- Oct 26, 2015
- 23
- 0
- 4,520
Hello all, I did a small amount of testing in Call of Duty: Black Ops 3 to compare the effects of having an i5 or an i7 at the same clock speeds.
For this test i used an i7-4790K, for all tests, all cores were enabled and set to 4.4 GHz.
Instead of switching between CPUs, as i dont own an i5 haswell, i simply disabled Hyperthreading support to emulate an i5, i know some will complain about the fact that its not exactly the same as an i5-4690K as it will have 8MB of cache but the effects of 2MB L3 cache will not really effect much here.
Test Machine:
CPU: i7-4790K @ 4.4 GHz
RAM: 16GB DDR3 1600 MHz CL8
Motherboard: MSI Z97M Gaming
GPU: GTX Titan X
Drive: Samsung 850 EVO M.2 (SATA-3)
PSU: 500w Corsair CX500M
Test Constants:
CPU will always have all cores enabled and will run at 4.4GHz at all times.
GPU will always be clocked at 1000MHz Core, and 8GHz VRAM.
RAM will always be 16GB DDR3 CL8 @ 1600 MHz
Game settings will always be set to max possible at 1080p (vsync off, frame cap 240, everything else at maximum available setting)
Test Variables:
CPU Hyperthreading will be Enabled (to emulate i7-4790K)
CPU Hyperthreading will be Disabled (to emulate i5-4690K)
Testing Method:
For both scenarios. I will run through Mission 5 (Hypocenter), then Mission 11 (Life), and record average FPS and average CPU Utilisation throughout each.
Testing Results:
Hyperthreading Enabled (i7-4790K):
-Mission 5 - FPS: Avg of 158 FPS, CPU: 76% Usage
-Mission 11 - FPS: Avg of 127 FPS, CPU: 84% Usage
Hyperthreading Disabled (i5-4690K):
-Mission 5 - FPS: Avg of 122 FPS, CPU: 100% Usage
-Mission 11 - FPS: Avg of 93 FPS, CPU: 100% Usage
Conclusion:
The effect of having an i5 vs an i7 is fairly clear here. As soon as the mission started the i5's usage choked up to 100%, fps was a little lower, and in certain points minor fps drops were noticable. With the i7 the cpu was able to power through, at no point hitting 100% usage and no frame drops were noticed. So guys i think we've found a scenario where having an i7 vs an i5 at same clocks actually does give some benefit!
Other interesting notes:
With hyperthreading disabled, although usage was higher, power consumption of the cpu and gpu were lower, and consequently, temps were lower too, by around 5c. I cant comment too much on temperatures as both GPU and CPU are in the same watercooling loop.
For this test i used an i7-4790K, for all tests, all cores were enabled and set to 4.4 GHz.
Instead of switching between CPUs, as i dont own an i5 haswell, i simply disabled Hyperthreading support to emulate an i5, i know some will complain about the fact that its not exactly the same as an i5-4690K as it will have 8MB of cache but the effects of 2MB L3 cache will not really effect much here.
Test Machine:
CPU: i7-4790K @ 4.4 GHz
RAM: 16GB DDR3 1600 MHz CL8
Motherboard: MSI Z97M Gaming
GPU: GTX Titan X
Drive: Samsung 850 EVO M.2 (SATA-3)
PSU: 500w Corsair CX500M
Test Constants:
CPU will always have all cores enabled and will run at 4.4GHz at all times.
GPU will always be clocked at 1000MHz Core, and 8GHz VRAM.
RAM will always be 16GB DDR3 CL8 @ 1600 MHz
Game settings will always be set to max possible at 1080p (vsync off, frame cap 240, everything else at maximum available setting)
Test Variables:
CPU Hyperthreading will be Enabled (to emulate i7-4790K)
CPU Hyperthreading will be Disabled (to emulate i5-4690K)
Testing Method:
For both scenarios. I will run through Mission 5 (Hypocenter), then Mission 11 (Life), and record average FPS and average CPU Utilisation throughout each.
Testing Results:
Hyperthreading Enabled (i7-4790K):
-Mission 5 - FPS: Avg of 158 FPS, CPU: 76% Usage
-Mission 11 - FPS: Avg of 127 FPS, CPU: 84% Usage
Hyperthreading Disabled (i5-4690K):
-Mission 5 - FPS: Avg of 122 FPS, CPU: 100% Usage
-Mission 11 - FPS: Avg of 93 FPS, CPU: 100% Usage
Conclusion:
The effect of having an i5 vs an i7 is fairly clear here. As soon as the mission started the i5's usage choked up to 100%, fps was a little lower, and in certain points minor fps drops were noticable. With the i7 the cpu was able to power through, at no point hitting 100% usage and no frame drops were noticed. So guys i think we've found a scenario where having an i7 vs an i5 at same clocks actually does give some benefit!
Other interesting notes:
With hyperthreading disabled, although usage was higher, power consumption of the cpu and gpu were lower, and consequently, temps were lower too, by around 5c. I cant comment too much on temperatures as both GPU and CPU are in the same watercooling loop.
Facebook
Twitter
Instagram