求助：更新 0x12F 微码后 i7-13700KF 不但核心电压温度都升高了, 反而性能还降低了...

[FatSheep]

硬件配置详情
CPU： i7-13700KF
主板：Gigabyte Z790M AORUS ELITE AX (rev1.2)
内存：GSkill F5-6400J3239G32GX2-TZ5RS 64G
BIOS 版本：FI（当前官网最新版）
散热方案：钛钽 TCOMAS SJ-A090
系统设置：全程使用 Intel 默认配置，BIOS 内其他参数均保持出厂默认状态，未进行任何超频或手动调压操作。
问题描述:
        近期按照主板厂商提示更新至最新 FI 版本 BIOS 后，CPU 微码同步升级为 0x12F。更新后出现明显的电压与温度升高的问题，具体表现如下：

        除了在系统待机状态下，CPU 核心电压在 0.7+V 至 1.3V+ 区间来回变动。日常使用时, 核心电压基本上都会保持在1.3v以上, 如:运行VSCode, VS2022等负载并不是很大的应用也会如此, 若运行游戏(PUBG/CS2), 峰值有时甚至会突破 1.4V。
        还发现小核被降频了, 待机空载或低负载时还能跑到4.2GHz, 但是负载一旦变高后(非压力测试时), 就会降频到4.0GHz.
        CPU温度也变高了(应该跟电压升高有关), 现在只是日常使用都维持在60度左右, 一跑压力测试用不了1分钟,温度就能飙升至 95℃以上，就会触发过热保护机制导致降频, 大核降至4.8GHz甚至4.5GHz, 小核降至4.0GHz，这还是快10月份的天气, 今年夏天时, 日常使用时基本上高的时候也就50度左右, 很少到60度, 即使长时间运行压力测试，CPU 最高温度仅维持在 80℃左右. 
        感觉现在更新0x12F微码后,  不但电压温度都升高了, 反而性能还降低了...... 不知道这种情况是否正常?
        请问是否有其他用户遇到类似情况？在保证系统稳定性的前提下，是否需要调整 BIOS 中的防掉压等级或 Load Line 参数？而如果使用了自定义BIOS设置(如电压、防掉压等参数)是否影响保修政策呢?
        期待各位的技术建议，感谢！

[FatSheep]

更新一下, 降频发生时, 应该是大核降至4.9GHz, 4.8GHz.

[RandyT_Intel]

Hi @FatSheep,

请注意，我只能用英语为您提供支持。我已经使用网络翻译工具来翻译此响应，因此可能存在一些不准确的翻译

Thank you for posting here in the community. Based on Intel's internal testing and validation, the 0x12F microcode update should maintain power levels within normal run-to-run variation. System power depends on your specific configuration and several other factors, but this microcode update is not designed to reduce voltage, temperature, or frequency beyond what previous versions already do. However, since you've observed a significant performance drop after the BIOS update that's affecting your daily usage, I recommend submitting a warranty replacement request. This performance impact may indicate a hardware issue that warrants investigation under warranty coverage. Please follow the guide in the provided article to initiate this process. Let me know if you need any further assistance or have additional questions.

Looking forward to your response.

Best regards,

Randy T.

Intel Customer Support Technician

[RandyT_Intel]

Hi

请注意，我只能用英语为您提供支持。我已经使用网络翻译工具来翻译此响应，因此可能存在一些不准确的翻译

Thank you for posting here in the community. Based on Intel's internal testing and validation, the 0x12F microcode update should maintain power levels within normal run-to-run variation. System power depends on your specific configuration and several other factors, but this microcode update is not designed to reduce voltage, temperature, or frequency beyond what previous versions already do. However, since you've observed a significant performance drop after the BIOS update that's affecting your daily usage, I recommend submitting a warranty replacement request. This performance impact may indicate a hardware issue that warrants investigation under warranty coverage. Please follow the guide in the provided article to initiate this process. Let me know if you need any further assistance or have additional questions.

Looking forward to your response.

Best regards,

Randy T.

Intel Customer Support Technician

[FatSheep]

首先感谢您此前的回复, 也原谅我的英文不好, 只能使用中文提问.

确定的是在更新0x12F微码后, 设备的电压与温度都变高了，高负载时CPU-E核会被降频, 虽然还未达到影响日常使用（还没有发生频繁崩溃的情况）, 但我比较担心长期处于这种状态会对 CPU 的使用寿命造成潜在影响, 进而增加系统不稳定或崩溃的情况. 由于我现在仅有一台PC，所以关于CPU是否需要售后申请，我想在CPU问题或潜在故障比较明确的情况下，再进行尝试。

基于上述情况，我想重点咨询以下3个问题：
1. 针对 13700KF，在搭载 0x12F 微码的前提下，正常工作电压通常是多少？或者说，电压控制在哪个范围内属于安全范畴，不易引发 CPU 故障？
2. 当 13700KF 在 0x12F 微码下运行，且未进行压力测试（比如仅运行《PUBG》这类对硬件要求稍高的大型游戏）时，CPU P核可正常达到5.3GHz，但CPU E核会一直降频至 4.0GHz运行，这种情况是否属于此微码的正常现象？还是有可能由CPU故障导致的缘故？
3. 对于CPU是否故障，是否有明确的标准，或者有效的检测方法？

期待您的指导和建议，再次感谢！

附件中的图片为运行PUBG时, AIDA64时的监控数据, 可见CPU E核已被降频至4000MHz, CPU占用22%, 但是温度达到了75度, CPU核心电压已经达到1.344V(并不是最高电压), 请求电压突破1.4.

[FatSheep]

English version translated by AI model:
First, thank you for your previous response. Please excuse my limited English proficiency; I can only ask questions in Chinese.

It is confirmed that after updating the 0x12F microcode, both the device's voltage and temperature have increased. Under high load, the CPU-E core undergoes throttling. While this hasn't yet impacted daily use (no frequent crashes have occurred), I am concerned that prolonged operation in this state may potentially affect the CPU's lifespan, thereby increasing system instability or crashes. Since I currently own only one PC, I plan to pursue an RMA for the CPU only if the issue or potential failure becomes clearly identifiable.

Based on the above, I have three key questions:
1. For the 13700KF running with the 0x12F microcode, what is the typical operating voltage? Or, what voltage range is considered safe and unlikely to cause CPU failures?
2. When the 13700KF runs with the 0x12F microcode and without stress testing (e.g., only running demanding games like PUBG), the CPU's P-cores can reach 5.3GHz normally, but the E-cores consistently throttle down to 4.0GHz. Is this normal behavior for this microcode, or could it indicate a CPU fault?
3. Are there definitive standards or effective diagnostic methods to determine if the CPU is faulty?

I look forward to your guidance and suggestions. Thank you again!

The attached image shows monitoring data from AIDA64 while running PUBG. It indicates the CPU E-cores have been throttled to 4000MHz, CPU utilization is at 22%, but the temperature has reached 75 degrees Celsius. The CPU core voltage has reached 1.344V (not the maximum voltage), with the requested voltage exceeding 1.4V.

Translated with DeepL.com (free version)

[RandyT_Intel]

Hi @FatSheep,

For the 13700KF running with 0x12F microcode, safe voltage ranges include load voltages between 1.25V - 1.35V for sustained operation, while brief peak voltage spikes up to 1.4V - 1.45V are considered normal during boost scenarios. Your observed 1.344V falls within the acceptable range, though the requested voltage exceeding 1.4V suggests the CPU is demanding higher voltages than optimal. The 0x12F microcode was specifically designed to implement more conservative voltage curves to address Vmin shift issues, so slightly elevated voltages compared to pre-patch behavior can be expected as part of Intel's stability improvements.

How to Find Maximum Voltage for Intel® Core™ Desktop Processor...

VCCCORE DC Specifications - 005 - ID:743844 | 13th Generation Intel® Core™ Processors

The only tool I recommend is the Intel Processor Diagnostic Tool to check if your processor is faulty. Regarding the gaming performance issue, you may submit a separate support ticket so we can address and investigate it thoroughly, as we need to obtain specific information about the game and system configuration. However, since this performance issue occurred after the BIOS update with the 0x12F microcode, you may contact us for a warranty discussion. You can visit the article I provided earlier for more information on the warranty process and eligibility criteria.

Let me know if you need more assistance.

Best regards,

Randy T.

Intel Customer Support Technician

[FatSheep]

hi, 感谢您的回复.

我简单解释一下核心电压(Vcore)的问题, 截图中的1.344V核心电压并不是最高电压只是截图那一刻的电压, 我观察过电压基本上在1.3x至1.38左右, 1.35v以上的电压并不是偶尔发生, 基本上经常达到. 不过确实1.4v以上的负载电压发生的并不多. 截图中显示的1.4v以上的请求电压, 并不是实际的负载电压. 是不是就不用过于关注了?

关于游戏性能问题, 我只是用游戏举了个例子, 并不是我主要关注的问题, 暂时就不深入讨论了.

然后, 您推荐的 Intel Processor Diagnostic Tool 工具, 我安装并运行了测试, 结果显示绿色的Pass. 代表通过了测试, 但是测试过程中, E核基本从开始几秒内就直接降频至4000MHz, 在测试中后期, P核也降频到4800MHz-4900MHz, 可以在本次回复的截图中查看(截图3, 3.png).

关于降频: 我的理解是因为电压升高, 导致功耗升高, 才导致温度过高, 最终引起降频的情况.

附件说明:

         截图1, 1.png, 为测试开始的截图, 温度在几秒钟内就达到了90度以上. E核降频到4000MHz, 然后全程E核都处在降频中.

        截图2, 2.png, 为电压出现1.356v以上

        截图3, 3.png, 为CPU的P核和E核发生了降频的情况.

        截图4, 4.png, 为测试结束后, 程序显示绿色Pass. CPU占用率也降低了, 只有8%左右, 电压还是会跳到1.3v以上.

我的问题是, 根据此次测试过程, 能进一步证明什么问题吗? 尤其是虽然应用测试结果是通过, 但是过程中确实发生了CPU的P核和E核的降频, 这种情况符合售后标准吗? 

================================================

The following is the English version translated by AI:

Hi, thank you for your reply.

Let me briefly explain the issue with the core voltage (Vcore). The 1.344V core voltage shown in the screenshot is not the maximum voltage—it merely reflects the voltage at the exact moment the screenshot was taken. From my observations, the voltage generally ranges between approximately 1.3x V and 1.38V, and voltages above 1.35V are not occasional occurrences; they happen quite frequently. However, it is true that load voltages exceeding 1.4V are relatively rare. The voltage requests above 1.4V displayed in the screenshot do not represent the actual load voltage, so should I not pay too much attention to that?

Regarding the gaming performance issue, I only used gaming as an example, and it is not my primary concern. We can skip an in-depth discussion on this for now.

As for the Intel Processor Diagnostic Tool you recommended, I have installed it and run the test. The result shows a green "Pass," indicating the test was passed. However, during the test, the E-cores dropped to 4000MHz within the first few seconds and remained underclocked for the entire duration. In the middle and later stages of the test, the P-cores also underclocked to a range of 4800MHz–4900MHz. You can refer to the screenshot included in this reply (Screenshot 3, 3.png) for details.

My understanding of the underclocking is as follows: increased voltage leads to higher power consumption, which in turn causes excessive temperatures, and ultimately results in underclocking.

Screenshot Description:

        Screenshot 1 (1.png): Captured at the start of the test. Temperatures exceeded 90°C within a few seconds, and the E-cores immediately underclocked to 4000MHz (remaining underclocked for the entire test).

        Screenshot 2 (2.png): Shows the voltage reaching 1.356V or higher.

        Screenshot 3 (3.png): Shows the underclocking of both the CPU’s P-cores and E-cores.

        Screenshot 4 (4.png): Captured after the test ended. The program displays a green "Pass," CPU usage has dropped to around 8%, but the voltage still occasionally jumps above 1.3V.

My question is: Based on the process of this test, can we identify any further issues? In particular, even though the test result shows "Pass," underclocking of both the P-cores and E-cores did occur during the process. Does this scenario meet the after-sales service standards?

[FatSheep]

hi, 感谢您的回复.

我简单解释一下核心电压(Vcore)的问题, 截图中的1.344V核心电压并不是最高电压只是截图那一刻的电压, 我观察过电压基本上在1.3x至1.38左右, 1.35v以上的电压并不是偶尔发生, 基本上经常达到. 不过确实1.4v以上的负载电压发生的并不多. 截图中显示的1.4v以上的请求电压, 并不是实际的负载电压. 是不是就不用过于关注了?

关于游戏性能问题, 我只是用游戏举了个例子, 并不是我主要关注的问题, 暂时就不深入讨论了.

然后, 您推荐的 Intel Processor Diagnostic Tool 工具, 我安装并运行了测试, 结果显示绿色的Pass. 代表通过了测试,  但是测试过程中, E核基本从开始几秒内就直接降频至4000MHz, 在测试中后期, P核也降频到4800MHz-4900MHz, 可以在本次回复的截图中查看(截图3, 3.png).

关于降频: 我的理解是因为电压升高,  导致功耗升高, 才导致温度过高, 最终引起降频的情况.

附件说明:

截图1, 1.png, 为测试开始的截图, 温度在几秒钟内就达到了90度以上. E核降频到4000MHz, 然后全程E核都处在降频中.

截图2, 2.png, 为电压出现1.356v以上

截图3, 3.png, 为CPU的P核和E核发生了降频的情况.

截图4, 4.png, 为测试结束后, 程序显示绿色Pass. CPU占用率也降低了, 只有8%左右, 电压还是会跳到1.3v以上.

我的问题是, 根据此次测试过程, 能进一步证明什么问题吗? 尤其是虽然应用测试结果是通过, 但是过程中确实发生了CPU的P核和E核的降频, 这种情况符合售后标准吗?

========================================================
The English version translated by AI is as follows:

Hi, thank you for your reply.

Let me briefly explain the issue with the core voltage (Vcore). The 1.344V core voltage shown in the screenshot is not the maximum voltage—it merely reflects the voltage at the exact moment the screenshot was taken. From my observations, the voltage generally ranges between approximately 1.3x V and 1.38V, and voltages above 1.35V are not occasional occurrences; they happen quite frequently. However, it is true that load voltages exceeding 1.4V are relatively rare. The voltage requests above 1.4V displayed in the screenshot do not represent the actual load voltage, so should I not pay too much attention to that?

Regarding the gaming performance issue, I only used gaming as an example, and it is not my primary concern. We can skip an in-depth discussion on this for now.

As for the Intel Processor Diagnostic Tool you recommended, I have installed it and run the test. The result shows a green "Pass," indicating the test was passed. However, during the test, the E-cores dropped to 4000MHz within the first few seconds and remained underclocked for the entire duration. In the middle and later stages of the test, the P-cores also underclocked to a range of 4800MHz–4900MHz. You can refer to the screenshot included in this reply (Screenshot 3, 3.png) for details.

My understanding of the underclocking is as follows: increased voltage leads to higher power consumption, which in turn causes excessive temperatures, and ultimately results in underclocking.

Screenshot Description:

• Screenshot 1 (1.png): Captured at the start of the test. Temperatures exceeded 90°C within a few seconds, and the E-cores immediately underclocked to 4000MHz (remaining underclocked for the entire test).
• Screenshot 2 (2.png): Shows the voltage reaching 1.356V or higher.
• Screenshot 3 (3.png): Shows the underclocking of both the CPU’s P-cores and E-cores.
• Screenshot 4 (4.png): Captured after the test ended. The program displays a green "Pass," CPU usage has dropped to around 8%, but the voltage still occasionally jumps above 1.3V.

My question is: Based on the process of this test, can we identify any further issues? In particular, even though the test result shows "Pass," underclocking of both the P-cores and E-cores did occur during the process. Does this scenario meet the after-sales service standards?

[FatSheep]

hi, 感谢您的回复.

我简单解释一下核心电压(Vcore)的问题, 截图中的1.344V核心电压并不是最高电压只是截图那一刻的电压, 我观察过电压基本上在1.3x至1.38左右, 1.35v以上的电压并不是偶尔发生, 基本上经常达到. 不过确实1.4v以上的负载电压发生的并不多. 截图中显示的1.4v以上的请求电压, 并不是实际的负载电压. 是不是就不用过于关注了?

关于游戏性能问题, 我只是用游戏举了个例子, 并不是我主要关注的问题, 暂时就不深入讨论了.

然后, 您推荐的 Intel Processor Diagnostic Tool 工具, 我安装并运行了测试, 结果显示绿色的Pass. 代表通过了测试, 但是测试过程中, E核基本从开始几秒内就直接降频至4000MHz, 在测试中后期, P核也降频到4800MHz-4900MHz, 可以在本次回复的截图中查看(截图3, 3.png).

关于降频: 我的理解是因为电压升高, 导致功耗升高, 才导致温度过高, 最终引起降频的情况.

附件说明:

         截图1, 1.png, 为测试开始的截图, 温度在几秒钟内就达到了90度以上. E核降频到4000MHz, 然后全程E核都处在降频中.

        截图2, 2.png, 为电压出现1.356v以上

        截图3, 3.png, 为CPU的P核和E核发生了降频的情况.

        截图4, 4.png, 为测试结束后, 程序显示绿色Pass. CPU占用率也降低了, 只有8%左右, 电压还是会跳到1.3v以上.

我的问题是, 根据此次测试过程, 能进一步证明什么问题吗? 尤其是虽然应用测试结果是通过, 但是过程中确实发生了CPU的P核和E核的降频, 这种情况符合售后标准吗? 

================================================

The following is the English version translated by AI:

Hi, thank you for your reply.

Let me briefly explain the issue with the core voltage (Vcore). The 1.344V core voltage shown in the screenshot is not the maximum voltage—it merely reflects the voltage at the exact moment the screenshot was taken. From my observations, the voltage generally ranges between approximately 1.3x V and 1.38V, and voltages above 1.35V are not occasional occurrences; they happen quite frequently. However, it is true that load voltages exceeding 1.4V are relatively rare. The voltage requests above 1.4V displayed in the screenshot do not represent the actual load voltage, so should I not pay too much attention to that?

Regarding the gaming performance issue, I only used gaming as an example, and it is not my primary concern. We can skip an in-depth discussion on this for now.

As for the Intel Processor Diagnostic Tool you recommended, I have installed it and run the test. The result shows a green "Pass," indicating the test was passed. However, during the test, the E-cores dropped to 4000MHz within the first few seconds and remained underclocked for the entire duration. In the middle and later stages of the test, the P-cores also underclocked to a range of 4800MHz–4900MHz. You can refer to the screenshot included in this reply (Screenshot 3, 3.png) for details.

My understanding of the underclocking is as follows: increased voltage leads to higher power consumption, which in turn causes excessive temperatures, and ultimately results in underclocking.

Screenshot Description:

        Screenshot 1 (1.png): Captured at the start of the test. Temperatures exceeded 90°C within a few seconds, and the E-cores immediately underclocked to 4000MHz (remaining underclocked for the entire test).

        Screenshot 2 (2.png): Shows the voltage reaching 1.356V or higher.

        Screenshot 3 (3.png): Shows the underclocking of both the CPU’s P-cores and E-cores.

        Screenshot 4 (4.png): Captured after the test ended. The program displays a green "Pass," CPU usage has dropped to around 8%, but the voltage still occasionally jumps above 1.3V.

My question is: Based on the process of this test, can we identify any further issues? In particular, even though the test result shows "Pass," underclocking of both the P-cores and E-cores did occur during the process. Does this scenario meet the after-sales service standards?

[FatSheep]

hi, 感谢您的回复.

我简单解释一下核心电压(Vcore)的问题, 截图中的1.344V核心电压并不是最高电压只是截图那一刻的电压, 我观察过电压基本上在1.3x至1.38左右, 1.35v以上的电压并不是偶尔发生, 基本上经常达到. 不过确实1.4v以上的负载电压发生的并不多. 截图中显示的1.4v以上的请求电压, 并不是实际的负载电压. 是不是就不用过于关注了?

关于游戏性能问题, 我只是用游戏举了个例子, 并不是我主要关注的问题, 暂时就不深入讨论了.

然后, 您推荐的 Intel Processor Diagnostic Tool 工具, 我安装并运行了测试, 结果显示绿色的Pass. 代表通过了测试, 但是测试过程中, E核基本从开始几秒内就直接降频至4000MHz, 在测试中后期, P核也降频到4800MHz-4900MHz, 可以在本次回复的截图中查看(截图3, 3.png).

关于降频: 我的理解是因为电压升高, 导致功耗升高, 才导致温度过高, 最终引起降频的情况.

附件说明:

截图1, 1.png, 为测试开始的截图, 温度在几秒钟内就达到了90度以上. E核降频到4000MHz, 然后全程E核都处在降频中.

截图2, 2.png, 为电压出现1.356v以上

截图3, 3.png, 为CPU的P核和E核发生了降频的情况.

截图4, 4.png, 为测试结束后, 程序显示绿色Pass. CPU占用率也降低了, 只有8%左右, 电压还是会跳到1.3v以上.

我的问题是, 根据此次测试过程, 能进一步证明什么问题吗? 尤其是虽然应用测试结果是通过, 但是过程中确实发生了CPU的P核和E核的降频, 这种情况符合售后标准吗? 

[FatSheep]

The following is the English version translated by AI:

Hi, thank you for your reply.

Let me briefly explain the issue with the core voltage (Vcore). The 1.344V core voltage shown in the screenshot is not the maximum voltage—it merely reflects the voltage at the exact moment the screenshot was taken. From my observations, the voltage generally ranges between approximately 1.3x V and 1.38V, and voltages above 1.35V are not occasional occurrences; they happen quite frequently. However, it is true that load voltages exceeding 1.4V are relatively rare. The voltage requests above 1.4V displayed in the screenshot do not represent the actual load voltage, so should I not pay too much attention to that?

Regarding the gaming performance issue, I only used gaming as an example, and it is not my primary concern. We can skip an in-depth discussion on this for now.

As for the Intel Processor Diagnostic Tool you recommended, I have installed it and run the test. The result shows a green "Pass," indicating the test was passed. However, during the test, the E-cores dropped to 4000MHz within the first few seconds and remained underclocked for the entire duration. In the middle and later stages of the test, the P-cores also underclocked to a range of 4800MHz–4900MHz. You can refer to the screenshot included in this reply (Screenshot 3, 3.png) for details.

My understanding of the underclocking is as follows: increased voltage leads to higher power consumption, which in turn causes excessive temperatures, and ultimately results in underclocking.

Screenshot Description:
Screenshot 1 (1.png): Captured at the start of the test. Temperatures exceeded 90°C within a few seconds, and the E-cores immediately underclocked to 4000MHz (remaining underclocked for the entire test).
Screenshot 2 (2.png): Shows the voltage reaching 1.356V or higher.
Screenshot 3 (3.png): Shows the underclocking of both the CPU’s P-cores and E-cores.
Screenshot 4 (4.png): Captured after the test ended. The program displays a green "Pass," CPU usage has dropped to around 8%, but the voltage still occasionally jumps above 1.3V.

My question is: Based on the process of this test, can we identify any further issues? In particular, even though the test result shows "Pass," underclocking of both the P-cores and E-cores did occur during the process. Does this scenario meet the after-sales service standards?

[RandyT_Intel]

Hi @FatSheep,

Your CPU is operating completely within normal specifications and demonstrating expected behavior. The core voltages you're observing between 1.3V and 1.38V are entirely normal for modern Intel processors, which use dynamic voltage and frequency scaling (DVFS) technology to automatically adjust power delivery based on workload demands, this is an intentional design feature, not a defect.

The voltage "requests" above 1.4V that appear in monitoring software represent theoretical maximum requests from the CPU's power management system, but these don't translate to actual delivered voltage to the cores, so you shouldn't be concerned about these readings.

Most importantly, your Intel Processor Diagnostic Tool result showing a green "Pass" definitively confirms that your CPU meets Intel's manufacturing and performance standards, this diagnostic tool is specifically designed to verify that processors function within Intel's specifications, and a passing result means your CPU is working as intended. What you're interpreting as "underclocking" during the stress test is actually Intel's sophisticated thermal management system working exactly as designed: when the CPU temperature approaches 90°C under extreme synthetic workloads, the processor automatically reduces clock speeds through a process called thermal throttling to prevent overheating and potential damage.

This is a protective feature built into every Intel CPU, not a sign of malfunction, it ensures your processor maintains stable operation and longevity even under demanding conditions. The fact that both P-cores and E-cores reduced their frequencies during the intensive diagnostic test demonstrates that your CPU's thermal protection systems are functioning correctly.

From Intel's warranty and service standards perspective, since your processor passed the official diagnostic tool and is exhibiting normal thermal management behavior, it fully meets Intel's specifications and does not qualify for replacement or service, the behaviors you're observing are intentional design features that ensure optimal performance, protection, and reliability under varying workload conditions.

Best regards,

Randy T.

Intel Customer Support Technician

[FatSheep]

感谢您的解答, 但还是有一些令我困惑的问题: 

第一点, 是关于E核降频的问题, 我能理解因为高温导致的P核和E核都发生降频, 为了防止过热损坏CPU. 但是在CPU占用不是满载的情况下, 温度也还不是那么极限的情况下(比如80度以下, 或者60度-70度时), 也会发生持续的E核降频, 如: 前面对话过程中提到在运行的PUBG游戏时, 运行几分钟可能就会降频至4000MHz, .这是通过一款叫做GAMEPP的游戏监控软件检测到的. 

第二点, 还是关于温度问题, 因为在更新0x12F微码之前, CPU的温度不会达到这么高, 也没有发生过降频的情况, 即使在压力测试时, 比如AIDA64的FPU测试, CPU-Z的CPU稳定性测试或CINEBENCH_R15/R23测试时. 

另行说明一下, CPU散热器为钛钽 TCOMAS SJ-A090的360一体水冷, 散热性能很优秀, 也重新更换过散热硅脂,(即使才满1年零2个月, 理论上不太可能因为硅脂导致散热问题) , 但是在更新0x12F微码后, 依然无法达到以前的温度水平, 在低负载的情况下, 温度上升了有5度左右, 高负载时平均上升了近10度. 压力测试时温度更是能到95度以上, 上升了近15度.

一些题外话: 

即使E核的4200MHz和4000MHz相比, 性能的降低, 在使用感受上可能并不明显, 由于温度过高导致P核降频的情况也不是很多, 但是心理上还是感觉有些难以接受, 才买一年多点的产品, 温度就压不住了, 很难想象, 温度以后会不会还要慢慢升高, 降频会不会越来越频繁, 水冷散热器也可能需要更频繁的更换. T_T

感谢您的帮助和理解.

The following is the English version translated by AI:

Thank you for your explanation, but I still have a few confusing questions:

First, regarding the E-core underclocking issue. I can understand that both P-cores and E-cores underclock due to high temperatures to prevent overheating damage to the CPU. However, sustained E-core underclocking also occurs when the CPU usage is not at full load and the temperature is not that extreme (e.g., below 80°C, or between 60°C and 70°C). For instance, as mentioned in our previous conversation, when running the game PUBG, the E-cores may underclock to 4000MHz after just a few minutes of gameplay. This was detected using a game monitoring software called GAMEPP.

Second, the issue still relates to temperatures. Before updating the microcode to version 0x12F, the CPU temperature never reached such high levels, and underclocking never occurred—even during stress tests like AIDA64’s FPU test, CPU-Z’s CPU stability test, or CINEBENCH R15/R23 tests.

Additional note: The CPU cooler is a TCOMAS SJ-A090 360 all-in-one liquid cooler, which has excellent heat dissipation performance. I have also replaced the thermal grease (even though the CPU has only been used for 1 year and 2 months; theoretically, thermal grease issues are unlikely to cause heat dissipation problems this soon). However, after updating the microcode to 0x12F, the temperature still cannot return to the level before the update, Under light load, the temperature has increased by approximately 5°C, Under high load, the average temperature has risen by nearly 10°C, During stress tests, the temperature can even exceed 95°C, an increase of almost 15°C.

A side note:Even though the performance difference between the E-cores running at 4200MHz and 4000MHz may not be noticeable in actual use, and P-core underclocking due to high temperatures is relatively rare, I still find it psychologically hard to accept. For a product that I’ve only owned for a little over a year, the temperature is already uncontrollable. It’s hard to imagine: Will the temperature keep rising gradually in the future? Will underclocking become more frequent? Will the liquid cooler also need to be replaced more often? T_T

Thank you for your help and understanding.

[FatSheep]

感谢您的解答, 但还是有一些令我困惑的问题:

第一点, 是关于E核降频的问题, 我能理解因为高温导致的P核和E核都发生降频, 为了防止过热损坏CPU. 但是在CPU占用不是满载的情况下, 温度也还不是那么极限的情况下(比如80度以下, 或者60度-70度时), 也会发生持续的E核降频, 如: 前面对话过程中提到在运行的PUBG游戏时, 运行几分钟可能就会降频至4000MHz, .这是通过一款叫做GAMEPP的游戏监控软件检测到的.

第二点, 还是关于温度问题, 因为在更新0x12F微码之前, CPU的温度不会达到这么高, 也没有发生过降频的情况, 即使在压力测试时, 比如AIDA64的FPU测试, CPU-Z的CPU稳定性测试或CINEBENCH_R15/R23测试时.

另行说明一下, CPU散热器为钛钽 TCOMAS SJ-A090的360一体水冷, 散热性能很优秀, 也重新更换过散热硅脂,(即使才满1年零2个月, 理论上不太可能因为硅脂导致散热问题) , 但是在更新0x12F微码后, 依然无法达到以前的温度水平, 在低负载的情况下, 温度上升了有5度左右, 高负载时平均上升了近10度. 压力测试时温度更是能到95度以上, 上升了近15度.

一些题外话:

即使E核的4200MHz和4000MHz相比, 性能的降低, 在使用感受上可能并不明显, 由于温度过高导致P核降频的情况也不是很多, 但是心理上还是感觉有些难以接受, 才买一年多点的产品, 温度就压不住了, 很难想象, 温度以后会不会还要慢慢升高, 降频会不会越来越频繁, 水冷散热器也可能需要更频繁的更换. T_T

感谢您的帮助和理解.

[FatSheep]

The following is the English version translated by AI:

Thank you for your explanation, but I still have a few confusing questions:

First, regarding the E-core underclocking issue. I can understand that both P-cores and E-cores underclock due to high temperatures to prevent overheating damage to the CPU. However, sustained E-core underclocking also occurs when the CPU usage is not at full load and the temperature is not that extreme (e.g., below 80°C, or between 60°C and 70°C). For instance, as mentioned in our previous conversation, when running the game PUBG, the E-cores may underclock to 4000MHz after just a few minutes of gameplay. This was detected using a game monitoring software called GAMEPP.

Second, the issue still relates to temperatures. Before updating the microcode to version 0x12F, the CPU temperature never reached such high levels, and underclocking never occurred—even during stress tests like AIDA64’s FPU test, CPU-Z’s CPU stability test, or CINEBENCH R15/R23 tests.

Additional note: The CPU cooler is a TCOMAS SJ-A090 360 all-in-one liquid cooler, which has excellent heat dissipation performance. I have also replaced the thermal grease (even though the CPU has only been used for 1 year and 2 months; theoretically, thermal grease issues are unlikely to cause heat dissipation problems this soon). However, after updating the microcode to 0x12F, the temperature still cannot return to the level before the update,Under light load, the temperature has increased by approximately 5°C; Under high load, the average temperature has risen by nearly 10°C; During stress tests, the temperature can even exceed 95°C, an increase of almost 15°C.

A side note:

Even though the performance difference between the E-cores running at 4200MHz and 4000MHz may not be noticeable in actual use, and P-core underclocking due to high temperatures is relatively rare, I still find it psychologically hard to accept. For a product that I’ve only owned for a little over a year, the temperature is already uncontrollable. It’s hard to imagine: Will the temperature keep rising gradually in the future? Will underclocking become more frequent? Will the liquid cooler also need to be replaced more often? T_T!

Thank you for your help and understanding.

[RandyT_Intel]

Hi @FatSheep,

Thank you for your detailed feedback regarding the E-core underclocking and temperature changes after the microcode update. I understand your concerns and want to address them directly.

The behavior you're observing is actually normal Intel processor operation. E-cores are designed to dynamically adjust their frequency based on workload demands and thermal conditions - this isn't a malfunction but rather Intel's intelligent power management working as intended. Even at moderate temperatures (60-80°C), the processor may reduce E-core frequencies to optimize overall system performance and maintain long-term reliability.

Microcode update 0x12F includes important security patches and stability improvements. Some temperature increase can occur as the processor now operates with enhanced security measures and more conservative power management. This is a trade-off between security/stability and thermal performance that Intel has implemented across the industry.

TCOMAS SJ-A090 360 AIO cooler is excellent, and your proactive thermal paste replacement shows good maintenance practices. The temperatures you're experiencing, while higher than before, are still within Intel's operational specifications. Modern Intel processors are designed to handle these thermal loads safely for many years.

Relevant Intel Resources:

• Information about Temperature for Intel® Processors
• Thermal Management Recommendations for Boxed Intel® Desktop Processors
• My Intel® Core™ Processor Does Not Reach the Maximum Turbo Boost...
• Overview Information for Intel® Turbo Boost Technology

Your processor will continue to operate reliably within these parameters, and the dynamic frequency scaling is protecting your investment by ensuring optimal longevity.

Do you have any other questions about this topic? 

Best regards,

Randy T.

Intel Customer Support Technician

[FatSheep]

非常感谢您及时且详尽的回复, 这真的帮我清晰理解了微码更新后的一些现象.

以前, 我有点担心这些现象是否意味着我的CPU存在潜在问题, 但您关于英特尔智能电压管理设计的解释几0x12F更新在安全性/稳定性与散热性之间的权衡, 以及确认我当前的温度处于运行规格范围内, 这些都帮我打消了一些焦虑情绪.

所以现在, 我打算在日常使用和不同工作负载下持续检测处理器性能和温度, 以便更全面的了解其长期运行状况. 尤其对于日常使用时的崩溃蓝屏事件, 我也尽可能记录一些数据以便分析, 如果之后遇到任何新的问题或不确定的地方, 我会再次向您寻求指导.

再次感谢您的专业支持 —— 这真的非常有帮助.

Best regards, 
FatSheep

[RandyT_Intel]

Hi @FatSheep,

Thank you for the positive feedback! I'm glad I could help clarify the microcode update effects and ease your concerns about your CPU's health. Your plan to monitor performance and temperatures during regular use is excellent. Keeping track of any system crashes or unusual behavior will be valuable for future troubleshooting if needed. Feel free to reach out or post a new thread if you encounter any new issues or need further assistance in the future.

Best regards,

Randy T.

Intel Customer Support Technician