SkatterBencher #48: AMD Ryzen 5 7600X Overclocked to 5544 MHz

We overclock the AMD Ryzen 5 7600X up to 5544 MHz with the ROG Strix X670E-F Gaming Wifi motherboard and EK custom loop water cooling.

https://www.youtube.com/watch?v=sEWYcpcZ2Zo

Even though the overclocking results may not be overly impressive, I hope you can still find value from the information I provide in this article.

AMD Ryzen 5 7600X: Introduction

The AMD Ryzen 5 7600X processor is part of the Ryzen 7000 Zen 4 AM5 desktop product line-up.

The Zen 4 desktop CPUs codenamed Raphael launched on August 29, 2022, during an AMD Livestream. At launch, there were 4 SKUs available. The entry-level processor is the Ryzen 5 7600X we’re overclocking today.

Zen 4 Raphael is very similar to Zen 3 Vermeer, but there are some noteworthy differences. Most importantly, Raphael requires the new AM5 socket and supports DDR5 memory. The CCDs, featuring the Zen 4 CPU cores, is AMD’s first chip using TSMC N5 manufacturing technology. Raphael also introduces AVX-512 support and operates at significantly higher clock frequencies.

The Ryzen 5 7600X has six cores with 12 threads, a base frequency of 4.7 GHz, and a maximum listed boost frequency of 5.3GHz. The TDP is 105W, and PPT is 142W.

In this blog post, we will cover six overclocking strategies:

  • First, we simply rely on AMD PBO and EXPO performance boost technologies
  • Second, we rely on ASUS AI Overclocking performance boost technology
  • Third, we tune Precision Boost Overdrive with Curve Optimizer
  • Fourth, we tune Precision Boost Overdrive with asynchronous eCLK
  • Fifth, we try to do a manual overclock
  • Lastly, we combine our best PBO and manual overclocking results using ASUS Dynamic OC Switcher

However, before we jump into overclocking, let us quickly review the hardware and benchmarks used in this article.

AMD Ryzen 5 7600X: Platform Overview

Along with the AMD Ryzen 5 7600X processor and ASUS ROG Strix X670E-F Gaming Wifi motherboard, in this guide, we will be using a pair of G.SKILL Trident Z5 DDR5-6400 memory sticks, an ASUS ROG STRIX RTX 2080TI graphics card, an 512GB M.2 NVMe SSD, a Seasonic Prime 850W Platinum power supply, the ElmorLabs Easy Fan Controller, the ElmorLabs EVC2SX, the EK-Quantum Magnitude water block, and EK-Quantum water cooling. All this is mounted on top of our favorite Open Benchtable V2. 

ElmorLabs EFC & EVC2

I explained how I use ElmorLabs products in SkatterBencher #34. By connecting the EFC and PMD to the EVC2 device, I monitor the ambient temperature (EFC), water temperature (EFC), fan duty cycle (EFC), and CPU input power (PMD). I include the measurements in my Prime95 stability test results.

I also use the ElmorLabs EFC to map the radiator fan curve to the water temperature. Without going into too many details: I have attached an external temperature sensor from the water in the loop to the EFC. Then, I use the low/high setting to map the fan curve from 25 to 40 degrees water temperature. I use this configuration for all overclocking strategies.

AMD Ryzen 5 7600X: Benchmark Software

We use Windows 11 and the following benchmark applications to measure performance and ensure system stability.

AMD Ryzen 5 7600X: Stock Performance

Before starting overclocking, we must check the system performance at default settings. Note that on this motherboard, Precision Boost Overdrive is enabled by default.

So, to check the performance at default settings, you must enter the BIOS and

  • Switch to the Advanced Mode view
  • Go to the Ai Tweaker menu
  • Enter the Precision Boost Overdrive submenu
  • Set Precision Boost Overdrive to Disabled

Then save and exit the BIOS.

Note that the Precision Boost 2 boosting algorithm is still active even though we disabled Precision Boost Overdrive. The standard parameters of the Precision Boost algorithm for the Ryzen 5 7600X are as follows:

7600x stock parameters

Here is the benchmark performance at stock:

  • SuperPI 4M: 30.051 seconds
  • Geekbench 5 (single): 2,161 points
  • Geekbench 5 (multi): 11,114 points
  • Cinebench R23 Single: 1,962 points
  • Cinebench R23 Multi: 14,699 points
  • CPU-Z V17.01.64 Single: 754.0 points
  • CPU-Z V17.01.64 Multi: 5,964.0 points
  • V-Ray 5: 11,210 vsamples
  • AI Benchmark: 4,205 points
  • 3DMark Night Raid: 68,232 points
  • CS:GO FPS Bench: 593.29 fps
  • Tom Raider: 194 fps
  • Final Fantasy XV: 188.31 fps
7600x stock benchmark performance

Here are the 3DMark CPU Profile scores at stock

  • CPU Profile 1 Thread: 1,097
  • CPU Profile 2 Threads: 2,179
  • CPU Profile 4 Threads: 4,124
  • CPU Profile 8 Threads: 6,333
  • CPU Profile 16 Threads: 6,975
  • CPU Profile Max Threads: 7,062
7600x stock 3dmark performance

When running Prime 95 Small FFTs with AVX-512 enabled, the average CPU effective clock is 4945 MHz with 1.343 volts. The average CPU temperature is 95.3 degrees Celsius. The ambient and water temperature is 26.4 and 34.1 degrees Celsius. The average CPU package power is 130.2 watts.

7600x stock prime95 avx

When running Prime 95 Small FFTs with AVX disabled, the average CPU effective clock is 5012 MHz with 1.385 volts. The average CPU temperature is 95.2 degrees Celsius. The ambient and water temperature is 26.7 and 34.1 degrees Celsius. The average CPU package power is 123.1 watts.

7600x stock prime95 non-avx

Now, let us try our first overclocking strategy.

However, before we get going, make sure to locate the CMOS Clear button

Pressing the Clear CMOS button will reset all your BIOS settings to default which is helpful if you want to start your BIOS configuration from scratch. However, it does not delete any of the BIOS profiles previously saved. The Clear CMOS button is located in the bottom right of the motherboard.

OC Strategy #1: PBO 2 + EXPO

In our first overclocking strategy, we take advantage of AMD exclusive features Precision Boost Overdrive and  Extended Profiles for Overclocking.

PBO 2 – Precision Boost Overdrive 2

With the launch of Zen 3, AMD introduced an improved version of the Precision Boost Overdrive toolkit, allowing for manual tuning of the parameters affecting the Precision Boost frequency boost algorithm.

Precision Boost Overdrive 2 builds on the PBO implementation of Zen 2. In addition to the platform overclocking knobs from Zen+ (PPT, TDC, EDC) and processor overclocking knobs from Zen 2 (Boost Override and Scalar), Precision Boost Overdrive 2 introduces Curve Optimizer.

There are essentially 3 levels of Precision Boost Overdrive

  1. AMD’s stock values, which can be set by disabling PBO
  2. The motherboard vendor values, which are programmed into the BIOS to match the motherboard VRM specification and can be set by enabling PBO
  3. Custom values, which can be programmed by the end-user

I explored the ins and outs of Precision Boost Overdrive 2 tuning in my AMD Raphael Overclocking launch article. If you want to learn more about the impact of each of these settings, I suggest you check out that post.

In this overclocking strategy, we’re just enabling Precision Boost Overdrive, whereas, in the following strategies, we’ll explore tuning the parameters. We rely on the motherboard pre-programmed PBO parameters by enabling Precision Boost Overdrive.

We find that the following values have changed:

7600x pbo expo parameters

Increasing the platform boost parameters typically have the most significant effect with processors that have a lot of cores. With the 6 cores of the 7600X, we don’t even hit the limits with stock values. During a Prime95 non-AVX workload, the maximum values for PPT, TDC, and EDC are 120W, 80A, and 82A, respectively. That’s well below the limits.

So, ultimately, we don’t expect much performance improvement.

EXPO – Extended Profiles for Overclocking

EXPO stands for AMD Extended Profiles for Overclocking. It is an AMD technology that enables ubiquitous memory overclocking profiles for AMD platforms supporting DDR5 memory. EXPO allows memory vendors such as G.SKILL to program higher performance settings onto the memory sticks. If the motherboard supports EXPO, you can enable higher performance with a single BIOS setting. So, it saves you lots of manual configuration.

While our memory kit is rated at DDR5-6400, I had to run it at DDR5-5600 for stability reasons as my system wasn’t stable at higher frequencies. I’m using an early CPU, motherboard, and BIOS combination, so memory overclocking will likely improve over time.

BIOS Settings & Benchmark Results

Upon entering the BIOS

  • Switch to the Advanced Mode view
  • Go to the Ai Tweaker menu
  • Set Ai Overclock Tuner to EXPO II
  • Set Memory Frequency to DDR5-5600MHz
  • Enter the Precision Boost Overdrive submenu
  • Set Precision Boost Overdrive to Enabled

Then save and exit the BIOS.

We re-ran the benchmarks and checked the performance increase compared to the default operation.

  • SuperPI 4M: +0.48%
  • Geekbench 5 (single): +0.46%
  • Geekbench 5 (multi): +2.86%
  • Cinebench R23 Single: +0.46%
  • Cinebench R23 Multi: +1.16%
  • CPU-Z V17.01.64 Single: +0.48%
  • CPU-Z V17.01.64 Multi: +0.86%
  • V-Ray 5: +0.04%
  • AI Benchmark: +6.80%
  • 3DMark Night Raid: +0.99%
  • CS:GO FPS Bench: +0.01%
  • Tomb Raider: +0.00%
  • Final Fantasy XV: +0.02%
7600x pbo expo benchmark performance

Here are the 3DMark CPU Profile scores at stock

  • CPU Profile 1 Thread: +0.18%
  • CPU Profile 2 Threads: +0.32%
  • CPU Profile 4 Threads: +1.58%
  • CPU Profile 8 Threads: +1.45%
  • CPU Profile 16 Threads: +0.65%
  • CPU Profile Max Threads: +0.58%
7600x pbo expo 3dmark performance

As expected, enabling Precision Boost Overdrive does little to improve the system’s performance. After all, the CPU is limited by operating temperature and programmed Fmax, not by any platform power parameters. Thanks to EXPO, we see a performance improvement of  6.80% in AI Benchmark.

When running Prime 95 Small FFTs with AVX-512 enabled, the average CPU effective clock is 4945 MHz with 1.335 volts. The average CPU temperature is 95.3 degrees Celsius. The ambient and water temperature is 26.7 and 34.1 degrees Celsius. The average CPU package power is 135.8 watts.

7600x pbo expo prime95 avx

When running Prime 95 Small FFTs with AVX disabled, the average CPU effective clock is 5021 MHz with 1.384 volts. The average CPU temperature is 95.3 degrees Celsius. The ambient and water temperature is 26.4 and 33.9 degrees Celsius. The average CPU package power is 131.0 watts.

7600x pbo expo prime95 non-avx

OC Strategy #2: AI Overclocking

In our second overclocking strategy, we use the Asus AI Overclocking feature integrated into the ASUS ROG BIOS.

ASUS AI Overclocking

For many years board vendors have tried to implement automatic overclocking features in their BIOS for more straightforward performance enhancement. This has always been a mixed bag, as most preset OC profiles are overly optimistic in frequency target or overly generous with the voltage selection. So often, you end up with a slightly unstable or overheating system.

ASUS AI overclocking uses a unique strategy for automatic overclocking. Instead of working with preset frequency and voltage profiles, the system will monitor the CPU and cooling system throughout an initial testing phase. Based on its findings, it will then predict the optimal settings. The system will automatically guide the overclocking process and adjust voltages and frequency to match your cooling system.

The better your cooling, the higher your AI overclock.

ASUS introduced AI Overclocking on its Z490 ROG motherboards as a next-generation automatic overclocking technology. Since then, it’s been present on every next Intel platform; however, not on any AMD platform. That changes with the AM5.

There are three steps to enabling AI overclocking. First, reset the BIOS to default settings. Then, reboot and enter the operating system. Run a couple of heavy workloads such as Prime95, Realbench, or Intel XTU for 10 to 30 minutes. Then return to the BIOS and enter the AI OC Guide menu from the top. Make sure to read through the explanation and click Enable AI when ready.

In addition to automatic overclocking, AI Overclocking provides a lot of advanced information and suggestions in the AI Features menu. The information includes:

  • P0 VID and SP values for each CPU core
  • Precision Boost Overdrive 2 suggested overclocking parameters
  • Dynamic OC Switcher suggested overclocking parameters

The SP value is based on the combination of maximum boost frequency, temperature, and P0 VID. Generally, it indicates the quality of a particular core. A higher SP value would indicate a better-quality core with superior overclocking capabilities, though it’s not an exact science. The overclocking suggestions are based on a continued evaluation of your CPU thermal solution.

After enabling AI Overclock, the following settings have changed:

  • CPU Core Ratio: AI Optimized
    • Preference: Performance
  • Precision Boost Overdrive: Enabled
    • Medium Load Boostit: Enabled
    • Max CPU Boost Clock Override: +100MHz
    • Scalar: 8X
  • Dynamic OC Switcher: Enabled
    • CCX0 Ratio: 52.75X
    • Core VID: 1.268V
    • Current Threshold: 48A

As you can see, AI Overclock provides a conservative +100 MHz increase of the Precision Boost maximum frequency. It also boosts the CPU to 5275 MHz in all-core workloads by enabling Dynamic OC Switcher. We’ll get back to the details on Dynamic OC Switcher later in this article.

BIOS Settings & Benchmark Results

Upon entering the BIOS

  • Enter the Ai Overclocking Guide
  • Go through the guide, then click Enable AI
  • Switch to the Advanced Mode view
  • Go to the Ai Tweaker menu
  • Set Ai Overclock Tuner to EXPO II
  • Set Memory Frequency to DDR5-5600MHz
  • Ensure CPU Core Ratio is set to AI Optimized and Preference is set to performance

Then save and exit the BIOS.

We re-ran the benchmarks and checked the performance increase compared to the default operation.

  • SuperPI 4M: +1.75%
  • Geekbench 5 (single): +1.80%
  • Geekbench 5 (multi): +4.25%
  • Cinebench R23 Single: +0.92%
  • Cinebench R23 Multi: +5.37%
  • CPU-Z V17.01.64 Single: +1.22%
  • CPU-Z V17.01.64 Multi: +3.45%
  • V-Ray 5: +3.41%
  • AI Benchmark: +7.71%
  • 3DMark Night Raid: +1.15%
  • CS:GO FPS Bench: +0.49%
  • Tomb Raider: +0.00%
  • Final Fantasy XV: -0.61%
7600x pbo ai overclock benchmark performance

Here are the 3DMark CPU Profile scores at stock

  • CPU Profile 1 Thread: +1.09%
  • CPU Profile 2 Threads: +1.01%
  • CPU Profile 4 Threads: +2.30%
  • CPU Profile 8 Threads: +1.97%
  • CPU Profile 16 Threads: +2.34%
  • CPU Profile Max Threads: +1.56%
7600x pbo ai overclock 3dmark performance

Increasing the Precision Boost maximum CPU frequency ceiling by 100 MHz is an increase of less than 2% over the standard limit of 5450 MHz. So, we don’t expect that much in terms of performance improvement. In all-core multi-threaded workloads, we get a more significant improvement of up to 5% with the manual overclock empowered by Dynamic OC Switcher.

When running Prime 95 Small FFTs with AVX-512 enabled, unfortunately, the CPU is not stable. AI Overclock targets the typical use-case and stability requirements, not a worst-case scenario like Prime95 with AVX.

When running Prime 95 Small FFTs with AVX disabled, the average CPU effective clock is 5275 MHz with 1.278 volts. The average CPU temperature is 83 degrees Celsius. The ambient and water temperature is 26.9 and 33.9 degrees Celsius. The average CPU package power is 111.9 watts.

7600x pbo ai overclock prime95 non-avx

OC Strategy #3: PBO Tuned with Curve Optimizer

In our third overclocking strategy, we will use the Curve Optimizer tool included in the Precision Boost Overdrive overclocker’s toolkit.

PBO 2: Curve Optimizer

As I mentioned earlier in this blog post, Curve Optimizer is an important new feature of Precision Boost Overdrive 2.

Curve Optimizer allows end-users to adjust the factory-fused VFT curve, or voltage-frequency-temperature curve, for each CPU core separately. The VFT curve is a unique curve for each core inside your CPU that defines the required voltage for a given frequency at a given temperature. Higher frequencies or higher operating temperatures require higher voltage.

Curve Optimizer adjusts the VFT curve by offsetting the voltages of the factory-fused VFT curve. By setting a positive offset, you increase the voltage point. Conversely, you decrease the voltage point by setting a negative offset.

You can offset the entire curve by up to 30 steps in a positive direction and up to 300 steps in a negative direction. Each step represents approximately 5 to 7 mV.

The traditional overclocking approach for AMD Ryzen CPUs is to set a negative curve optimizer. Two things happen when you adjust the VFT curve with a negative point offset.

  1. You effectively tell the CPU that it needs less voltage for a given frequency. And, as a consequence, at a given voltage, it can apply a higher frequency. So, when the Precision Boost 2 algorithm determines sufficient power and temperature headroom to use 1.35V, with the negative point offset, it will target a higher frequency.
  2. The CPU temperature will be lower because you use less voltage at a given frequency. That extra thermal headroom will also encourage the Precision Boost algorithm to target higher voltages and frequencies.

In my AMD Raphael overclocking launch article, I explored the ins and outs of Curve Optimizing with Precision Boost Overdrive 2. If you want to learn more about the impact of Curve Optimizer, I suggest you check out that post.

As I mentioned already, Curve Optimizer is available on a per-core basis. In the past, per-core tuning offered a tangible benefit as negative curve optimizer settings would provide some cores with much more frequency headroom. Unfortunately, the frequency headroom on this Ryzen 5 7600X is quite limited.

Curve Optimizer Tuning Process

The manual tuning process for Curve Optimizer can become quite convoluted since it affects the CPU core voltage in all scenarios ranging from very light single-threaded workloads to heavy all-core workloads.

My Curve Optimizer process for this Ryzen 5 7600X was as follows.

First, I use a broad tuning approach to get a big-picture understanding of the tuning opportunity. Then, I narrow down the per-core curve optimizer settings to achieve stability in a light workload. Then, I verify the stability in all-core multi-threaded workloads. Lastly, I run through all my benchmarks to validate stability in a wide range of test scenarios.

The broad tuning process includes:

  • Using Shamino’s Boost Curve tool to track the average frequency across an increasing amount of active threads
  • Gradually increasing an all-core negative curve optimizer until the point of instability
  • Gradually increasing Fmax Boost Override in case the boost frequency is hitting the frequency ceiling

The narrow tuning process includes:

  • Using SuperPI 32M with affinity to a specific thread to check the maximum effective clock frequency
  • Gradually increasing the per core negative curve optimizer until the point of instability for each core
  • Gradually increasing Fmax Boost Override in case the boost frequency is hitting the frequency ceiling

The all-core stability process includes running Prime 95 Small FFTs with both AVX-512 enabled and disabled.

The benchmark validation process includes completing all the benchmarks I’ve included in this guide. If a benchmark is unstable, I increase every core’s per-core curve optimizer value by 2 until I reach stability.

I will show you the BIOS configuration in a minute. But first, note that the per-core curve optimizer settings are CPU-specific, and the optimal values of your CPU may differ substantially.

7600x pbo curve optimized parameters

The following table shows the boost profile for this specific CPU. I use two applications to characterize the boost: SuperPI 32M, assigned to a particular thread, for checking the 1T light load boost, and Shamino’s Boost Curve, which traces the average frequency as increasingly more threads are working on a medium-load workload.

7600x pbo curve optimized boost chart
7600x pbo curve optimized boost profile

As you can see, at stock, every core hits boost frequencies slightly over the listed maximum boost frequency of 5450 MHz. While most cores are close to the VID limit of 1.475V, Core 1 and Core 5 seem to offer plenty of voltage headroom.

When we lift the frequency ceiling with a +200 MHz Fmax boost override with Zen 3, we’d see the maximum boost frequency rise spectacularly. Especially on cores with a lot of voltage headroom. Surprisingly, on this Ryzen 5 7600X, we see slight improvement as most cores are limited to 5462 MHz, and the two best cores only reach 5500 MHz. That’s well below our ceiling of 5650 MHz.

Based on previous Ryzen overclocking experience, adding an all-core negative curve optimizer should help lift the maximum boost frequency. Again, that’s not what we get. Not only is our negative curve optimizer range limited to -12, but the resulting frequency increase is also very minor.

In the end, I found that the most stable configuration is a curve optimizer setting of -12 and a Fmax ceiling of 5500 MHz (+50 over default). For the other Precision Boost parameters, we just use the ones from our previous overclocking strategy.

BIOS Settings & Benchmark Results

Upon entering the BIOS

  • Switch to the Advanced Mode view
  • Go to the Ai Tweaker menu
  • Set Ai Overclock Tuner to EXPO II
  • Set Memory Frequency to DDR5-5600MHz
  • Enter the Precision Boost Overdrive submenu
    • Set Precision Boost Overdrive to Enabled
    • Set CPU Boost Clock Override to Enabled (Positive)
    • Set Max CPU Boost Clock Override to 50
    • Enter the Curve Optimizer submenu
      • Set Curve Optimizer to All Cores
      • Set All Core Curve Optimizer Sign to –
      • Set All Core Curve Optimizer Magnitude to 12
    • Leave the Curve Optimizer submenu

Then save and exit the BIOS.

We re-ran the benchmarks and checked the performance increase compared to the default operation.

  • SuperPI 4M: +1.27%
  • Geekbench 5 (single): +1.39%
  • Geekbench 5 (multi): +4.28%
  • Cinebench R23 Single: +0.97%
  • Cinebench R23 Multi: +2.83%
  • CPU-Z V17.01.64 Single: +2.16%
  • CPU-Z V17.01.64 Multi: +2.52%
  • V-Ray 5: +1.72%
  • AI Benchmark: +8.49%
  • 3DMark Night Raid: +2.13%
  • CS:GO FPS Bench: +1.16%
  • Tomb Raider: +2.06%
  • Final Fantasy XV: +0.30%
7600x pbo curve optimized benchmark performance

Here are the 3DMark CPU Profile scores at stock

  • CPU Profile 1 Thread: +1.37%
  • CPU Profile 2 Threads: +1.38%
  • CPU Profile 4 Threads: +3.83%
  • CPU Profile 8 Threads: +2.27%
  • CPU Profile 16 Threads: +3.17%
  • CPU Profile Max Threads: +3.10%
7600x pbo curve optimized 3dmark performance

Unsurprisingly, the performance uplift after Curve Optimizing is very limited. After all, we could only increase the frequency ceiling by 50 MHz. We see a slightly better improvement in multi-threaded applications, likely due to the lower voltage used after curve optimization. We see the highest performance improvement of +8.49% in AI Benchmark.

When running Prime 95 Small FFTs with AVX-512 enabled, the average CPU effective clock is 5051 MHz with 1.324 volts. The average CPU temperature is 95.3 degrees Celsius. The ambient and water temperature is 26.9 and 34.4 degrees Celsius. The average CPU package power is 135.7 watts.

7600x pbo curve optimized prime95 avx

When running Prime 95 Small FFTs with AVX disabled, the average CPU effective clock is 5126 MHz with 1.373 volts. The average CPU temperature is 95.4 degrees Celsius. The ambient and water temperature is 26.9 and 34.4 degrees Celsius. The average CPU package power is 130.8 watts.

7600x pbo curve optimized prime95 non-avx

OC Strategy #4: PB Supercharged with ECLK

In our fourth overclocking strategy, we take advantage of the return of the ECLK mode. ECLK stands for external clock and is precisely what the term suggests: an external clock generator. It was previously available on Ryzen 2000 Pinnacle Ridge processors but was removed afterward.

Raphael ECLK Overview

The standard Raphael platform has a 48 MHz crystal input to the integrated CGPLL clock generator. The CGPLL then generates a 48 MHz clock for the USB PLL and a 100 MHz reference clock for the FCH, which contains the CCLK PLL for the CPU cores and several SOC PLLs.

In addition to the standard internal CGPLL, Raphael supports up to two external clock modes. They’re called eCLK0 Mode and eCLK1 Mode.

In eCLK0 Mode, an external 100MHz reference clock is used for both the CPU and SOC PLLs. In other words, it’s a reference clock that affects the CPU core clocks and the PCIe and SATA clocks. The USB PLL is still driven by the 48 MHz crystal via the CGPLL.

In eCLK1 Mode, there are two distinct external 100MHz reference clocks. One clock provides the 100MHz input for the CPU PLL, and another provides the 100MHz reference clock for the SOC PLLs. AMD suggests up to 140 MHz can be expected for the CPU core reference clock, but your mileage may vary. The USB PLL is still driven by the 48 MHz crystal via the CGPLL.

In previous SkatterBencher guides, I referred to the technique of overclocking the reference clock as supercharging PBO. Unlike previous Ryzen CPUs, the overclocking of the reference clock is very well supported and provides a viable path to overclocking.

The overclocking strategy with ECLK is the polar opposite of what we’re used to with Ryzen CPUs. Typically, as we showed in OC Strategy #3, Ryzen overclocking consisted of using a negative curve optimizer. This would offset the CPU core VFT curve and achieve higher boost frequencies.

With ECLK, we still build on the factory-fused VFT curve but adjust the frequency by adjusting the reference clock. For example, if the Precision Boost has a VFT point for 5000 MHz at 1.25V at 50C, with an ECLK of 105 MHz, the actual point will be 5250 MHz at 1.25V at 50C.

Obviously, the default voltage for this VFT point won’t suffice for stable operation. So, counter-intuitively, we use a positive curve optimizer to increase the voltages of the VFT point. For example, a +30 curve optimizer may increase the voltage by 150mV. Thus, the resulting VFT point will be 5250 MHz at 1.40V at 50C.

Suppose Curve Optimizer doesn’t provide you with sufficient additional voltage. In that case, you can always add a voltage offset via the VRM controller configuration. As you’ll see in a bit, that’s what I had to do in this strategy.

One more important point: ECLK also affects the maximum frequency ceiling. With an ECLK of 109.25 MHz, the new Fmax for the 7600X is 5450 MHz x 1.0925 = 5954 MHz. That’s obviously way too high for this CPU. Luckily, we can use a negative Fmax boost override to fix this.

ECLK Tuning Process

The manual tuning process for eCLK tuning can become quite convoluted since it affects the CPU core stability in all scenarios ranging from very light single-threaded workloads to heavy all-core workloads.

My ECLK tuning process for this Ryzen 5 7600X is similar to the Curve Optimizer process.

First, I use a broad tuning approach to get a big-picture understanding of the tuning opportunity. Then, I narrow down the per-core curve optimizer settings to achieve stability in a light workload. Then, I verify the stability in all-core multi-threaded workloads. Lastly, I run through all my benchmarks to validate stability in a wide range of test scenarios.

The broad tuning process includes:

  • Using Shamino’s Boost Curve tool to track the average frequency across an increasing amount of active threads
  • Gradually increasing the ECLK frequency until the point of instability
  • In case of instability, reduce the programmed Fmax by configuring a negative boost override

The narrow tuning process includes:

  • Using SuperPI 32M with affinity to a specific thread to check the maximum effective clock frequency
  • Gradually reduce the negative Fmax Boost Override in case the boost frequency is hitting the frequency ceiling
  • In case of instability, set a positive per-core curve optimizer

The all-core stability process includes running Prime 95 Small FFTs with both AVX-512 enabled and disabled.

The benchmark validation process includes completing all the benchmarks I’ve included in this guide. If a benchmark is unstable, I increase every core’s per-core curve optimizer value by 2 until I reach stability.

In my case, I used an ECLK of 109.25 MHz, a Curve Optimizer of +30, and an additional voltage offset of +50mV. I use a Fmax boost override of -375, resulting in a new Fmax ceiling of (5450 – 375) x 1.0925 = 5544 MHz.

7600x pbo eclk parameters

I will show you the BIOS configuration in a minute. First, please note that these settings, especially the curve optimizer, are CPU-specific, and the optimal values of your CPU may differ substantially.

Again, I checked the boost profile for this specific CPU using the two previously mentioned.

7600x pbo eclk boost chart
7600x pbo eclk boost profile

As you can see from the table, the ECLK approach provides an additional 44MHz higher maximum boost frequency for each core and a very similar boost curve.

BIOS Settings & Benchmark Results

Upon entering the BIOS

  • Switch to the Advanced Mode view
  • Go to the Ai Tweaker menu
  • Set Ai Overclock Tuner to EXPO II
  • Set eCLK Mode to Asynchronous
  • Set BCLK2 Frequency to 109.25 MHz
  • Set Memory Frequency to DDR5-5600MHz
  • Enter the Precision Boost Overdrive submenu
    • Set Precision Boost Overdrive to Enabled
    • Set CPU Boost Clock Override to Enabled (Negative)
    • Set Max CPU Boost Clock Override to 375
    • Enter the Curve Optimizer submenu
      • Set Curve Optimizer to All Cores
      • Set All Core Curve Optimizer Sign to +
      • Set All Core Curve Optimizer Magnitude to 30
    • Leave the Curve Optimizer submenu
  • Leave the Precision Boost Overdrive submenu
  • Set CPU Core Voltage to Offset Mode
    • Set Offset Mode Sign to +
    • Set CPU Core Voltage Offset to 0.05

Then save and exit the BIOS.

We re-ran the benchmarks and checked the performance increase compared to the default operation.

  • SuperPI 4M: +2.44%
  • Geekbench 5 (single): +2.27%
  • Geekbench 5 (multi): +4.68%
  • Cinebench R23 Single: +1.33%
  • Cinebench R23 Multi: +3.27%
  • CPU-Z V17.01.64 Single: +1.79%
  • CPU-Z V17.01.64 Multi: +3.03%
  • V-Ray 5: +2.31%
  • AI Benchmark: +8.80%
  • 3DMark Night Raid: +2.63%
  • CS:GO FPS Bench: +1.56%
  • Tomb Raider: +3.09%
  • Final Fantasy XV: +0.56%
7600x pbo eclk benchmark performance

Here are the 3DMark CPU Profile scores at stock

  • CPU Profile 1 Thread: +2.19%
  • CPU Profile 2 Threads: +1.47%
  • CPU Profile 4 Threads: +5.33%
  • CPU Profile 8 Threads: +2.67%
  • CPU Profile 16 Threads: +4.40%
  • CPU Profile Max Threads: +3.17%
7600x pbo eclk 3dmark performance

While the performance improvement over stock is limited, we still see the highest performance of any overlocking strategy. We get the best performance improvement of +8.80% in AI Benchmark

When running Prime 95 Small FFTs with AVX-512 enabled, the average CPU effective clock is 5051 MHz with 1.324 volts. The average CPU temperature is 95.3 degrees Celsius. The ambient and water temperature is 26.9 and 34.4 degrees Celsius. The average CPU package power is 135.7 watts.

7600x pbo eclk prime95 avx

When running Prime 95 Small FFTs with AVX disabled, the average CPU effective clock is 5126 MHz with 1.373 volts. The average CPU temperature is 95.4 degrees Celsius. The ambient and water temperature is 26.9 and 34.4 degrees Celsius. The average CPU package power is 130.8 watts.

7600x pbo eclk prime95 non-avx

OC Strategy #5: Manual Overclock

In our fifth overclocking strategy, we will pursue a manual overclock.

One could question the use-case for manual overclocking an AMD Ryzen CPU. Just like with all past Ryzen processors, the major downside of manual overclocking is that you lose the benefits of Precision Boost technology in low-threaded benchmark applications. So, whereas this 7600X can boost up to 5450 MHz with Precision Boost, it will be limited to your set fixed frequency when manually overclocking.

However, manual overclocking is not all negative and even has some benefits.

  1. Automatic overclocking and frequency boosting technologies leave a little margin on the table. We can exploit this margin and fine-tune it for application stability when manually overclocking with our specific hardware configuration.
  2. In multi-threaded applications, Precision Boost Overdrive applies a single frequency to every core. However, on AMD Ryzen CPUs, you can set the frequency for each CCX separately. So, we can exploit the fact that some CCXs may overclock better than others. This point is not necessarily relevant for the 7600X, which only has one CCX, but it’s good to know.
  3. By setting a fixed voltage, we avoid the CPU dynamically and rapidly updating its VID requests to the VRM controller. This alleviates stress on the VRM and typically yields lower temperatures.

To better understand the performance tuning opportunities embedded in the Ryzen 5 7600X processor, let’s look at its topology in more detail.

CPU Core Clocking Topology

The clocking of AMD Raphael is similar to the previous generation of Zen 3 Vermeer desktop CPUs.

The standard Raphael platform has a 48 MHz crystal input to the integrated CGPLL clock generator. The CGPLL then generates a 48 MHz clock for the USB PLL and a 100 MHz reference clock for the FCH, which contains the CCLK PLL for the CPU cores and several SOC PLLs.

The CCLK PLL 100MHz reference clock drives the 200 MHz VCO, which is then multiplied by an FID and divided by a DID. As a whole, this provides CPU clock frequency granularity of 25 MHz.

As with Vermeer, each CCX has its own PLL, with the cores within that CCX running at the same frequency. In a typical operation, all cores within a CCX will run at the same frequency. The effective clock will differ if the core is in a different P-state.

The SOC PLLs include a wide range of PLLs present on the IO die. The ones most relevant for overclocking are:

  • FCLK for the data fabric
  • UCLK for the memory controller
  • MCLK for the system memory
  • GFXCLK for the integrated graphics

The SOC PLLs are not particularly relevant for manual CPU Core overclocking.

As I mentioned in OC Strategy #4, in addition to the standard internal CGPLL, Raphael supports up to two external clock modes. We won’t be using eCLK in this overclocking strategy.

CPU Core Voltage Topology

From the voltage topology perspective, there are a few minor changes. Like Vermeer, the processor still relies on an internal and external power supply to generate the processor voltages.

There are four primary power supplies from the motherboard VRM to the processor: VDDCR, VDDCR_SOC, VDDCR_MISC, and VDDIO_MEM_S3. I have covered the details extensively in my Raphael overclocking launch article.

The VDDCR voltage rail provides the external power for two internal voltage regulators: VDDCR_CPU and VDDCR_VDDM.

VDDCR_CPU provides the voltage for the CPU cores in CCX. On CPUs with multiple CCXs, each CCX has its own VDDCR_CPU voltage rail, but the voltage will be identical. The voltage rails can work in either regular or bypass mode, but on Raphael, it is always in bypass mode. That means the voltage is always equal to the VDDCR external voltage.

VDDCR_VDDM provides the voltage for the L2, L3, and, if present, 3D V-Cache on a CCX. On CPUs where there are multiple CCXs, each CCX has its own VDDCR_VDDM voltage rail. This rail cannot work in bypass mode; therefore, it is internally regulated from the VDDCR external voltage rail. The default VDDM voltage is 0.95V

The VDDCR voltage rail can be directly controlled via the SVI3 interface. VDDCR_VDDM cannot be controlled by the end-user.

Choosing the proper manual voltage is always a matter of finding a suitable trade-off between three factors: increased overclocking potential, the thermal challenges that come with increased voltage, and of course, consideration of CPU lifespan.

CCX Frequency Tuning Process

With that last thought, we kick off our manual overclocking process.

Our limiting factor will ultimately be the cooling solution as power consumption increases exponentially with operating voltage and temperature scales (somewhat) linearly with power consumption.

The maximum voltage will be determined by the application we’re tuning for. So, the first step in our tuning process would be deciding on the stress test representing our worst-case scenario. Usually, I pick Prime95 Small FFTs with AVX512 enabled, but for the Ryzen 5 7600X, I think it’s reasonable to forego the AVX requirement.

The next step is to set a fixed CPU ratio and check the maximum temperature when running our workload. If there’s thermal headroom left, increase the operating voltage.

Once we know the maximum voltage, we can tune the CCX ratio. Simply increase the CPU ratio until the application shows instability, then back off. This approach will give you the maximum stable per-CCX frequency for a given voltage.

In our case, we end up with a CPU ratio of 53.75X and a Core VID of 1.365V. Our worst-case stress test yields a CPU voltage of 1.379V and a CPU temperature of 96 degrees Celsius.

BIOS Settings & Benchmark Results

Upon entering the BIOS

  • Switch to the Advanced Mode view
  • Go to the Ai Tweaker menu
  • Set Ai Overclock Tuner to EXPO II
  • Set Memory Frequency to DDR5-5600MHz
  • Enter the CPU Core Ratio (Per CCX) submenu
    • Set Core VID to 1.365
    • Set CCX0 Ratio to 53.75

Then save and exit the BIOS.

We re-ran the benchmarks and checked the performance increase compared to the default operation.

  • SuperPI 4M: -3.94%
  • Geekbench 5 (single): -0.60%
  • Geekbench 5 (multi): +4.54%
  • Cinebench R23 Single: -1.17%
  • Cinebench R23 Multi: +6.78%
  • CPU-Z V17.01.64 Single: -1.05%
  • CPU-Z V17.01.64 Multi: +4.25%
  • V-Ray 5: +3.82%
  • AI Benchmark: +8.75%
  • 3DMark Night Raid: +1.95%
  • CS:GO FPS Bench: +0.65%
  • Tomb Raider: +0.52%
  • Final Fantasy XV: +0.12%
7600x manual overclock benchmark performance

Here are the 3DMark CPU Profile scores

  • CPU Profile 1 Thread: -0.82%
  • CPU Profile 2 Threads: -0.50%
  • CPU Profile 4 Threads: +2.52%
  • CPU Profile 8 Threads: +1.82%
  • CPU Profile 16 Threads: +4.50%
  • CPU Profile Max Threads: +4.40%
7600x manual overclock 3dmark performance

As expected, we see a performance deficit against stock performance in single-threaded applications as manual overclocking loses the benefit of the Precision Boost 1T frequency. In multi-threaded applications, however, we see a nice performance uplift up to +8.75% in AI Benchmark.

When running Prime 95 Small FFTs with AVX-512 enabled, unfortunately, the CPU is not stable. This is a deliberate choice as we decided that Prime95 with AVX512 enabled is not a representative worst-case workload for our system.

When running Prime 95 Small FFTs with AVX disabled, the average CPU effective clock is 5375 MHz with 1.376 volts. The average CPU temperature is 96.0 degrees Celsius. The ambient and water temperature is 24.2 and 31.5 degrees Celsius. The average CPU package power is 136.4 watts.

7600x manual overclock prime95 non-avx

OC Strategy #6: Dynamic OC Switcher

In our final overclocking strategy, we rely on Dynamic OC Switcher to bring together the best of PBO and manual overclocking.

ASUS Dynamic OC Switcher

Dynamic OC Switcher, or DOS for short, is ASUS’ clever way of addressing a core challenge when overclocking AMD Ryzen processors. It was first introduced with the ASUS ROG Crosshair VIII Dark Hero motherboard and has since been included in several other ASUS motherboards.

As we know from the Raphael CPU Overclocking overview, there are two approaches to AMD Ryzen CPU core overclocking: Precision Boost Overdrive and OC mode.

  • Precision Boost Overdrive has the benefit of retaining all the automatic boost algorithms, including the peak single thread frequency and associated performance.
  • OC Mode has the benefit of enabling precise per-CCX fine-tuning for the maximum stable all-core frequency.

So, you typically have to choose to compromise either the best single-threaded performance or the best multi-threaded performance.

ASUS Dynamic OC Switcher gives us the best of both worlds, allowing the system to actively switch between Precision Boost Overdrive and manual OC Mode. It’s most easy to conceptualize DOS as retaining all automatic Precision Boost benefits but with a manually configured frequency floor for all-core workloads.

DOS requires little additional configuration work. We need to know two things:

  1. What is the lowest frequency we will allow
  2. At which point do we want DOS to switch between PBO and OC Mode

Sadly, we cannot simply configure a minimum frequency and have the system switch based on that. Instead, we need to use a proxy metric: a specific current or temperature threshold. There is no perfect method of determining the ideal threshold, so I’ll show you one example using Prime95 Small FFTs without AVX.

The first step is to determine your desired manual overclock. The fastest way to get to this point is by selecting your stress test tool of choice and finding the maximum CPU voltage within your thermal budget. Afterward, you can determine the maximum stable frequency for that voltage and temperature. In our case, we can rely on our overclocking settings from OC Strategy #5

  • Stress Test: Prime95 non-AVX
  • OCVID: 1.365V (1.376V under load)
  • CCX0 Ratio: 53.75

We can write the settings down and switch back to tuning with Precision Boost Overdrive.

The next step is to apply all your Precision Boost Overdrive tuning settings, including any custom Curve Optimizer or Fmax Override settings. In our case, we can rely on the settings from OC Strategy #4.

Then, go into the operating system and open your stress test tool and HWiNFO. We aim to find where the Precision Boost frequency drops below our target manual overclock of 5375 MHz and check the current use. In HWiNFO, we will monitor the CPU Core Effective Clocks and ASUS EC VRM Vcore Current.

Then we start the Prime95 non-AVX stress test and change the affinity to 1 core in Task Manager. Now monitor the core clock frequency. It will be higher than our target of 5375 MHz.

Now you can gradually increase the active thread count. When you reach 5375 MHz or below, check the ASUS EC VRM Vcore Current reading. This value will be our input for the DOS Current Threshold setting.

In my case, we reach the frequency of 5375 MHz at 3 active threads, and the current reads about 45A. This will be our Current Threshold value for Dynamic OC Switcher.

So, to reiterate what’s happening: Dynamic OC Switcher will actively switch between OC mode and Precision boost when the CPU current hits 45A. Anything above 45A engages manual OC mode; anything below 45A will engage Precision Boost.

Before we forget, one final thing!

Our Precision Boost Overdrive ECLK OC strategy includes an adjusted ECLK frequency of 109.25 MHz and a manual voltage offset of +50mV. When Dynamic OC Switcher switches between PBO and OC mode, it does not change these settings. So, we must adjust our manual overclock settings accordingly:

  • Core VID from 1.365 to 1.305
  • CCX0 Ratio from 53.75 to 49.00

BIOS Settings & Benchmark Results

Upon entering the BIOS

  • Switch to the Advanced Mode view
  • Go to the Ai Tweaker menu
  • Set Ai Overclock Tuner to EXPO II
  • Set eCLK Mode to Asynchronous
  • Set BCLK2 Frequency to 109.25 MHz
  • Set Memory Frequency to DDR5-5600MHz
  • Enter the CPU Core Ratio (Per CCX) submenu
    • Set Core VID to 1.305
    • Set CCX0 Ratio to 49
    • Set Dynamic OC Switcher to Enabled
    • Set Current Threshold to Switch to OC Mode to 45
  • Leave the CPU Core Ratio (Per CCX) submenu
  • Set CPU Core Voltage to Offset Mode
    • Set Offset Mode Sign to +
    • Set CPU Core Voltage Offset to 0.05

Then save and exit the BIOS.

We re-ran the benchmarks and checked the performance increase compared to the default operation.

  • SuperPI 4M: +2.30%
  • Geekbench 5 (single): +2.55%
  • Geekbench 5 (multi): +5.25%
  • Cinebench R23 Single: +1.12%
  • Cinebench R23 Multi: +6.07%
  • CPU-Z V17.01.64 Single: +1.86%
  • CPU-Z V17.01.64 Multi: +4.14%
  • V-Ray 5: +3.93%
  • AI Benchmark: +8.68%
  • 3DMark Night Raid: +2.62%
  • CS:GO FPS Bench: +1.81%
  • Tomb Raider: +3.09%
  • Final Fantasy XV: +0.36%
7600x dynamic oc switcher benchmark performance

Here are the 3DMark CPU Profile scores at stock

  • CPU Profile 1 Thread: +2.01%
  • CPU Profile 2 Threads: +1.70%
  • CPU Profile 4 Threads: +2.84%
  • CPU Profile 8 Threads: +3.27%
  • CPU Profile 16 Threads: +3.70%
  • CPU Profile Max Threads: +3.68%
7600x dynamic oc switcher 3dmark performance

On the upside, we see the highest performance in all our benchmarks. On the downside, it’s a little disappointing that after hours of testing and tuning, we can only squeeze less than 10% extra performance. We see the maximum performance improvement of +8.68% in AI Benchmark

When running Prime 95 Small FFTs with AVX-512 enabled, unfortunately, the CPU is not stable. This is a deliberate choice on our side as we decided that Prime95 with AVX512 enabled is not a representative worst-case workload for our system.

When running Prime 95 Small FFTs with AVX disabled, the average CPU effective clock is 5305 MHz with 1.379 volts. The average CPU temperature is 94.7 degrees Celsius. The ambient and water temperature is 25.0 and 32.1 degrees Celsius. The average CPU package power is 126.6 watts.

7600x dynamic oc switcher prime95 non-avx

AMD Ryzen 5 7600X: Conclusion

Alright, let us wrap this up.

The Ryzen 5 7600X overclocking experience was pretty rough. Not because we cannot reach high frequencies but because AMD left little to no overclocking headroom on the table. The Ryzen 7000 chips are definitely pushed to the limit out of the box. The maximum frequency we could squeeze out of this particular chip was 5550 MHz, not even 300 MHz more than the listed maximum boost frequency.

Frankly, it’s a pity there’s so little headroom because the Ryzen 7000 overclocking tools are fantastic. ECLK adds a new dimension to Precision Boost tuning and provides us with new avenues for overclocking. Overall, I’d say the Raphael overclocking toolkit is the best we’ve seen from AMD (yet).

Ultimately, the overclocking judgment will boil down to what happens with future batches of Ryzen 7000 CPUs. If the overclocking headroom further improves such that 7600X CPUs can match the 5.8GHz boost of the 7950X, then this will be a great platform. If AMD left too little margin, we might see the worst chips not meet the stock boost specifications.

Anyway, that’s all for today!

I want to thank my Patreon supporters for supporting my work. As per usual, if you have any questions or comments, feel free to drop them in the comment section below. 

See you next time!

Liked it? Take a second to support Pieter on Patreon!

Leave a Reply