Forum Topic
PC Bintsmok - Benchmarks, Reviews, and more (see page 1)
-
Frame Time Benchmarking Explained
Article continued from page 1 of this thread <click here for link>
Analyzing the Frame Time Data
Let me show you frame time graphs of The Witcher 3 using a GeForce GTX 970. These are actual frame time graphs generated by FRAPS Bench Viewer.
The frame time graphs above are for the 1st 300 frames of the 45-second benchmark. There are about 2,000+ frames in the 45-second benchmark. The graph would look cluttered if I show all the frame times for the whole 45-second benchmark.
You would notice that the Core i7 4790K has a much more consistent frame time than Core i3 4160K. Even though they have almost the same average frame rate, the gaming experience will be very different.
Now, you might be wondering how will you use the frame time data to compare different systems. It would be hard to look at frame time graphs when there are hundreds or thousands of frames.
This is where the 99th percentile frame time comes in. The 99th percentile frame time is a measure of the overall \"smoothness\" of game play. For example, a 99th percentile frame time of 20 ms means that 99% of the frames were displayed within 20 ms. If there are 2,500 frames, it means that 2,475 (99%) of the 2,500 frames were displayed within 20 ms. Lower 99th percentile frame time means smoother game play.
-- edited by bintsmok on Oct 31 2015, 01:33 AM -
Frame Time Benchmarking Explained
To be continued -
*** Note: This is not a paid review. No person or company sponsored the hardware used in this review
The Witcher 3 v1.12 - Performance Test
Test System
[CPU] Intel Core i7 4790K @ 4.0 GHz (Turbo disabled)
[GPU] Gigabyte GTX 970 ITX
[MOB] Gigabyte Z97-D3H (F9 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 359.06
Test Methodology
Gaming performance was measured by using FRAPS to record the frame rate and frame time. The results shown below are the average of 3 runs. For those who are not yet familiar with frame time benchmarking, please read this article <click here for link>
Everything setting was set to Ultra except for Shadow Quality and Foliage Visibility Range which will be both set only to High. Here are the complete details of the settings I used
Motion Blur - OFF
Blur - OFF
Anti-aliasing - ON
Bloom - ON
Sharpening - HIGH
Ambient Occlusion - SSAO
Depth of Field - ON
Chromatic Aberration - OFF
Vignetting - ON
Light Shafts - ON
Resolution - 1920 x 1080
Frame Rate - Unlimited
V-sync - OFF
NVIDIA HairWorks - OFF
Number of Background Characters - ULTRA
Shadow Quality - HIGH
Terrain Quality - ULTRA
Water Quality - ULTRA
Grass Density - ULTRA
Texture Quality - ULTRA
Foliage Visibility Range - HIGH
Detail Level - ULTRA
Performance Results
NOTES:
Core i7 running on 2 cores with HT enabled = Core i3
Core i7 running on 4 cores with HT disabled = Core i5
Watch this for proof <click here for link>
<click here for link>
<click here for link>
Conclusion
The results above show the importance of where you benchmark a game. In the Woesong Bridge area, all the CPU configurations delivered the same performance and same gaming experience. However, results of the Novigrad City benchmark reveals that the CPU has a large impact on performance and there is noticeable stuttering with the simulated Core i3. The stuttering was reduced greatly with the simulated Core i5 while the Core i7 delivered the smoothest gaming experience.
If you are using a mid-range GPU or high-end GPU, Core i5 is the minimum CPU you should use for the Witcher 3. While the game is playable with Core i5, you can get a smoother gaming experience with Core i7.
-- edited by bintsmok on Feb 09 2016, 05:34 PM -
Eto ang comprehensive review... thumbs up...
-
may ganito k p lng thread.... up haha
-
what if you have HBAO+ on?
-
*** Note: This is not a paid review. No person or company sponsored the hardware used in this review
Testing the Intel Core i3 6100 with NVIDIA GeForce GTX 970
Test System
[CPU] Intel Core i3 6100 3.7 GHz
[GPU] Gigabyte GTX 970 ITX
[MOB] Gigabyte Z170N Gaming (F5 BIOS)
[RAM] G. Skill Trident Z 8GBx2 DDR4 @ 3200 MHz 16-18-18-38 2T
[HSF] Intel stock cooler
Windows 10
GeForce 373.06
Test Methodology
Gaming performance was measured by using FRAPS v3.5.99 to record the frame rate and frame time. The results shown below are the average of 3 runs. For those who are not yet familiar with frame time benchmarking, please read this article <click here for link>
The Witcher 3 (v1.31) <click here for link>
Resolution: 1920 x 1080
HairWorks: Off
Ambient Occlusion: SSAO
Every graphics option was set to the highest setting available. Blur, Motion Blur, and Chromatic Aberration were disabled.
Grand Theft Auto V (build 877) <click here for link>
Resolution: 1920 x 1080
FXAA: On
MSAA: Off
Every graphics option was set to the highest setting available except for Grass Quality which is set only to Very High and Soft Shadows which is set only to Softest.
Performance Results
Conclusion
The games tested we very playable despite the GeForce GTX 970 being bottlenecked by Core i3 6100. In Grand Theft Auto V, the CPU usage is consistently at 98 - 100% while the GPU usage drops to 65 - 75%. Core i3 6100 is an acceptable gaming CPU but I would still recommend a Core i5 6500 for new builds even if the gaming PC to be built will only use a GeForce GTX 1050 Ti. Expect upcoming games in 2017 to demand more from the CPU and the Core i3 6100 is likely not to suffice.
-- edited by bintsmok on Nov 22 2016, 06:21 PM -
Replaying Tomb Raider in ultrawide :)
<click here for link>
-- edited by bintsmok on Jan 03 2018, 10:07 PM -
Serious Sam 3: BFE is an underrated game. I'm glad I bought this game. It supports ultrawide (21:9), Vulkan, and Linux.
<click here for link>
-
ang bangis! salamat bro bintsmok! ^_^
-
Short game play of Descenders on Ubuntu MATE 17.10.
<click here for link>
-- edited by bintsmok on Feb 11 2018, 12:48 AM -
Testing the Linux beta of Dead Cells on Ubuntu MATE 17.10.
<click here for link>
-- edited by bintsmok on Jun 23 2018, 03:35 PM -
Playing Batman: Arkham Origins on Ubuntu MATE 18.04 LTS through Steam Play's Proton.
V-Sync had to be enabled because OBS produces video with lots of stuttering when in-game frame rate is way above 60 FPS and the recording frame rate is 60 FPS. The game runs very smooth when there is no video recording.
My gaming PC:
Intel Core i7 7700 3.6 GHz
AMD Radeon RX 580 8GB
8GBx2 DDR4 @ 2,133 MHz
Ubuntu MATE 18.04 LTS
Mesa 18.1.7
<click here for link>
The Steam Linux client now has a new feature that allows Windows games without Linux version to be played within the Steam Linux client. Manual configuration of Wine is not needed anymore to play Windows games on Linux. <click here for link> -
Playing Hellblade: Senua's Sacrifice on Ubuntu MATE 18.04 LTS through Steam Play's Proton.
V-Sync had to be enabled for this game because OBS produces video with lots of stuttering when V-Sync is disabled. The game runs at 45 - 50 FPS when there is no video recording.
My gaming PC:
Intel Core i7 7700 3.6 GHz
AMD Radeon RX 580 8GB
8GBx2 DDR4 @ 2,133 MHz
Ubuntu MATE 18.04 LTS
Mesa 18.1.7
<click here for link> -
Playing The Witcher 3 on Ubuntu MATE 18.04.1 LTS through Steam Play 2.0.
V-Sync had to be enabled for this game because OBS produces video with lots of stuttering when V-Sync is disabled. The game runs at 60 - 70 FPS when there is no video recording.
My gaming PC:
Intel Core i7 7700 3.6 GHz
AMD Radeon RX 580 8GB
8GBx2 DDR4 @ 2,133 MHz
Ubuntu MATE 18.04.1 LTS
Mesa 18.3.0
<click here for link> -
Playing Bayonetta on Manjaro 18.1.5 KDE through Steam Play 2.0.
TL;DW - Constant 60 FPS at 1440p with RX 580. Everything displays correctly using Vulkan.
<click here for link>
-- edited by bintsmok on Feb 02 2020, 11:10 AM -
Playing Streets of Rage 4 on Manjaro 20.0.1 KDE through new version of Steam Play. Game is playable with Radeon RX 550, constant 60 FPS at 1080p.
<click here for link>