Multi Tasking on Intel Haswell CPU’s

Test System

[CPU] Intel Core i7 4790K
[GPU] Gigabyte GTX 970 mini-ITX
[MOB] Gigabyte Z97-D3H (F7 bios)
[RAM] G. Skill Trident X 4GBx2 DDR3 @ 2400 MHz 10-12-12-31 2T
[HSF] Noctua NH-U12S

Windows 7 64-bit with Service Pack 1
GeForce 355.82

CPU’s Tested

Core i5 was simulated using Core i7. Hyper Threading was configured in the BIOS. Turbo was disabled for all CPU configurations.

Core i5 = Core i7 with Hyper Threading disabled

Core i7 4790K @ 4.4 GHz
4 cores / 8 threads

Core i5 4690K @ 4.4 GHz
4 cores / 4 threads

Test Methodology

Multi tasking workload will be running a game while HandBrake is encoding video in the background. The short movie (about 10 minutes long) Big Buck Bunny was encoded from AVI to MP4 using High Profile preset. Video resolution is 1080p. Big Buck Bunny is an open movie which was created using open software. You can download the movie here

Gaming performance will be measured by using FRAPS to record the frame rate and frame time. For those who are new to frame times, it is the measure of time taken to render a frame and its unit is milliseconds (ms). The lower the frame time, the better. The 99th percentile frame time is a measure of the overall “smoothness” of game play. For example, a 99th percentile frame time of 20ms means that 99% of the frames were rendered within 20ms. A more detailed explanation of frame time benchmarking can be found here

For those who want to replicate my gaming benchmarks, check the YouTube links below to see what part of the game I benchmarked. The settings used for each game are also indicated below. All the games are updated to the latest version as of September 14, 2015.

Grand Theft Auto V
Resolution: 1920 x 1080
MSAA: 2x
Every graphics option was set to the highest setting available except for Grass Quality which was set only to High

Crysis 3
Resolution: 1920 x 1080
Texture Resolution: Very High
Anti-aliasing: SMAA T2X
System Spec: Very High

Performance Results

The results below are the average of 3 runs

no MT – no multi tasking, just the game running
with MT – game + HandBrake



Performance Analysis and Conclusion

Here is the summary of performance difference between No Multi Tasking and With Multi Tasking. The percentage values below are relative to No Multi Tasking

Crysis 3
average frame rate: Core i7 @ -10.2% / Core i5 @ -13.8%
minimum frame rate: Core i7 @ -11.5% / Core i5 @ -17.6%
99th percentile frame time: Core i7 @ +47.4% / Core i5 @ +34.2%

Grand Theft Auto V
average frame rate: Core i7 @ -23.0% / Core i5 @ -12.9%
minimum frame rate: Core i7 @ -25.0% / Core i5 @ -15.8%
99th percentile frame time: Core i7 @ +51.7% / Core i5 @ +82.2%

Performance reduction is every noticeable even without a frame rate monitoring tool. With multi tasking, I noticed a reduction in game play smoothness with Core i7 but the games were still playable. However, doing multi tasking with Core i5 makes the game unplayable due to frequent and noticeable stuttering. I was also surprised by the very big performance reduction with 99th percentile frame times. I expected a reduction but not that big especially for Core i7.

4 thoughts on “Multi Tasking on Intel Haswell CPU’s”

  1. Did you try setting the video encoder to lowest priority? I'd expect game FPS to be relatively unchanged but the encoding rate to slow down more on i5. If you set it to high or maybe even default priority the game performance will drop a ton though.

  2. Just an FYI, the definition of “Core i5 = Core i7 with Hyper Threading disabled” is not the whole story. i7 has a larger cache on average, which may increase CPU performance on some titles and most synth benchmarks.

  3. I render videos with Sony Vegas while playing Arma 3 all the time, with my 4770K @ 4GHz. In Task Manager, right click on the processes for the game and HandBrake, then set the Affinity for them to use separate CPUs/Threads and try again. For example, i'd have arma3.exe set to CPU 0,1 and 2, while vegas13.exe would be set to the remaining 5.

Leave a Reply

Your email address will not be published. Required fields are marked *