F5F Stay Refreshed Hardware Desktop Processing speed of the CPU

Processing speed of the CPU

Processing speed of the CPU

Pages (2): 1 2 Next
I
ItzCh3nTi_YT
Member
96
03-01-2018, 02:49 PM
#1
Hello, I'm curious about the effect of CPU clock/frequency on gaming performance. What are typical core/clock settings for smooth play at 720p, 1080p, and 1440p resolutions? I'm not focusing on basic or entry-level systems, but rather mid to high-end models. Could anyone run benchmarks with different clock speeds and share their results? My setup includes an x470 processor, R7 2700X, 32GB DDR4 at 3200MHz, RX6700XT graphics card, W11 motherboard, and NVMe SSD. At 720p I noticed noticeable differences in FPS, with CPU utilization around 50-70%. Do you have any insights into other scenarios where clock speed plays a significant role in gaming?
I
ItzCh3nTi_YT
03-01-2018, 02:49 PM #1

Hello, I'm curious about the effect of CPU clock/frequency on gaming performance. What are typical core/clock settings for smooth play at 720p, 1080p, and 1440p resolutions? I'm not focusing on basic or entry-level systems, but rather mid to high-end models. Could anyone run benchmarks with different clock speeds and share their results? My setup includes an x470 processor, R7 2700X, 32GB DDR4 at 3200MHz, RX6700XT graphics card, W11 motherboard, and NVMe SSD. At 720p I noticed noticeable differences in FPS, with CPU utilization around 50-70%. Do you have any insights into other scenarios where clock speed plays a significant role in gaming?

F
flyer78
Senior Member
425
03-01-2018, 08:26 PM
#2
To make frequency truly impactful, you must be CPU-intensive. For an authentic test, try adjusting low in competitive shooters or revert to lower settings in AC Valhalla and watch the visuals improve. You’ll notice the change.
F
flyer78
03-01-2018, 08:26 PM #2

To make frequency truly impactful, you must be CPU-intensive. For an authentic test, try adjusting low in competitive shooters or revert to lower settings in AC Valhalla and watch the visuals improve. You’ll notice the change.

C
CelticGila
Senior Member
454
03-02-2018, 05:28 AM
#3
It focuses more on the ability of a single core and, to some degree, multiple cores rather than just clock speed. Clock speed is only part of the picture, alongside IPC and branch prediction, among other things.
C
CelticGila
03-02-2018, 05:28 AM #3

It focuses more on the ability of a single core and, to some degree, multiple cores rather than just clock speed. Clock speed is only part of the picture, alongside IPC and branch prediction, among other things.

S
S1ayer11
Junior Member
29
03-03-2018, 02:54 AM
#4
It depends on how much work you assign to it and what other components are involved. For reference, let's say you have an RTX 3090 with 32GB of fast dual-channel RAM and an 8-core Alder Lake CPU with only the performance cores active. At speeds around 1-2 GHz, you'll notice a significant boost—about twice the frame rate in many games. Between 2-4GHz, gains will be roughly 30-100% depending on the title and settings. Switching from 4GHz to 5GHz won't make much difference. For most gamers who aren't doing heavy multitasking, the CPU isn't a major concern unless you're using a very powerful GPU at low settings.
S
S1ayer11
03-03-2018, 02:54 AM #4

It depends on how much work you assign to it and what other components are involved. For reference, let's say you have an RTX 3090 with 32GB of fast dual-channel RAM and an 8-core Alder Lake CPU with only the performance cores active. At speeds around 1-2 GHz, you'll notice a significant boost—about twice the frame rate in many games. Between 2-4GHz, gains will be roughly 30-100% depending on the title and settings. Switching from 4GHz to 5GHz won't make much difference. For most gamers who aren't doing heavy multitasking, the CPU isn't a major concern unless you're using a very powerful GPU at low settings.

S
Space_Triks
Member
106
03-14-2018, 01:44 AM
#5
Used low preset settings; CPU consumption stayed around 50-60%. Overall performance is a bit better, with only a slight increase. I notice a noticeable 10-15% gap now, but it doesn’t feel like a major issue or a performance bottleneck.
S
Space_Triks
03-14-2018, 01:44 AM #5

Used low preset settings; CPU consumption stayed around 50-60%. Overall performance is a bit better, with only a slight increase. I notice a noticeable 10-15% gap now, but it doesn’t feel like a major issue or a performance bottleneck.

F
freekieboy77
Member
52
03-14-2018, 06:45 AM
#6
I was looking for a standard spec like 8 cores/thread with 3.0+ GHz clock speed, but it won’t always limit performance in games.
F
freekieboy77
03-14-2018, 06:45 AM #6

I was looking for a standard spec like 8 cores/thread with 3.0+ GHz clock speed, but it won’t always limit performance in games.

D
DueZulu
Senior Member
252
03-14-2018, 08:40 AM
#7
It depends on the specific game. Aiming for "not CPU bound" might not be the best approach. Focus more on meeting a certain performance threshold for your needs. For example, aim for over 120 FPS most of the time in CS GO on X map and above 60 FPS 99.9% of the time. The importance of clock speed changes based on the CPU and RAM you have. There are many components involved. Usually, the CPU isn't something to fixate on—even older CPUs like the 2600k can still work well today for many tasks. Fast GPUs from 2012, such as the GTX 680, don’t have the same limitations.
D
DueZulu
03-14-2018, 08:40 AM #7

It depends on the specific game. Aiming for "not CPU bound" might not be the best approach. Focus more on meeting a certain performance threshold for your needs. For example, aim for over 120 FPS most of the time in CS GO on X map and above 60 FPS 99.9% of the time. The importance of clock speed changes based on the CPU and RAM you have. There are many components involved. Usually, the CPU isn't something to fixate on—even older CPUs like the 2600k can still work well today for many tasks. Fast GPUs from 2012, such as the GTX 680, don’t have the same limitations.

X
XJustPizzaX
Member
114
03-18-2018, 08:39 AM
#8
Technically, when analyzing percentage changes around a threshold, a drop or rise is generally classified as a loss or gain. For example, if a frequency of 4.0GHz decreases by 3.3GHz (a 17% reduction), the resulting FPS might stay stable or even improve slightly in this instance. This pattern repeats across different scenarios, offering a general understanding—though specific results will vary depending on the setup.
X
XJustPizzaX
03-18-2018, 08:39 AM #8

Technically, when analyzing percentage changes around a threshold, a drop or rise is generally classified as a loss or gain. For example, if a frequency of 4.0GHz decreases by 3.3GHz (a 17% reduction), the resulting FPS might stay stable or even improve slightly in this instance. This pattern repeats across different scenarios, offering a general understanding—though specific results will vary depending on the setup.

S
StijnDragonFly
Junior Member
3
03-18-2018, 04:15 PM
#9
After reviewing additional tests focusing on the mhz/fps relationship, it seems the optimal frequency for an 8/16 core setup is around 2700MHz (just under 60% lower than 60fps). Values shifting either way tend to be less ideal. I’m curious about whether similar trends exist with 4/8 or 6/12 core processors, or specifically with Intel chips.
S
StijnDragonFly
03-18-2018, 04:15 PM #9

After reviewing additional tests focusing on the mhz/fps relationship, it seems the optimal frequency for an 8/16 core setup is around 2700MHz (just under 60% lower than 60fps). Values shifting either way tend to be less ideal. I’m curious about whether similar trends exist with 4/8 or 6/12 core processors, or specifically with Intel chips.

I
InfinitePop
Junior Member
25
03-20-2018, 07:20 AM
#10
I would consider improvements in terms of Ahmdah's law https://en.wikipedia.org/wiki/Amdahl's_law. Picture a system with three key components: CPU, RAM, and graphics card. The time needed to render each frame—estimated at around 16ms for smooth performance—could break down as 75% GPU, 20% CPU, and 5% RAM, totaling roughly 12ms, then 3.2ms, 0.8ms respectively. If the GPU speed doubles, the frame time drops to about 10ms, which equals 100FPS. Now consider another scenario where the CPU speeds up twice. Its calculation time shrinks from 3.2ms to 1.6ms. The overall frame time becomes 14.4ms, roughly matching 70FPS. From a simple GHz standpoint... we don’t need to jump from 5GHz to 10GHz in this case. In reality, most CPUs perform similarly, especially at 1080p resolution. A 1080p display will show only minor differences between a 4K and a 1080p setup. Generally, changes in CPU performance aren’t linear—big jumps from low to high frequencies, but small variations otherwise. Notice the big impact when CPUs are truly different, like comparing a 5GHz to a 4GHz processor; that gap is substantial. When it comes to overall feel, it’s more about how often the CPU really matters rather than constantly being the bottleneck. Those critical frames where CPU shines can make a noticeable difference.
I
InfinitePop
03-20-2018, 07:20 AM #10

I would consider improvements in terms of Ahmdah's law https://en.wikipedia.org/wiki/Amdahl's_law. Picture a system with three key components: CPU, RAM, and graphics card. The time needed to render each frame—estimated at around 16ms for smooth performance—could break down as 75% GPU, 20% CPU, and 5% RAM, totaling roughly 12ms, then 3.2ms, 0.8ms respectively. If the GPU speed doubles, the frame time drops to about 10ms, which equals 100FPS. Now consider another scenario where the CPU speeds up twice. Its calculation time shrinks from 3.2ms to 1.6ms. The overall frame time becomes 14.4ms, roughly matching 70FPS. From a simple GHz standpoint... we don’t need to jump from 5GHz to 10GHz in this case. In reality, most CPUs perform similarly, especially at 1080p resolution. A 1080p display will show only minor differences between a 4K and a 1080p setup. Generally, changes in CPU performance aren’t linear—big jumps from low to high frequencies, but small variations otherwise. Notice the big impact when CPUs are truly different, like comparing a 5GHz to a 4GHz processor; that gap is substantial. When it comes to overall feel, it’s more about how often the CPU really matters rather than constantly being the bottleneck. Those critical frames where CPU shines can make a noticeable difference.

Pages (2): 1 2 Next