Post your JetStream 2 speeds!

[SchumannFrequency]

So you presume this topic will exist for the whole day? A year from now how you will understand what version of a particular browser was used at a random time earlier?

OS is a non-factor, period. 99% of people here run bog-standard Windows 10 or 11. Among die-hard Linux fans maybe 0.01% run Clear Linux as their daily driver. Absolute most Linux users run pre-compiled distros including Ubuntu and its derivatives, Arch and Fedora. All the others barely register.

We can occasionally mention the version so that people have an idea. Here you will find mainly Windows users. There are currently possibly even more windows7 users than windows11 users according to statistics.

People of any nationality can eventually participate in Techpowerup. You should also keep in mind that less than 50 percent of Russians currently use Windows. In China it is less than 80 percent. In South and North Korea you also have that people want to use Linux for government desktops. In India you also have people who use Linux. In the Linux world you have more fragmentation than before, Ubuntu is less dominant. You also have quite a few people using macOS, iOS, Android or ChromeOS.

Regarding Clear Linux, quite a few people have already asked me to post a detailed tutorial online on how to use it for gaming. Many people tell me that Clear Linux can become very popular with a little help.

 

[P4-630]

People of any nationality can eventually participate in Techpowerup.

Most of them here are using Bill Gate$' software as far as I know...

 

[AlwaysHope]

What does your i7-11700K do?

This, but with Edge.

 

[R-T-B]

Most of them here are using Bill Gate$' software as far as I know...

Considering the last thing I think he worked on was XP, I hope not.

 

[cvaldes]

Here are my results on my daily driver (not the build in my System Specs), Windows 10 21H2

Ryzen 7 3700X (PBO enabled), 32GB RAM 3200MHz, MSI MPG B550-I Gaming Wifi, EVGA RTX 3050 OC 8GB, Sabrent Rocket 4 Plus 1TB m.2

These results are with ungoogled Chromium (Marmaduke build):

View attachment 256589

Unsurprisingly Firefox is a total dog:
View attachment 256590

And now a different machine:

Mac mini 2018 (Intel Core i7-8700B, 6 core/12 thread), 16GB RAM, 1TB SSD running macOS Monterey 12.4

Once again ungoogled Chromium (Marmaduke build)

and another result with Safari 15.5

which compares pretty favorably next to Chromium-based browsers.

 

[Athlonite]

Vivaldi: 5.3.2679.70 (Stable channel) (64-bit)
Edition Windows 11 Pro
Version 22H2
OS build 22621.290
Experience Windows Feature Experience Pack 1000.22632.1000.0

 

[P4-630]

Considering the last thing I think he worked on was XP, I hope not.

LOL, I mean in general "Windows"

 

[Tatty_Two]

@SchumannFrequency if you don't mind make this thread's posts editable. I tried to add my browser version to my original post to no avail.
Thanks.

You can now.

 

[oobymach]

This, but with Edge.
View attachment 256693

Here's my edge result, better than all my other browsers and comparable to your result.

 

[lmille16]

Just for kicks, I ran it on my Verizon iPhone 13 iOS 15.6 using Chrome and scored a 201.069. Safari scored a 201.437.

 

[cvaldes]

Just for kicks, I ran it on my Verizon iPhone 13 iOS 15.6 using Chrome and scored a 201.069. Safari scored a 201.437.

And here's my iPhone 12 mini (Apple A14 SoC, 2 performance cores + 4 efficiency cores, 2GB RAM) running iOS 15.6:

I expect other iPhone browsers will perform similarly since they are all required to use the WebKit engine on iOS.

It's even faster on my iPad mini 6th generation (Apple A15 SoC, 2 performance cores, 4 efficiency cores, 4 GB RAM):

From a performance-per-watt standpoint, my iOS devices destroy my PCs (Windows and Mac) in this benchmark.

 

[Athlonite]

I just ran this again and watched Task managers CPU usage it seems it's not very CPU intensive and when it is it only uses 4~5 of the available 16 threads and only then it's lucky to be pushing 10.6% usage until the very end when it calculates the score where it hikes upto 76% for half a second. To me it just doesn't seem very multicore aware

@oobymach what sort of OC do you have on your R5 5600

Microsoft Edge Version: 103.0.1264.77 (Official build) (64-bit)

Hmmm seems Edge is quite a bit faster at this than Vivaldi I wonder why as they're both using the exact same version of chromium under the hood

 

[SchumannFrequency]

I just ran this again and watched Task managers CPU usage it seems it's not very CPU intensive and when it is it only uses 4~5 of the available 16 threads and only then it's lucky to be pushing 10.6% usage until the very end when it calculates the score where it hikes upto 76% for half a second. To me it just doesn't seem very multicore aware

It runs several test cases from real-world applications and libraries. In reality, your CPU will never go to 100% while browsing the internet, even if you have multiple websites open. However, this does not mean that it is not a good benchmark to measure your CPU performance while browsing the web. The most powerful AMD processors are only going to have 1% CPU usage when running Cyberpunk 2077. But that doesn't mean Cyberpunk 2077 is necessarily going to get the most FPS (the highest performance results) on a 64 core CPU.

Hmmm seems Edge is quite a bit faster at this than Vivaldi I wonder why as they're both using the exact same version of chromium under the hood

Chrome and Edge use proprietary technology on top of this engine, so Chormium and Chrome, for example, have different performance in certain things.
A typical example as pointed out by a Google developer on Reddit is the rendering of PDFs, where Google uses a much faster proprietary product other than an open-sourced one.
Due to this, some streaming sites are unavailable for Chromium but work with Google Chrome.

 

[bonehead123]

Here is mine, using the latest Chrome and the hardware in my specs, which I think is ok considering it's a few gens old:

141.210
Score

3d-cube-SP​

294.744​

156.250First344.828Worst475.240Average

3d-raytrace-SP​

282.179​

138.889First333.333Worst485.318Average

acorn-wtb​

40.371​

24.272First47.170Worst57.471Average

ai-astar​

384.477​

263.158First344.828Worst626.316Average

Air​

322.319​

131.579First285.714Worst890.719Average

async-fs​

102.945​

104.167First82.873Worst126.377Average

Babylon​

428.004​

147.059First416.667Worst1279.570Average

babylon-wtb​

44.017​

24.155First51.020Worst69.204Average

base64-SP​

285.788​

166.667First344.828Worst406.143Average

Basic​

465.875​

178.571First540.541Worst1047.535Average

bomb-workers​

29.400​

30.488First27.174Worst30.675Average

Box2D​

213.522​

54.945First232.558Worst761.844Average

cdjs​

140.750​

69.444First189.873Worst211.470Average

chai-wtb​

84.322​

64.103First92.593Worst101.010Average

coffeescript-wtb​

31.612​

24.155First30.864Worst42.373Average

crypto​

813.443​

357.143First952.381Worst1582.447Average

crypto-aes-SP​

416.630​

208.333First512.821Worst676.906Average

crypto-md5-SP​

242.399​

156.250First285.714Worst319.035Average

crypto-sha1-SP​

278.930​

138.889First357.143Worst437.500Average

date-format-tofte-SP​

140.525​

128.205First136.054Worst159.091Average

date-format-xparb-SP​

145.908​

142.857First130.719Worst166.341Average

delta-blue​

872.409​

172.414First1333.333Worst2888.350Average

earley-boyer​

380.991​

172.414First400Worst801.887Average

espree-wtb​

25.385​

19.455First26.316Worst31.949Average

first-inspector-code-load​

148.596​

156.250First135.135Worst155.393Average

FlightPlanner​

509.148​

250First571.429Worst923.913Average

float-mm.c​

11.319​

11.521First10.881Worst11.568Average

gaussian-blur​

257.314​

172.414First294.118Worst335.968Average

gbemu​

75.223​

32.051First74.906Worst177.294Average

gcc-loops-wasm​

21.416​

217.391Startup2.110Runtime

hash-map​

435.091​

200First476.190Worst864.826Average

HashSet-wasm​

30.676​

161.290Startup5.834Runtime

jshint-wtb​

48.090​

33.557First53.191Worst62.305Average

json-parse-inspector​

149.169​

147.059First144.928Worst155.738Average

json-stringify-inspector​

124.422​

156.250First87.719Worst140.533Average

lebab-wtb​

45.633​

27.473First51.020Worst67.797Average

mandreel​

88.220​

42.017First121.212Worst134.812Average

ML​

72.645​

49.020First86.957Worst89.939Average

multi-inspector-code-load​

271.800​

161.290First155.039Worst802.969Average

n-body-SP​

808.558​

500First952.381Worst1110.075Average

navier-stokes​

585.132​

200First833.333Worst1202.020Average

octane-code-load​

904.165​

833.333First833.333Worst1064.401Average

octane-zlib​

20.571​

12.563First25.974Worst26.677Average

OfflineAssembler​

103.891​

56.818First128.205Worst153.936Average

pdfjs​

120.986​

53.191First137.931Worst241.379Average

prepack-wtb​

33.507​

22.624First36.496Worst45.558Average

quicksort-wasm​

279.946​

454.545Startup172.414Runtime

raytrace​

433.778​

121.951First714.286Worst937.008Average

regex-dna-SP​

431.769​

416.667First400Worst482.955Average

regexp​

316.862​

277.778First294.118Worst389.398Average

richards​

602.979​

250First869.565Worst1008.475Average

richards-wasm​

64.679​

625Startup6.693Runtime

segmentation​

38.776​

33.113First38.071Worst46.247Average

splay​

236.420​

294.118First121.951Worst368.421Average

stanford-crypto-aes​

317.800​

142.857First444.444Worst505.523Average

stanford-crypto-pbkdf2​

551.282​

312.500First645.161Worst831.006Average

stanford-crypto-sha256​

501.261​

312.500First540.541Worst745.614Average

string-unpack-code-SP​

552.364​

384.615First540.541Worst810.627Average

tagcloud-SP​

209.354​

142.857First206.186Worst311.518Average

tsf-wasm​

54.600​

357.143Startup8.347Runtime

typescript​

13.241​

7.837First14.006Worst21.148Average

uglify-js-wtb​

26.016​

18.587First28.090Worst33.727Average

UniPoker​

280.466​

178.571First312.500Worst395.349Average

WSL​

0.981​

1.653Stdlib0.583MainRun

 

[rethcirE]

Left Chrome - Version 103.0.5060.134 (Official Build) (64-bit)
Right Opera GX - LVL 4 (core: 89.0.4447.64)
OS - Windows 10 Pro 21H2 19044.1826

Detailed Specs in Signature

Curious how Edge would fare but not curious enough to actually install it. Banished that browser from my OS almost immediately.

 

[SchumannFrequency]

I noticed today that Brave (via Linux compatibility) is (usually) faster than the native Chromium on FreeBSD:

My assumption is that with an Intel i7 12700K or an R7 5800X without overclocking I will end up above 300 anyway via FreeBSD and Brave.

 

[bug]

Ooh, a browser benchmark, I haven't seen one of these in ages, ever since Futuremark phased out Peacekeeper

It's different. Peacekeeper was mostly about HTML/CSS, this one seems to be more JS and WebAssembly focused (more like Octane, I'd say). Nothing inherently bad with either, given a choice I'd prefer we had both around.

 

[SchumannFrequency]

It's different. Peacekeeper was mostly about HTML/CSS, this one seems to be more JS and WebAssembly focused (more like Octane, I'd say). Nothing inherently bad with either, given a choice I'd prefer we had both around.

If you are interested in CSS performance there is still StyleBench: https://perftest.netlify.app/stylebench/
You also have the Maze Solver benchmark for CSS: https://testdrive-archive.azurewebsites.net/Performance/MazeSolver/Default.html
I don't know of any benchmark that focuses solely on HTML, but CanvasMark also tests HTML performance to a large extent: http://www.kevs3d.co.uk/dev/canvasmark/

As everyone knows, JavaScript is not efficient for web apps that require more computing power. We will probably see a lot more heavier web apps in the future. For WebAssembly you have the following benchmarks:

https://openbenchmarking.org/embed.php?i=2207047-NE-ALDERLAKE22&sha=5ad092f891fa&p=2

https://openbenchmarking.org/embed.php?i=2207047-NE-ALDERLAKE22&sha=32e00d66d318&p=2

 

[I hit the lottery]

i512600k @ 5.4
16 gig shit bird micron ddr5
6700xt
champ mindset.
and chrome.

 

[Det0x]

5950x @ 24/7 settings
score = 275

Ill be back in ~1months with a 7950x

 

[I hit the lottery]

yeah man. Imma check out another i5 I think or MAYBE the i7... but the i5 13600 6+8 sounds pretty dank for 300$.. and it looks to be smoking the 12900k in all aspects....so... if I catch the chip lottery this time, imma be a problem for the higher paying folk, more so even than with this 12600k.. but yeah the 7950x is guna be dank...wish I could warrant shelling at 700-800 got it.. but the real world performance difference outside of benchmarks is like 5%... real talk ....between the i5 Ks and everything else...Im trying to ditch this 6700xt as good as it is... i need a 6900 or a 3090... this i5 is maxing out this 6700xt not even on kill. smh....still jealous tho... im must admit lol..But a gpu serious upgrade would throw me so much more than the 7xxx series could get me above the i5s.. tbh.

 

[Athlonite]

Vivaldi

5.4.2753.47 (Stable channel) (64-bit)

Looks like Vivaldi have been making improvements since the last time I ran this

 

[SchumannFrequency]

It's different. Peacekeeper was mostly about HTML/CSS, this one seems to be more JS and WebAssembly focused (more like Octane, I'd say). Nothing inherently bad with either, given a choice I'd prefer we had both around.

I tested the CSS performance of my 10 year old cheap hardware today and was quite surprised at the result:

You see here that the newer and more expensive R9 5950X doesn't get better results than 3.7 in Maze Solver.
Lower results are better in this benchmark.

Firefox 95 vs. Chrome 97 Browser Performance On Linux - Phoronix

www.phoronix.com

Brave is lightning fast on my old and cheap hardware. I think FreeBSD explains why it currently feels like a supercomputer and gives these sort of inexplicable results.

 

[I hit the lottery]

I tested the CSS performance of my 10 year old cheap hardware today and was quite surprised at the result: View attachment 260924
You see here that the newer and more expensive R9 5950X doesn't get better results than 3.7 in Maze Solver.
Lower results are better in this benchmark.

Firefox 95 vs. Chrome 97 Browser Performance On Linux - Phoronix

www.phoronix.com

Brave is lightning fast on my old and cheap hardware. I think FreeBSD explains why it currently feels like a supercomputer and gives these sort of inexplicable results.

old hardware defeated.

 

[SchumannFrequency]

old hardware defeated.

We should be able to see the second digit after the decimal point to know this

On the same page you can generate a second maze (by clicking on the circular arrow)

Here I get sometimes 2.1, and sometimes 2.2, what is your result there?