local.txt" file in Prime95's folder. However, since Core temperatures will be the same as 26.6, it's easier to just use 26.6. AVX doesn't affect Core i 1st Generation, Core 2, Pentium or Celeron processors since they don't have AVX Instruction Sets. If you’re overclocked and run AVX apps such as for rendering or transcoding, you may need to reduce Vcore and Core speed or upgrade your cooler and case fans so Core temperatures don’t reach 85°C. Many 6th, 7th and 8th Generation motherboards address the AVX problem by providing offset adjustments in BIOS. An offset of -2 or -3 (200 or 300 MHz) is usually sufficient. 4th Generation Haswell processors have no AVX offset adjustments in BIOS. Asus RealBench runs a realistic AVX workload typically within +/- a few % of TDP, and is an excellent utility for testing overall system stability, whether you're overclocked or not. RealBench is also the utility that Silicon Lottery uses for testing CPU stability. • Asus RealBench - http://rog.asus.com/rog-pro/realbench-v2-leaderboard/ Here’s a sample of thermal and stability test utilities shown according to % of TDP at Default BIOS settings: TDP … Thermal Test - Steady Workload 129% … Prime95 v27.7 through v29.4 - Small FFT’s (AVX, No Offset) 101 % <-- Prime95 v26.6 - Small FFT’s 89% … HeavyLoad v3.4.0.234 - Stress CPU 87% … FurMark v1.19.1.0 - CPU Burner 78% … CPU-Z v1.82.0 - Bench - Stress CPU 66% … AIDA64 v5.95.4500 - System Stability Test - Stress CPU 55% … Intel Processor Diagnostic Tool v4.0 - CPU Load TDP … Stability Test - Fluctuating Workload (Peak) 123% … OCCT v4.5.1 - CPU: OCCT (AVX, No Offset) 118% … LinX v0.6.5 - Default 116% … IntelBurn Test v2.54 - High 113% … OCCT v4.5.1 - CPU: Linpack (AVX, No Offset) 110% … AIDA64 v5.95.4500 - System Stability Test - Stress FPU 99 % <-- Asus RealBench v2.56 - Stress Test (AVX, No Offset) 94% … Sandra 2017.09.24.41 - Burn in - Processor Tests 92% … CineBench v15.0 - CPU - Render Test 79% … Intel Extreme Tuning Utility v6.4.1.15 - CPU Stress Test All tests will show 100% CPU Utilization in Windows Task Manager, regardless of actual Workload. Higher TDP tests produce higher Core temperatures. Results will vary according to Microarchitecture, Core count, Cache, Core speed, Turbo Boost, Core voltage, Hyperthreading, Instruction Sets, BIOS versions and CPU microcode. Intel tests their processors on an open bench, without a case, under carefully controlled conditions at 100% TDP. So when performing your own thermal test, the goal is to run a utility that will get you as close as possible to 100% TDP, which for your 4670K is 84 Watts. Prime95 version 26.6 Small FFT's is ideal for CPU thermal testing, because it's a steady 100% workload with steady Core temperatures that typically runs Core i variants with Hyperthreading and Core 2 processors within +/- a few % of TDP. No other utility so closely replicates Intel's proprietary test conditions. This is also the utility that Real Temp uses to test Core temperature sensors. Prime95's default test, Blend, is a fluctuating workload for testing memory stability, and Large FFT's combines CPU and memory tests. As such, Blend and Large FFT's both have fluctuating workloads which aren’t well suited for CPU thermal testing. Other stability tests such as OCCT have cycles that exceed 120% workload, which again aren’t well suited for CPU thermal testing. However, OCCT will by default, terminate the CPU tests at 85°C. The "Charts" in SpeedFan span 13 minutes, and show how each test creates distinct thermal signatures. Shown above from left to right: Small FFT's, Blend, Linpack and Intel Burn Test. Note the steady thermal signature of Small FFT's, which allows accurate measurements of Core temperatures. A steady 100% workload is crucial for thermal testing, as the CPU, socket, motherboard, VRM's and power delivery components can heat soak and stabilize. Shown above from left to right: Small FFT's, Intel Extreme Tuning Utility CPU Test, and AIDA64 CPU Test. Intel Extreme Tuning Utility is also a fluctuating workload. Although AIDA64's CPU test is a steady workload, it's far below TDP, which is insufficient for thermal testing. All other AIDA64 CPU test combinations are fluctuating workloads, which again aren't well suited for thermal testing. Guys, if you'd like to get yourself up to speed on this topic, then check out this Sticky near the top of the CPU's Forum: Intel Temperature Guide CT If i had any money, i would give it to you. That was almost too much information 😀 Thanks a lot 😉
andaja :
wicked_sticky :
I chose to experiment with my CPU's overclocking. As mentioned, it's an Intel i5 4670K paired with a Cooler Master Hyper 212 Evo cooler. I adjusted the multiplier to 44 (which equals 4.4Ghz) and set the voltage to 1.250. After running a stability test on Prime95 (using the blend preset), everything seemed fine at first—temperatures hovered between 55-60°C and power consumption around 134W. However, after about 30 minutes, some unexpected behavior appeared.
Is this typical? (see image: https://snag.gy/5ECdqW.jpg)
Please let me know.
It's been a while since I worked with a Haswell chip, but unless there have been any updates, Haswell models seem to struggle with Prime95, possibly due to a code bug. The temperatures are much higher than expected. Please verify your readings with another stress test.
Also, the 212 cooler is on the lower side, and a multiplier of 4.4 is quite high for that heat sink. Some processors can handle it better. Have you turned off adaptive voltage? That could cause overheating. Did you increase both the uncore multiplier and voltage? That would help stabilize the overclock.
Yes, adaptive voltage was disabled. I wouldn't say the 212 is too low or the multiplier too high. I had a lucky chip this time, with default settings giving me almost 1V. I was testing overclocking back in 2014 and achieved stable 4.4 at 1.16V. Also, I don’t use any rendering software, so even in games I rarely exceed 50-55°C, even when overclocked.
Thank you for your reply.
If stability under stress isn't a concern, results in casual gaming will drop about 20°C compared to the best. I couldn't exceed 4.0 with the 212 without a thermal shutdown. I tried a CryoCore H5 Ultimate and a Phantek PH-TC14PE; both are stable at 4.4 (beyond that, it's just a heat limitation on my chip). The H5 is what I currently use and keep at 4.3 (a 9°C difference from 4.4 under stress).
I chose to experiment with my CPU's overclocking. As mentioned, it's an Intel i5 4670K paired with a Cooler Master Hyper 212 Evo cooler. I adjusted the multiplier to 44 (which equals 4.4Ghz) and set the voltage to 1.250, then performed a stability test using Prime95 (blend preset). Initially, everything worked well—CPU temperatures hovered between 55-60°C and power consumption around 134W. However, after about 30 minutes, unusual behavior appeared.
Is this typical? (see image: https://snag.gy/5ECdqW.jpg)
Thank you ahead of time.
It's been a while since I worked with a Haswell chip, but unless there have been any fixes, Haswell models seem to struggle with Prime95, possibly due to a code bug. The temperatures are much higher than expected. Please verify your readings with another stress test.
Also, the 212 cooler is on the lower end, and a multiplier of 4.4 is quite high for that heat sink. Some processors handle it better than others. Did you remember to turn off adaptive voltage? That could cause overheating. Have you increased your uncore multiplier and voltage? (this helps keep things stable during overclocking)
Yes, adaptive voltage was turned off. I wouldn't say the 212 is too low or that 4.4 is excessive. I had luck with a chip, and my default voltage is nearly 1V. I actually tested overclocking back in 2014 and achieved 4.4 at 1.16V stability. Also, I don’t use any rendering programs, so even in games I rarely exceed 50-55°C, even when overclocked.
Thanks for your reply.
If you're not concerned about stability under stress, results in gaming will be about 20°C lower than usual during the time.
I couldn't exceed 4.0 under stress with the 212 without a thermal shutdown. I tried a CryoCore H5 Ultimate and a Phantek PH-TC14PE; both are stable at 4.4 (beyond that, it's just a heat limitation on my chip). The H5 is what I use now and keeps at 4.3 (a 9°C difference between 4.3 and 4.4 under stress).
I completely understand that gaming isn't as demanding on the CPU as rendering tasks.
Just completed a couple of hours of stress testing according to CompuTronix's advice—it was very stable with temperatures around 70°C (4.4Ghz, 1.25V).