WebXPRT 4 performance scores

[SchumannFrequency]

I use the following hardware: Intel i3-3240 + 4GB RAM @1600Mhz DDR3 single channel + NVIDIA GTX 650 1GB + EVO 850 500GB
I use Firefox 111 and FreeBSD as software.

You can take the test on this page. https://www.principledtechnologies.com/benchmarkxprt/webxprt/

When you post, please include your specific hardware and software.

 

[SchumannFrequency]

Today I did the test with Chromium (with the hardware mentioned in my signature).

I also have a Thinkcentre-M75 with Ryzen 5 Pro 3400G with dual-channel ddr4 RAM @ 2666Mhz and it gets 130 on windows11 with Edge.

Intel i3-3240 + 4GB DDR3 @1600MHZ single channel + NVIDIA GTX 650 1GB + EVO 850 500GB + FreeBSD + Chromium = 120
Ryzen 5 Pro 3400G with dual-channel ddr4 RAM @ 2666Mhz + windows11 + Edge = 130

The Ryzen 5 Pro 3400G is almost exactly 3x faster in multithreading performance : https://hwbench.com/cpus/amd-ryzen-5-3400g-vs-intel-core-i3-3240
My i3-3240 has single channel RAM which gives no performance boost while the Ryzen 5 Pro 3400G in this test has dual channel memory and the RAM is also clocked 1000 Mhz higher.

Perhaps the performance difference between FreeBSD and windows11 largely closes this large hardware performance gap.

 

[SchumannFrequency]

This is an older version of the same benchmark.

According to notebookcheck, the i9 13900K reaches about 368 (in combination with RAM @ 3000 Mhz). The Ryzen 7950x would score around 398 according to cpuagent.

Today's best CPUs are about twice as fast as my eleven-year-old $110 dual core CPU with single channel ddr3 RAM.

You also have to look at everything in context.

 

[SchumannFrequency]

Firefox 111 on MX Linux (a Debian variant) with fluxbox.

This means FreeBSD will be on average about 6.5% faster than most Linux distros in WebXPRT3.
Many Linux distros score slightly lower than MX Linux.
Clear Linux is going to be almost as fast as FreeBSD in WebXPRT 3, but it's usually one or two points slower.

It's kind of strange that FreeBSD isn't used in supercomputers anymore. It is faster than Linux in many situations.
Most supercomputers are not being used optimally. Mainly due to a lack of skills common to most engineers.

 

[SchumannFrequency]

Ryzen 5 Pro 3400G with dual-channel ddr4 RAM @ 2666Mhz + windows11 + Chrome = 194 (WebXPRT 3)

It seems that if I gave Intel i3-3240 8GB DDR3 @1600MHZ dual channel I would score higher with FreeBSD and Firefox (than the Ryzen 5 Pro 3400G with dual-channel ddr4 RAM @ 2666Mhz + windows11 + Chrome).

I don't currently have a second ddr3 stick I can use, but I'm going to get it, as it would be worth it if my i3-3240 with FreeBSD then performed (in the browser) similar to the Ryzen 5 Pro 3400G with windows11.

 

[SchumannFrequency]

The use of dual channel DDR3 memory yields 4.2% higher performance in this benchmark.

You see here an i3-3240 getting a result of 125 in WebXPRT 4

I think this is only possible on a BSD system. Linux and windows are slower than BSD in this WebXPRT 4 test.

 

[Ferrum Master]

For the lulz

 

[SchumannFrequency]

AMD R5 PRO 3400G + Clear Linux & Opera browser

Not a bad result for WebXPRT4 for this budget CPU.
Firefox on Clear Linux is still 7.5 percent faster than Opera in the most realistic browser benchmark.

 

[SchumannFrequency]

AMD R5 PRO 3400G + Clear Linux & Firefox. This is without overclocking. I think the AMD R5 PRO 3400G allows for overclocking.
I'm not going to do that because it's in the compact housing of the ThinkCentre, which I don't think cools optimally.
If I overclock it, I would have to score above 170, which seems like a high result for the AMD R5 PRO 3400G.

 

[SchumannFrequency]

iPhone 13 + Safari 16.4 + WebKit 604.1

 

[SchumannFrequency]

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

Hardware: Intel i3-3240 + 8GB DDR3 @1600MHz dual-channel + NVIDIA GTX 650 1GB + EVO 850 500GB
Software: FreeBSD 13.1 + bspwm + Firefox 115.0.2 + Nvidia proprietary driver

Clear Linux has become almost exactly as fast as FreeBSD in this test.
Also interesting to note that with the AMD R5 3400G PRO + 2666MHz dual-channel RAM, I had a score of about 130 six months ago on windows11.

 

[Marcus Vinicus]

System as in signature below
Browser = Google Chrome Canary 117.0.5927.0
Also ran Edge latest stable build it scored 346

 

[SchumannFrequency]

Hardware: Intel i3-3240 + 8GB DDR3 @1600MHz dual-channel + NVIDIA GTX 650 1GB + EVO 850 500GB
Software: ALT Sisyphus + LXQt + Firefox 117 + proprietary Nvidia driver

As we can see, ALT Sisyphus is slower than FreeBSD and Clear Linux in this test. But the difference is not that big.
It is faster than 85% of Linux systems in the most useful browser benchmark.

 

[AVATARAT]

Ryzen 7700x
2x32GB @6200

Waterfox

Edge

 

[SchumannFrequency]

Software: ALT Sisyphus -- LXQt -- Chromium version 120.0.6051.2 -- proprietary Nvidia driver
Hardware: Intel 12600KF (stock) -- Kingston 6000 MHz CL40 -- GTX 650 1GB -- BIOSTAR B760MZ-E PRO -- Antec P6 -- Xilence XP550 -- ARCTIC i35 -- EVO 850 500GB

One year ago.

Software: ALT Sisyphus -- LXQt -- Firefox Nightly 120.0a1 -- proprietary Nvidia driver
Hardware: Intel 12600KF (stock) -- Kingston 6000 MHz CL40 -- GTX 650 1GB -- BIOSTAR B760MZ-E PRO -- Antec P6 -- Xilence XP550 -- ARCTIC i35 -- EVO 850 500GB

 

[SchumannFrequency]

Hadware: DELL Latitude E6540 and Intel® Core™ i7-4610M CPU @ 3.70GHz
Software: ALT Linux and Firefox beta 127.0b7-1

Ryzen 5 Pro 3400G with dual-channel ddr4 RAM @ 2666Mhz + windows11 + Firefox stable = 135 (WebXPRT 4 and checked +- 9 months ago)
The much older Intel 4610M notebook CPU achieves a surprisingly good result here considering the form factor and age.

Thinkcentre-M75 with Ryzen 5 Pro 3400G with dual-channel ddr4 RAM @ 2666Mhz got 130 on windows11 with Edge +- 14 months ago.

 

[lolek86]

chrome 125 , windows 11 24h2, ryzen 7600@5.5ghz , 6200mt tuned memory

 

[SchumannFrequency]

Hardware: Intel 12600KF (stock) -- Kingston 6200 MHz CL36 -- Sapphire RX 7600 -- BIOSTAR B760MZ-E PRO -- Antec P6 -- Xilence XP550 -- ARCTIC i35 -- 980 PRO 500GB
Software: OpenBSD, bspwm, open-source GPU driver, UFS file system, Chromium 125.0.6422.76

WebXPRT 4 does not work well on Chromium + OpenBSD so I tested the previous version. (WebXPRT 3)
OpenBSD focuses more on security and performance comes second.
It's still a good result when you compare it to how this CPU performed with browser versions from a few years ago (in WebXPRT 3).

Gentoo with Chromium only scores 4% higher than OpenBSD in WebXPRT 3.
With some optimizations, OpenBSD could score similar to or higher than the average Linux system in WebXPRT 3 (Chromium).

 

[SchumannFrequency]

The (WebXPRT 3) performance difference between Epiphany and Chromium is less than 1% on OpenBSD

Epiphany + OpenBSD + Intel 12600KF (WebXPRT 3)

Chromium + OpenBSD + Intel 12600KF (WebXPRT 3)

 

[SchumannFrequency]

Chromium + OpenBSD (bspwm) + Intel 12600KF (WebXPRT 3) = 405
Chromium + Gentoo (KDE Plasma) + Intel 12600KF (WebXPRT 3) = 428

The exact difference between OpenBSD and Linux seems to be around 5.7% for this (Chromium) test.
Chrome, Vivaldi, Nyxt and Brave are faster than Chromium in this test on Gentoo, but the difference is never big.

Only Firefox manages to create a big gap in performance with Chromium in this test on Gentoo, which you see here.

 

[SchumannFrequency]

Hardware: Intel 12700KF (stock) -- G.SKILL RIPJAWS @3600 CL18 (stock) -- Sapphire RX 7600 -- ASRock B760M-ITX/D4 WiFi -- fractal design DEFINE NANO S -- bequiet! SYSTEM POWER 10 550W -- DeepCool AG500BK ARGB -- EVO 850 500GB
Software: mageia, LXQt, Mesa open-source driver, XFS file system, Firefox 130.0a1 (AppImage)

WebXPRT 3 results from 17/10/2023

 

[Paramount]

 

[SchumannFrequency]

View attachment 357140

How much do you get with this CPU in WebXPRT 3?

 

[Paramount]

How much do you get with this CPU in WebXPRT 3?

FireFox 130

 

[Athlonite]

System 7800X3D, 32GB DDR5 6000MT/s

This seems to be a pretty poor CPU testing benchmark it fails to push all threads to anywhere near their maximum and very poorly multi core aware wit the max core's/threads being used is ~4 but on average it's mostly single core/thread leading to poor utilization of ~14% at most