• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.
  • The forums have been upgraded with support for dark mode. By default it will follow the setting on your system/browser. You may override it by scrolling to the end of the page and clicking the gears icon.

Post your JetStream 2 speeds!

Attachments

  • 1662562796084.png
    1662562796084.png
    208.3 KB · Views: 117
Last edited:
??
I'm getting 1.1 score in this benchmark with 24/7 settings
View attachment 260945

This is because you are using a different maze. My screenshot and the Phoronix benchmarks use the first maze. I think you are using the third maze.

CSS performance is something that is still an important element to determine browser performance. Because there are many websites where the CSS part is heavier than the JS part. Why this is so is explained here.

Myth Busting: CSS Animations vs. JavaScript

Once upon a time, most developers used jQuery to animate things in the browser. Fade this, expand that; simple stuff. As interactive projects got more aggressive and mobile devices burst onto the scene, performance became increasingly important. Flash faded away and talented animators pressed HTML5 to do things it had never done before. They needed better tools for complex sequencing and top-notch performance. jQuery simply wasn’t designed for that. Browsers matured and started offering solutions.

The most widely-acclaimed solution was CSS Animations (and Transitions). The darling of the industry for years now, CSS Animations have been talked about endlessly at conferences where phrases like “hardware accelerated” and “mobile-friendly” tickle the ears of the audience. JavaScript-based animation was treated as if it was antiquated and “dirty”.
 
This is because you are using a different maze. My screenshot and the Phoronix benchmarks use the first maze. I think you are using the third maze.
Also getting 3.5 seconds with that other maze after a restart..
Dont think this is a good benchmark for modern hardware, but atleast that 5950x minimum 3.7sec is not correct :)

*edit*
And it seems like 40*40 grid is even faster? (~3.0 sec with lots of background running)
lol it seems to be random what maze your getting there also..
 
Last edited:
Also getting 3.5 seconds with that other maze after a restart..
Dont think this is a good benchmark for modern hardware, but atleast that 5950x minimum 3.7sec is not correct :)
The benchmarks were on Linux and you are using Windows. It was an older browser he was using in the test, so it may be true that the 5950x was slower in January 2022.

lol it seems to be random what maze your getting there also..
The maze you get when you refresh the page is always exactly the same maze, it is used in the Phoronix tests.

As for how accurate this benchmark is to describe real-world performance, I'd say it's quite usefull.

Many systems with an SSD will perform about the same on certain parts of web browsing. Take my old dual core i3 3240 for example. When I look up things on Google, the search results come up instantly, I don't think my CPU is a bottleneck there in any way. Say I'm browsing The Guardian's website, that's just instant too, I click on something and it loads the entire webpage in about half a second.

It is correct to say that a CPU is not a significant bottleneck in many everyday tasks. Software is often a bigger bottleneck than hardware. Take, for example, the boot time of Windows 10/11. Windows doesn't actually boot up completely, it does a form of hibernation because windows is too slow without this trick. This hibernation makes windows more prone to driver bugs. However, it will still boot slower than if you let Alpine Linux or Void Linux boot completely on the same hardware.

When I log into my FreeBSD system, it takes exactly one second in XFCE. While you can't make Windows 10/11 log in faster even with the most powerful CPUs. Suppose I am viewing photos with Viewnior or with Feh, the transition to the next photo is going to be faster than on the most powerful CPUs with Windows 10/11.

Let's say I open a file manager with XFCE, this takes about one second. It's as fast as the most powerful CPUs with Windows 10/11.

Suppose I open my email app (Claws Mail), this will load my emails faster than is possible in Windows 10/11.

Suppose I open Gnumeric, I am going to achieve higher performance in most tasks than what is possible in windows 10/11, and Gnumeric is a more advanced spreadsheet app than MS Excel.

The benchmark is appropriate to see how fast the CPU is if there is no major bottleneck. This is the situation for most of the day-to-day tasks that most people do.

Here's a benchmark that measures the bottlenecks in a specific situation: https://perftest.netlify.app/stylebench/

But the question is, in which situations do you encounter this bottleneck in reality. I can say that more than 90% of the websites on my systems are fully loaded within a maximum of 2 seconds. One of the slower websites is Reddit which takes 4 seconds before it is completely loaded. I think this is because they switched from Lisp to Python which made Reddit very slow.
If they had made some changes to their Lisp code instead of switching to Python, they would have had a much faster platform.
 
Last edited:
Actually, it is. It is a good way to demonstrate the lack of difference between most hardware for basic tasks.
EXACTLY, which is why i KNOW he used this method which i just made a video of demonstrating 1.1seconds....there is zero and i MEAN ZERO chance he did 2.4 s faster than my pc, which has been edging his on every actual bench in the past 2 weeks we've both posted on.. and discussed in one case..and those benches are actually designed to show small differences for us to tweak...This benchemark shows the very lack of differences between well tuned PCS.. Like The count said... Again... this is how he cheated a 1.1s https://rumble.com/v1j1yfv-yep.html , not to mention in all the benchmarks we've both hit these past 2 weeks, I've been edging his system out by 5-10%.....and now I get 3.5s and he gets 1.1? Now all the sudden you're blowing me out of the water by... whats the percentage difference there? 104%? xD cmon man.
 
Last edited by a moderator:
This is not an official contest of any sort. Please play nice or you will be thread banned. No bickering or arguing.
 
M1 Air on Safari with 8GB RAM and 8-core GPU
 

Attachments

  • Screen Shot 2022-09-07 at 1.30.16 PM.png
    Screen Shot 2022-09-07 at 1.30.16 PM.png
    1.7 MB · Views: 133
Windows11 doesn't work that well I must say. I think the hardware of this system is something like 3x more powerful than my hardware. If Google published a well-optimized build of Chrome for FreeBSD then my i3-3240 with slow single channel DDR3 RAM would be exactly as fast in JetStream 2 as this R5 Pro Mobile 3400G with fast DDR4 dual channel RAM. Is that still normal with this difference in hardware? https://www.cpu-monkey.com/en/compare_cpu-amd_ryzen_5_pro_3400g-1262-vs-intel_core_i3_3240-391 When I have time I will test the Void Linux, FreeBSD and AlmaLinux OS scores on this specific R5 Pro 3400G so that we know exactly what the performance differences are.
Schermafdruk 2022-08-09 12.44.52.png

Schermafdruk 2022-08-09 12.07.58.png
 
I tested the same R5 Pro 3400G today with Linux and these were the results.
When JetStream 2 runs on Lubuntu or Fedora for the first time, I got a result that fluctuated between 102 and 106:
Screenshot from 2022-09-12 13-16-23.png

On FreeBSD and windows11 the results stayed the same on the following attempts, but on Linux the second attempt immediately gives a higher result:
screen5.jpg

If you then delete the browser history completely, the result on Linux will go back between 102 and 106.
With Brave on FreeBSD I got a result above 99 on an i3 3240 very old dual core with slow single channel DDR3 memory so it seems FreeBSD with Brave is a lot faster than Ubuntu or Fedora for browsing (JavaScript and CSS)
I have to say that I'm a bit disappointed with the work of Canonical and Red Hat as they lose out to a system that is much less used for browsing.
If you want a Linux system that will browse as fast or faster as windows10/11 on the first try, you should probably use Clear Linux with Chrome, or a Linux distro almost as fast as Clear Linux, or FreeBSD with Brave.

Today JetStream 2.1 was released and this is my result:
Screenshot from 2022-09-12 12-53-24.png
 
Considering the cpu age I guess it did good.

E5 1660V3 @ 4.5GHz (HT Disabled) + 32GB RAM @3200MHZ quad channel + RTX 3070 8GB + Patriot Burst 240GB
Windows 10 Pro 21H2 19044.1899
Chrome 105.0.5195.102


js2.PNG
 
Considering the cpu age I guess it did good.

E5 1660V3 @ 4.5GHz (HT Disabled) + 32GB RAM @3200MHZ quad channel + RTX 3070 8GB + Patriot Burst 240GB
Windows 10 Pro 21H2 19044.1899
Chrome 105.0.5195.102
E5 1660V3 @ 4.5GHz That's why it still performs well. Browsers are well optimized for multi-threading and it has 8 cores.

The performance in the Brave browser on the first try.
Screenshot at 2022-09-12 20-31-19.png

The performance in the Brave browser on the second try.
Screenshot at 2022-09-12 20-35-21.png

I knew from my previous results that certain Linux systems will get faster browser results than windows11. Solus has a few of Clear Linux's optimizations, which is why it's already on par with Edge on w11 and faster than Chrome on w11. On the second try it is faster than probably all browsers on windows.
Interesting to know is that in my experiments Void Linux is slightly faster than Solus in Brave for Speedometer 2.0 so I suspect Void will also be faster in JetStream 2.0
I will also test AlmaLinux with Brave in the future, because it uses the p-state, which makes it close to Clear Linux in most wokloads. So it could theoretically be faster than Solus and Void.

I think the above performance of Solus gives a good idea of how the faster Linux systems perform. They are slightly faster than windows11. Clear Linux with Chrome/Brave is still 4% faster than Solus for browsing but there are no faster Linux systems than Clear Linux.

Today I tested GhostBSD's performance and these were my results. The performance in the Brave browser on the first try.
Screenshot_2022-09-13_11-23-46.png

The performance in the Brave browser on the second try.
Screenshot_2022-09-13_11-29-17.png

We can see that GhostBSD with Brave is faster than any windows or Linux system in JetStream 2.

I ran three more benchmarks to see if this was a coincidence, or if FreeBSD is generally the fastest system for browsing.
In Octane 2.0 I got around 47 000, unsurprisingly it performed well here as this is similar to JetStream 2. These were the results of the last two benchmarks:
Screenshot_2022-09-13_11-49-11.png

To put this Kraken 1.1 result in context:

Screenshot_2022-09-13_12-41-37.png

To put this ARES-6 result in context:
Intel Core i9 9900K with ASUS PRIME Z390-A motherboard, 16GB RAM, Samsung 970 EVO 256GB NVMe SSD, and AMD Radeon RX Vega 56 graphics in the year 2019.
 

Attachments

  • Screenshot at 2022-09-12 20-31-19.png
    Screenshot at 2022-09-12 20-31-19.png
    266.2 KB · Views: 106
  • Screenshot at 2022-09-12 20-35-21.png
    Screenshot at 2022-09-12 20-35-21.png
    266.9 KB · Views: 95
12700K MS Edge and Brave below that.

1666753056880.png

1666752880270.png

Brave browser :
1666752922585.png

1666752937042.png


1666753006124.png
 
Hmm some more under the hood improvements from the Vivaldi team
Vivaldi5.5.2805.38 (Stable channel) (64-bit)

2022-10-26 17.24.33 browserbench.org 2b2e895ba199.jpg


Previous score was 245
 
62.223
Firefox browser
Laptop Spec i5 6200u, 8g ram, intel hd graphics and kingston ssd
 
62.223
Firefox browser
Laptop Spec i5 6200u, 8g ram, intel hd graphics and kingston ssd
Firefox is slow in this specific benchmark. You will score around 98 with a browser based on Chromium.

However, in WebXPRT, Firefox is faster than all Chromium-based browsers.
 
Score: 151.580
Specs: R7 5800X, 32GB DDR4-3600 (dual-channel), RX 6800 XT, 1TB WD SN750
Software: Windows 10 LTSC 2021, Firefox 106.0.3

One tab was open, and all of my browser extensions were enabled.

IVgMGDm.png
 
Screenshot from 2022-12-18 12-34-32.png

Clear Linux and FreeBSD are two of the fastest systems for JavaScript, HTML5, CSS and WASM.
In most benchmarks they score about exactly the same.
FreeBSD (now) scores above 98 in this benchmark in Chromium, and I'm using Chrome for the above benchmark which is slightly faster than Chromium.
They have about the same CPU performance, but FreeBSD has more IOPS than Clear Linux, making apps launch faster.

If you find a system that is generally faster than FreeBSD, I'd love to hear about it.
 
Vivaldi seem to be on a run lately couple of more updates (6~7 actually) and more perf again

Vivaldi5.6.2867.46 (Stable channel) (64-bit)

Vivaldi jetst2 2022-12-20 193552.png
 
Chrome Version 108.0.5359.125
Win10 22H2 19045.2364
Specs over there <--
1671527191804.png
 
Back
Top