Post your JetStream 2 speeds!

[Marcus Vinicus]

System as in specs below
Google Chrome Canary 117.0.5922.2

 

[SchumannFrequency]

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

 

[b0urne_]

Browser = SlimJet portable x64
Version 40.0.2.0 (based on Chromium 114.0.5735.16) (Official Build) (64-bit)

 

[SchumannFrequency]

Software: Archcraft -- Brave version 1.61.5 Chromium: 118.0.5993.54 nightly -- bspwm -- Nvidia proprietary GPU driver
Hardware: Intel 12600KF -- Kingston dual-channel 6000 MHz CL40 -- GTX 650 1GB -- BIOSTAR B760MZ-E PRO -- Antec P6 -- Xilence XP550 -- ARCTIC i35 -- EVO 850 500GB

 

[thexman5023]

hahahh looking at the list i win.......

have a go if you thinks yours is faster

 

[jiajo]

version chromée 118.0.5993.72

 

[SchumannFrequency]

Software: ALT Sisyphus -- LXQt -- Brave (snap edge channel) -- 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

 

[MSDOS Retro]

CPU: i5-4440
RAM: 8GB
GPU: Gigabyte GTX 1060 Windforce OC 3GB
SSD: Samsung 860 QVO 1TB
OS: Debian 12
Browser: Firefox 115.3.0esr (64-bits)

 

[SchumannFrequency]

Software: ALT Sisyphus -- LXQt -- Chrome (beta version) -- 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

 

[Det0x]

Chrome latest version: 118.0.5993.89

7950x3d PBO CO
8000MT/s 2133mhz FCLK (crappy IMC)

JetStream2 Score = 408.751 points

 

[Frozoken]

Chrome latest version: 118.0.5993.89

7950x3d PBO CO
8000MT/s 2133mhz FCLK (crappy IMC)

JetStream2 Score = 408.751 points
View attachment 318752

Damn, the one other guy with tuned ram. My 13600k got real close to your 7950x3d tho ahaha. 5.6ghz all core, 4.9ghz on the ring, 4.5ghz on the ecores and 6666mhz ddr5.

 

[SchumannFrequency]

Software: PCLinuxOS -- XFCE -- Nvidia proprietary driver -- F2FS (root) file system -- Vivaldi 6.5.3189.3
Hardware: Intel 12600KF (stock) -- Kingston 6000 MHz CL40 -- GTX 650 1GB -- BIOSTAR B760MZ-E PRO -- Antec P6 -- Xilence XP550 -- ARCTIC i35 -- EVO 850 500GB

This Intel 12600KF also supports overclocking and RAM @8200 MHz so it can normally score higher in JetStream 2.1 in optimal conditions.

 

[lolek86]

Ryzen 7600 @ 5.5GHZ , 6000MT CL28 Memory , chrome 120, windows 11 23h2

Once again serious gains for zen 4 systems with tuned memory compared to scores from reviews posted above.

 

[Marcus Vinicus]

Jetstream 2.1
Vivaldi 6.5.3206.42
Windows 11 Pro Beta Insider Channel 23H2 22635.2915
See signature below for other system specs.

 

[SchumannFrequency]

JetStream version 2.0

Software: ROSA Fresh Desktop 12.4 -- LXQt -- Nvidia proprietary driver -- XFS as root -- Vivaldi 6.5.3222.3
Hardware: Intel 12600KF (stock) -- Kingston 6000 MHz CL40 -- GTX 650 1GB -- BIOSTAR B760MZ-E PRO -- Antec P6 -- Xilence XP550 -- ARCTIC i35 -- EVO 850 500GB

 

[Paramount]

Ryzen 7600x PBO2 CO-38
Tuned DDR5 6400@CL30/FCLK 2133

I think Zen 4 is win here vs Intel

 

[SchumannFrequency]

I think Zen 4 is win here vs Intel

Specifically in JetStream 2, the Ryzen 7600X scores very well considering its price.
Many other browser benchmarks exist, and it is fairly difficult to say which one gives the most realistic picture of real-world performance.

I'm curious though what result you get in these tests.

Post your SilverBench multi-core result

Software: ALT Sisyphus -- LXQt -- Chromium version 121.0.6115.2 -- 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 You can take the test on this page...

www.techpowerup.com

SunSpider 1.0 JavaScript benchmark results

http://proofcafe.org/jsx-bench/js/sunspider-1.0/results.html?%7B%22v%22:%20%22sunspider-1.0%22,%20%223d-cube%22:%5B633,642,643,631,630,633,643,641,638,644%5D,%223d-morph%22:%5B846,938,900,850,902,873,922,908,911,899%5D,%22access-binary-trees%22:%5B432,442,398,391,432,431,431,428,422,431%5D,%22acc...

www.techpowerup.com

Hixie's DOM core performance tests

Result: 17 ms Hardware: AMD R5 PRO 3400G + dual-channel DDR4 @2666MHz Software: Epiphany browser 44.3 (WebKit version 605.1) + NixOS + Cinnamon You can take the test on this page. http://www.hixie.ch/tests/adhoc/perf/dom/artificial/core/001.html When you post, please include your specific...

www.techpowerup.com

JSBench.Me test results

I score 1.1 Md ops/s (Md or Mrd = milliard and M = million) Hardware: Intel i3-3240 + 8GB RAM @1600MHz DDR3 dual-channel + NVIDIA GTX 650 1GB + EVO 850 500GB Software: Epiphany browser 44.2 (WebKit version 605.1) + NixOS + Cinnamon Meanwhile, version 44.3 of the Epiphany browser is now...

www.techpowerup.com

Performance testing: setting image src attributes dynamically

Result: 402 ms Hardware: Intel i3-3240 + 8GB RAM @1600MHz DDR3 dual-channel + NVIDIA GTX 650 1GB + EVO 850 500GB Software: NixOS 23.11 & Firefox 114.0.1 You can take the test on this page. https://www.hixie.ch/tests/adhoc/perf/dom/real-world/005.html When you post, please include your specific...

www.techpowerup.com

Dromaeo DOM Core Tests

I score 6376.84 runs/s Hardware: Intel i3-3240 + 8GB RAM @1600MHz DDR3 dual-channel + NVIDIA GTX 650 1GB + EVO 850 500GB Software: Epiphany browser 44.3 (WebKit version 605.1) + NixOS + Cinnamon You can take the test on this page. https://dromaeo.com/?dom When you post, please include your...

www.techpowerup.com

Benchmarking Unity graphics performance in WebGL

You can take the test on this page. https://files.unity3d.com/jonas/benchmark2015/ When you post, please include your specific hardware and software. Hardware: Intel i3-3240 + 8GB RAM @1600MHz DDR3 dual-channel + NVIDIA GTX 650 1GB + EVO 850 500GB Software: FreeBSD 13.1 + Chromium version...

www.techpowerup.com

WebAssembly Fibonacci benchmarking

I score 350.7 ms I use the following hardware: Intel i3-3240 + 8GB RAM @1600MHz DDR3 dual-channel + NVIDIA GTX 650 1GB + EVO 850 500GB I use Epiphany browser 44.2 (WebKit version 605.1) + NixOS + Cinnamon as software. You can take the test on this page...

www.techpowerup.com

Speed test: innerHTML versus DOM manipulation redux

Result: innerHTML - 33 milliseconds ; DOM nodes - 32 milliseconds Hardware: Intel i3-3240 + 8GB RAM @1600MHz DDR3 dual-channel + NVIDIA GTX 650 1GB + EVO 850 500GB Software: NixOS 23.11 & Firefox 114.0.1 You can take the test on this page. https://andrew.hedges.name/experiments/innerhtml/ When...

www.techpowerup.com

WebAssembly imageConvolute benchmarking

I score 45.7700 ms I use the following hardware: Intel i3-3240 + 8GB RAM @1600MHz DDR3 dual-channel + NVIDIA GTX 650 1GB + EVO 850 500GB I use Chromium 111.0.5563.147 and FreeBSD 13.1 as software. You can take the test on this page...

www.techpowerup.com

Post your PSPDFKit WebAssembly Benchmark speeds!

You can take the test on this page: https://pspdfkit.com/webassembly-benchmark/ I myself score this result: I use the following hardware: Intel i3-3240 + 4GB RAM @1600Mhz DDR3 single channel + NVIDIA GTX 650 1GB + EVO 850 500GB If I were to overclock this CPU I should score around 3800. The...

www.techpowerup.com

JavaScript library test results

Results: DOM attributes (prototype): 3846.35 runs/s DOM attributes (jQuery): 3200.36 runs/s DOM events (prototype): 3906.55 runs/s DOM events (jQuery): 82.22 runs/s Hardware: Intel i3-3240 + 8GB DDR3 @1600MHz dual-channel + NVIDIA GTX 650 1GB + EVO 850 500GB Software: Void Linux + i3wm + Nyxt...

www.techpowerup.com

Furthermore, a benchmark is not always indicative of reality.
Consider Metro Exodus where Nvidia scored better than AMD in the benchmark. But in-game measurements showed that AMD performed relatively better than Nvidia in the game itself.

Benchmark results RX 580 vs GTX 1060 Metro Exodus:

Real world results RX 580 vs GTX 1060 Metro Exodus: https://www.techspot.com/review/1990-geforce-1060-vs-radeon-580/

You should always give the final word to measurements from realistic situations, not benchmarks.

I also think that in reality the Intel CPUs draw less power than their AMD counterparts for people who do a lot of browsing and light office tasks.

 

[lolek86]

Ryzen 7600x PBO2 CO-38
Tuned DDR5 6400@CL30/FCLK 2133

I think Zen 4 is win here vs Intel

Try geekbench 6, there is a thread here. It is very sensitive to memory tuning or overclock. I had gained about 12% single 24% multithread performance tinkering with PBO/OC/memory/fancurve on my nonX 7600 compared to lousy stock bios settings.

ASUS System Product Name vs ASUS System Product Name - Geekbench

Specifically in JetStream 2, the Ryzen 7600X scores very well considering its price.
Many other browser benchmarks exist, and it is fairly difficult to say which one gives the most realistic picture of real-world performance.

Yeah YMMV, especially with many benchmarks still favoring compiling for intel. Compared to many initial reviews, it's funny to see ZEN4 architecture spreading its wings with simple(for enthusiasts) tweaks.
I mean, almost any cheap 6000 sticks can do cl28 trfc 300-500 etc and almost any CPU 7600 can reach 5.5GHZ, hanging there quite easly with undervolt even with simple 4 heatpipe air cooler

 

[Paramount]

Ryzen 7600x PBO2 CO-38
Tuned DDR5 6400@CL30/FCLK 2133

I think Zen 4 is win here vs Intel

View attachment 328324

same setting but change power plan from balance to ultimate
7600x FTW

Try geekbench 6, there is a thread here. It is very sensitive to memory tuning or overclock. I had gained about 12% single 24% multithread performance tinkering with PBO/OC/memory/fancurve on my nonX 7600 compared to lousy stock bios settings.

ASUS System Product Name vs ASUS System Product Name - Geekbench

Yeah YMMV, especially with many benchmarks still favoring compiling for intel. Compared to many initial reviews, it's funny to see ZEN4 architecture spreading its wings with simple(for enthusiasts) tweaks.
I mean, almost any cheap 6000 sticks can do cl28 trfc 300-500 etc and almost any CPU 7600 can reach 5.5GHZ, hanging there quite easly with undervolt even with simple 4 heatpipe air cooler

i don't found the thread , but result look good what you think ?

 

[lolek86]

Thread is here, maybe mod can move our geekbench OT to proper place

geekbench6 scores

VDD: 1.44v VDDQ/IO: 1.30v

www.techpowerup.com

Very good score. So far my best is 3093 single 14820 multi under windows and 3220/15565 under catchyOS linux. Finally someone beat my 7600 score on geekbench browser!

Look at this , most people without proper tune are only pushing about 2700ish.

Geekbench Search - Geekbench

Where is your cpu boosting clocks at exactly? Air or water cooling? As for memory, in my case so far i mostly tinkered at 6000 searching lowest timings. For me 6400 was somehow unstable on beta bioses but i see official agesa 1.1.1.1 is here so i might try soon again.

 

[Paramount]

Thread is here, maybe mod can move our geekbench OT to proper place

geekbench6 scores

VDD: 1.44v VDDQ/IO: 1.30v

www.techpowerup.com

Very good score. So far my best is 3093 single 14820 multi under windows and 3220/15565 under catchyOS linux. Finally someone beat my 7600 score on geekbench browser!

Look at this , most people without proper tune are only pushing about 2700ish.

Geekbench Search - Geekbench

Where is your cpu boosting clocks at exactly? Air or water cooling? As for memory, in my case so far i mostly tinkered at 6000 searching lowest timings. For me 6400 was somehow unstable on beta bioses but i see official agesa 1.1.1.1 is here so i might try soon again.

i think the point is my cpu can do CO -38 all cores fully stable even in SHA/FPU Julia
the temp on cinebench never reach 82 for hottest one
according to HWinFO , max core reach 5.65Ghz but never see this no. during gaming ..
also my ram kit is precisely tuned ( Aida64 latency is 53.7 )/ I used AGESA 1.0.0.7c
for cooling i used LT720 with Thermaltake riing plus fan

 

[lolek86]

i think the point is my cpu can do CO -38 all cores fully stable even in SHA/FPU Julia
the temp on cinebench never reach 82 for hottest one
according to HWinFO , max core reach 5.65Ghz but never see this no. during gaming ..
also my ram kit is precisely tuned ( Aida64 latency is 53.7 )/ I used AGESA 1.0.0.7c
for cooling i used LT720 with Thermaltake riing plus fan

Ahh 5.65 allowed i see, that explains a lot. I run at stable 5.5ghz / -40 CO so far. Wonder if my 7600 nonX can do a little more or it is maxed out. From briefly trying nothing was totaly stable, just curious, what is your core vdd value at that speed?

 

[Marcus Vinicus]

Jetstream 2.1
Vivaldi 6.5.3206.48
Windows 11 Pro Beta Insider Channel 23H2 22635.2921
See signature below for other system specs.

 

[Paramount]

Ahh 5.65 allowed i see, that explains a lot. I run at stable 5.5ghz / -40 CO so far. Wonder if my 7600 nonX can do a little more or it is maxed out. From briefly trying nothing was totaly stable, just curious, what is your core vdd value at that speed?

According to HWinfo its 1.28v
core 2 & 3 reach 5.65 in certain app like cb 2024 ( they both the hottest two )

 

[lolek86]

According to HWinfo its 1.28v
core 2 & 3 reach 5.65 in certain app like cb 2024 ( they both the hottest two )

Looks like your X version is from bit better chip bin as should be expected. My needs precisely 1.285 vdd for that fixed 5.5G but its stable there. Even 0.005V more or less is quick bluscreen. Either thats end of the road for my chip or there is one more way.


During ryzen master scanning for suggested undervolt i noticed it tipped different values at different max frequencies. At stock 5150 it suggested -30, at 5350(+200 limit) it was -40.

For my 5.5 fixed OC i used 5350 curve settings as a base with slowly adding clock and voltage step by step. Given ryzen master insight, no wonder this falls apart sharply after certain threshold, especially near edge of chip possibilities.

When I will be in the mood for few tens of reboots i might try again by scraping negative curve, again searching manually if there is anything higher than 5.5 possible and only then search again for new negative curve.

Either way I won't complain. I was lucky with memory kit too. Cheap kingston sticks only rated for cl40 tightened up at cl28 with tight other timings, certainly not expected that when buying it. It is tearing through my use cases, not planning any upgrade until last supported am5 line along with whole different ddr5 kits available by then.

Userbenchmark is funny too, it gives funny description of your system. Mine is "UFO" and " operated by technical master " comparing performance to typical for given parts.