Windows 10 Meltdown/Spectre Patch Performance Impact Assessment

Conclusion - Will end-users notice a dip to general performance when using Intel processors?

Windows 10 Meltdown/Spectre Patch Performance Impact Assessment

Conclusion

When initial reports regarding Spectre and Meltdown were emerging from the shadows, a performance impact of 5-30% was a comment that was mentioned everywhere online, leaving many to wonder whether or not the final update would affect their user experience. 

While the 5-30% number was always said to apply to web servers and virtualisation environments the fear amongst Intel users was how much performance they would need to sacrifice on the altar of safety. Would our games suddenly be jerkier than before? Would our render times increase? Will I have to change to an AMD CPU? These were all questions that people were asking, but thankfully the results are in, and we can assess how much the damage is. 

Before we get into the testing, we must preface this with the fact that AMD systems will not be affected in the same way as an Intel-based system by this update, because AMD CPU architectures are not vulnerable to Meltdown. This difference means that only Intel systems will require page table isolation (PTI) and other similar security solutions. 

In general, our system performance was down across the board, the impact of the patch was so minor that the whole could be called inconsequential from a performance standpoint. The most noticeable performance changes we found were in the region of 3%, too low to even notice outside of graph. 

When gaming with our i7 6800K with all cores active we found that there was no noticeable difference in gameplay, with only a minor framerate variance, though when we moved to a dual-core configuration, we found the difference to be much more consistent across multiple gameplay runs. These tests were in a scenario where each game was in an extremely CPU limited scenario, but even then the performance difference was 3% or lower, only enough to lower the minimum framerate by only a handful of FPS. The difference was consistent but consistently meh.   

One other factor that must be remembered for our games testing is that we intentionally reduced our CPU to a dual-core, quad-thread configuration to make these consistent framerate differences appear. We would hope that no PC gamers today is using a GTX 1080 with a mere i3-class processor, a last generation i3-class CPU configuration at that. With a system with a more balanced approach to CPU and GPU selection will not see such a performance impact, making this patch's performance impact on gamers nonexistent in most cases.   

To finish off our tests show that in general workloads that Intel users should not see a noticeable difference in overall system performance, though please bear in mind that the difference does vary from workload-to-workload and some of Intel's pre-Haswell CPUs are said to be affected more, even if marginally. 

Intel users should have no qualms installing Windows 10's latest security update, at least from a performance standpoint, kinda anticlimactic really. 

You can join the discussion on the performance impact of Windows 10's Meltdown/Spectre security update on the OC3D Forums.  

«Prev 1 2 3 4 5 6 7 8 9 10 11 Next»

Most Recent Comments

05-01-2018, 15:11:33

AlienALX
Just going to a pic dude.Quote

05-01-2018, 15:26:48

RedSymphony
"we must preface this with the fact that AMD systems will not be affected in the same way as an Intel-based system by this update, because AMD CPU architectures are vulnerable to Meltdown".
I think you mean invulnerable, right?
Nice to see the performance hit for the average user is minimal.Quote

05-01-2018, 16:25:51

WYP
Quote:
Originally Posted by AlienALX View Post
Just going to a pic dude.
Quote:
Originally Posted by RedSymphony View Post
"we must preface this with the fact that AMD systems will not be affected in the same way as an Intel-based system by this update, because AMD CPU architectures are vulnerable to Meltdown".
I think you mean invulnerable, right?
Nice to see the performance hit for the average user is minimal.
Fixed and fixed.

Yeah, I am very happy to know that the performance hit is minimal for most standard users. Only slightly annoying that so much time was spend testing it.Quote

05-01-2018, 17:48:33

donp
Being new here, Hello first of all. Plus thank you for taking the time and effort with the testing Thank you again, regardsQuote

06-01-2018, 18:22:58

WYP
Quote:
Originally Posted by donp View Post
Being new here, Hello first of all. Plus thank you for taking the time and effort with the testing Thank you again, regards
You are welcome. Your thanks are greatly appreciated.Quote
Reply
x

Register for the OC3D Newsletter

Subscribing to the OC3D newsletter will keep you up-to-date on the latest technology reviews, competitions and goings-on at Overclock3D. We won't share your email address with ANYONE, and we will only email you with updates on site news, reviews, and competitions and you can unsubscribe easily at any time.

Simply enter your name and email address into the box below and be sure to click on the links in the confirmation emails that will arrive in your e-mail shortly after to complete the registration.

If you run into any problems, just drop us a message on the forums.