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

Post your SilverBench multi-core result

Joined
Jul 15, 2022
Messages
1,018 (0.99/day)
VySond8.png


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. https://silver.urih.com/
Please include your specific hardware and software when you submit a result.
 
To others who'd post screenshots, be aware that website shows your IP so make sure to cut that part out (especially if you have static IP).

5800x3D, CO -30, 103Mhz BCLK, rest of the system in my specs

Browser, Opera GX, Chromium version:116.0.5845.188

Edit: I restarted the system to plug my new fans and ssd and I got over 77K score now (First result was 72k) :confused: Nothing changed settingswise and temperatures arent really effected. Maybe I just needed a restart...
Edit2: got slightly higher score after trying again
 

Attachments

  • Opera Snapshot_2023-11-14_162443_silver.urih.com.png
    Opera Snapshot_2023-11-14_162443_silver.urih.com.png
    584.9 KB · Views: 92
  • Opera Snapshot_2023-11-14_174205_silver.urih.com.png
    Opera Snapshot_2023-11-14_174205_silver.urih.com.png
    574.7 KB · Views: 75
  • Opera Snapshot_2023-11-14_180143_silver.urih.com.png
    Opera Snapshot_2023-11-14_180143_silver.urih.com.png
    572.3 KB · Views: 77
Last edited:
Quick run before work

Screenshot 2023-11-14 073302.png
 
Edit: I restarted the system to plug my new fans and ssd and I got over 77K score now (First result was 72k) :confused: Nothing changed settingswise and temperatures arent really effected. Maybe I just needed a restart...
Edit2: got slightly higher score after trying again

In terms of performance, I have a very stable system because I have reduced the startup processes to the bare minimum.

I did the SilverBench test 10 times and all the values fluctuated between 848xx and 856xx. The results of the test normally do not fluctuate as much as in your case.

My suspicion is that perhaps there was a background process running when you did the initial tests. Either Opera updated to a newer version.
 
lkNV9pq.png


Result: 88 253
Hardware: Intel 12600KF (stock) -- Kingston 6200 MHz CL36 -- Sapphire RX 7600 -- BIOSTAR B760MZ-E PRO -- Antec P6 -- Xilence XP550 -- ARCTIC i35 -- EVO 850 500GB
Software: Alpine Linux -- Chrome (unstable) Flatpak version -- river wm -- open source GPU driver

The highest result I got with this version of Chrome was around 88 500.
 
Browser: Vivaldi
System as per specs below

Screenshot 2024-03-20 072130.png
 
chrome Version 123.0.6312.58 :peace:

Capture d'écran 2024-03-24 065425.png
 
silverbench.jpg
 

Attachments

  • silverbench.jpg
    silverbench.jpg
    55.1 KB · Views: 79
Screenshot.jpg


Samsung Galaxy S23, Snapdragon Gen 2. Just a couple of years of mobile development and you guys are done for!
 
Last edited:
Comparison.jpg


It's interesting to see the difference in efficiency between a 5 year old top gaming laptop CPU and a recent phone CPU. Almost 5 times better efficiency!
 
View attachment 350035

It's interesting to see the difference in efficiency between a 5 year old top gaming laptop CPU and a recent phone CPU. Almost 5 times better efficiency!

iOS is better optimized for JavaScript, so the iPhone 15 may score higher than 27 000 in SilverBench.

I think it would be interesting if smartphone CPUs were used for real desktop systems running desktop windows/BSD/Linux operating systems.
I find it strange that this is not yet happening because this way you can use much more compact desktop cases, which is what most people want anyway.

You also don't need fans for smartphone CPU/APUs. It's a better auditory experience too.
 
I'm still sorting my clocks but I've noticed the site randomizes threads. I'm using 9 for this one.
Brave 1.66.118
R5 3600↗4.0GHz
64GB DDR4-3200↗3800
MCLK/FCLK/UCLK =1900
and NO video drivers

1717581724678.png
 
iOS is better optimized for JavaScript, so the iPhone 15 may score higher than 27 000 in SilverBench.
Ah, makes sense. I suppose my comparison is not quite accurate because it doesn't take into account how well JavaScript is being run on each device. Though, to be fair, iPhone 15 A16 CPU is ~30% faster than Snapdragon 8 Gen 2, so even under same optimization, over 27000 makes sense.

I'm still sorting my clocks but I've noticed the site randomizes threads. I'm using 9 for this one.
Brave 1.66.118
R5 3600↗4.0GHz
64GB DDR4-3200↗3800
MCLK/FCLK/UCLK =1900
and NO video drivers

View attachment 350190
I've only seen Brave browser do that (it detects 3 threads out of 12 for me). Chrome works much better
 
I've only seen Brave browser do that (it detects 3 threads out of 12 for me). Chrome works much better

I also noticed that Brave doesn't use all the threads for this test.
In my case it only used 10 of the 16 threads, so it seems to be using the number of cores without using hyper-threading. (in my case)
 
1717587046581.png

i7-13800H (Dell Precision 5680)
 
Chrome works much better

In terms of performance it works better. It works less well in terms of security.

The Hyper-threading feature comes enabled on computers by default for performance boosting, but in a detailed post published Tuesday, OpenBSD maintainer Mark Kettenis said such processor implementations could lead to Spectre-style timing attacks.

Greg KH on disabling hyper-threading: "OpenBSD was right"


Intel 15th-Generation Arrow Lake-S could abandon Hyper-Threading technology

As far as I can tell, OpenBSD left hyper-threading disabled 6 years later.
 
It looks like quite a low score for an M3 Pro. Are you sure it was using maximum power? Perhaps try a different browser.

It is Safari that is slower than Chrome in this test and maybe it doesn't use all cores in Safari either.
But honestly, I find it strange that Safari is slower than Chrome because the test only uses JS.
 
Back
Top