• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

Share your Speedometer 2.0 benchmark here

Capture d’écran du 2023-06-08 13-21-40.png

Intel i3-3240 + NixOS 23.11 + Cinnamon + Vivaldi 6.0.2979.22
 
See signature for system info
Brave 152.126

speedometer2-29june2023.jpg
 
Capture d’écran du 2023-06-20 02-35-24.png

Hardware: AMD R5 PRO 3400G (134$ CPU+GPU 2019) + dual-channel DDR4 @2666MHz
Software: Vivaldi + NixOS + Cinnamon
 
iPhone 8 on safari that I bought 2nd hand for $100 CAD

IMG_2748.png
 
With my daily driver system running at stock & its iGPU + Brave browser inside win 10-22H2.
BrowserBenchmark_iGPU - Copy.JPG
 
Daily driver, Google Chrome.

Version 114.0.5735.134 (Official Build) (64-bit)

Screenshot 2023-06-19 214933.jpg
 
See signature for system info
Google Chrome Canary 116.0.5842.0

The benchmark seems to score well with Google Chrome?

speedometer2-Google Chrome Canary 20june2023.jpg
 
@SchumannFrequency
while i get the comparing/sharing of results, stating that even the latest/fastest (bestes) cpu is only small amounts faster than some 5/10/20y old one (or phone for taht matter), is kind of lost on me.

ppl usually dont have stuff like a 7950X (incl the supporting hw like board/ram/gpu/psu) for web browsing, and i doubt those great low-cost cpus from over 10y ago will properly feed a 4080 for gaming.

the same way a 200HP car from today might not be much faster than one from +20y ago, but almost everything else on/in it, will be ("better")..
 
My computer at work: HP prodesk I3-10100T, 8GB RAM running at 3200MHz

1687244984616.png
 
Screenshot from 2023-07-22 18-56-06.png

Hardware: Intel i3-3240 + 8GB DDR3 @1600MHz dual-channel + NVIDIA GTX 650 1GB + EVO 850 500GB
Software: Clear Linux + Brave browser + nouveau open-source GPU driver

I ran the test just one time and I can probably score higher.
As you can see I can almost reach the red area with the old i3 CPU which is awesome.

Screenshot from 2023-07-23 09-40-23.png

Hardware: Intel i3-3240 + 8GB DDR3 @1600MHz dual-channel + NVIDIA GTX 650 1GB + EVO 850 500GB
Software: Clear Linux + i3-wm + Brave browser + nouveau open-source GPU driver

Screenshot from 2023-07-23 12-04-02.png

Hardware: Intel i3-3240 + 8GB DDR3 @1600MHz dual-channel + NVIDIA GTX 650 1GB + EVO 850 500GB
Software: Clear Linux + i3-wm + Firefox 115.0.2 + nouveau open-source GPU driver

Firefox had a reasonable lag on browsers like Brave in this particular test. But that seems to have been largely eliminated in recent months.
If it continues to evolve like this, Firefox might be faster than Chromium-based browsers in a few months (on Clear Linux in the Speedometer 2.0 test).
 
Screenshot from 2023-07-27 09-21-24.png

Hardware: Intel i3-3240 + 8GB DDR3 @1600MHz dual-channel + NVIDIA GTX 650 1GB + EVO 850 500GB
Software: Clear Linux + i3-wm + Brave browser + nouveau open-source GPU driver

I think this is about exactly the same performance as the Samsung Galaxy S22 Ultra 5G plus + Samsung Exynos 2200, Xclipse 920, 8192

You can say that the performance of the i3-3240 is still from this moment, since he scores the same as a very recent version of the most popular Android flagship.
 
After clean install of Windows a couple of weeks back.
Browser Google Canary build 117.0.5913.0

See signature for system specs.

speedometer2-27july2023.jpg
 
Screenshot from 2023-07-28 15-17-23.png

Hardware: Intel i3-3240 + 8GB DDR3 @1600MHz dual-channel + NVIDIA GTX 650 1GB + EVO 850 500GB
Software: Clear Linux + i3 wm + Brave browser + nouveau open-source GPU driver

agent_x007 also posted a result of 100 with the default “Balanced” power plan and 6950X 4,26GHz + RX 570 8GB.
6950X = +45% single core speed.

That's kind of strange that windows in balanced mode performs so slowly in the default mode.
I can tweak Clear Linux to get more performance, the result is just the standard mode and the standard Brave version: 1.56.11 Chromium: 115.0.5790.102 (Official Build)

It would be interesting to compare how much power windows 10/11 draw in high performance mode versus Clear Linux + i3 wm.
Generally speaking, Linux uses less power at idle than Windows.

Clear Linux kernel is more power efficient while playing YouTube videos?

It also seems to me that i3 wm is going to use less power than the heavyweight desktop of GNOME, Plasma, windows and macOS.
But without testing we don't know for sure.

Power consumption and performance are not the only two advantages i3 has to offer:
 
Last edited:
agent_x007 also posted a result of 100 with the default “Balanced” power plan and 6950X 4,26GHz + RX 570 8GB.
6950X = +45% single core speed.

That's kind of strange that windows in balanced mode performs so slowly in the default mode.
In my case, "default" power plan simply doesn't allow CPU to boost over 3GHz clock with this benchmark (too low CPU usage).
Performance plan forces full clock on any workload (I modified it to force it even on Idle state), which should explain the massive difference in score you can see with my results.
 
5800X3D / Brave with 5 extensions / W10
Screenshot 2023-07-29 212806.png
 
1.png

Hardware: Intel i3-3240 + 8GB DDR3 @1600MHz dual-channel + NVIDIA GTX 650 1GB + EVO 850 500GB
Software: ROSA Fresh 12.4 + LXQt + Chrome version 117.0.5911.2 + proprietary Nvidia driver

ROSA is a bit of an odd distro in terms of performance. In terms of CPU performance, it is one of the slowest Linux systems similar to AlmaLinux.
But then in terms of speed to open apps, it is much faster than NixOS and also faster than Clear Linux.
It seems to have a lot of IOPS. It also has a reasonably fast package manager.

In my case, "default" power plan simply doesn't allow CPU to boost over 3GHz clock with this benchmark (too low CPU usage).
Performance plan forces full clock on any workload (I modified it to force it even on Idle state), which should explain the massive difference in score you can see with my results.
It is single threaded. It's going to look to you like it's using very little CPU, but meanwhile there is one thread running at 100% all the time. It should detect this and that core should be allowed to boost to higher speeds. Windows' current balanced mode is fairly primitive in its behaviour.

Screenshot from 2023-07-30 15-07-33.png

Hardware: Intel i3-3240 + 8GB DDR3 @1600MHz dual-channel + NVIDIA GTX 650 1GB + EVO 850 500GB
Software: Clear Linux + i3 wm + Chrome version 117.0.5911.2 + nouveau open-source GPU driver

Clear Linux is faster than ROSA Linux if we are going to use the same software versions, but not a big gap.
ROSA Linux has the advantage that you can install it without Flatpak (directly via RPM), which makes the app start much faster.

Screenshot from 2023-07-30 16-19-41.png
 
11.png

Hardware: Intel i3-3240 + 8GB DDR3 @1600MHz dual-channel + NVIDIA GTX 650 1GB + EVO 850 500GB
Software: ROSA Fresh 12.4 + LXQt + Chrome version 117.0.5911.2 + proprietary Nvidia driver

Conclusion: ROSA Fresh is almost exactly as fast as Clear Linux in Chrome 117. In Firefox 115, however, I see a much bigger gap between the two systems.

Screenshot from 2023-07-31 11-28-48.png

Hardware: Intel i3-3240 + 8GB DDR3 @1600MHz dual-channel + NVIDIA GTX 650 1GB + EVO 850 500GB
Software: Clear Linux + i3 wm + Chrome version 117.0.5911.2 + nouveau open-source GPU driver
 
I signed up because at first I thought the iPhone was winning this and I couldn't stand for that. LOL. But then I just noticed someone else did a 400+ run as well.

13900k DDR5, 5.9/4.7/5.2, DDR5 8200CL36

1690836229041.png
 
Screenshot from 2023-08-02 14-00-08.png

Hardware: AMD R5 PRO 3400G + dual-channel DDR4 @2666MHz + integrated graphics
Software: Clear Linux + i3 wm + Chrome version 117.0.5911.2 + Mesa open-source GPU driver
 
fgdgf2.png

Hardware: Intel i3-3240 + 8GB DDR3 @1600MHz dual-channel + NVIDIA GTX 650 1GB + EVO 850 500GB
Software: ALT Sisyphus + LXQt + Chrome unstable + open-source Nvidia driver
 
an I5 8400h beats a 13900K :confused:
 
Back
Top