Windows 11 VS. Windows 10 Performance Comparison – W10 + Superlite vs W11
20 June 2021
20H2 Benchmark Benchmarks BL3 Borderlands Borderlands 3 CINEBENCH Cinebench R15 Cinebench R20 Cinebench R23 FH4 Forza Forza Horizon Forza Horizon 4 Ghost Spectre Performance Comparison SYNTHETIC BENCHMARKS W10 W11 WINDOWS WINDOWS 10 Windows 11

So Windows 11 got leaked ahead of its rumored reveal on June 24th.

And as expected all the usual faggot GrabblerTubers have rushed in with their own verdict primarily regarding the UI and other aspects of the new operating system. I mean it was kinda expected to be honest, with Microsoft falling back on their statement that Windows 10 would be their last OS they instead pushed for its demise sometime during the middle of 2025.

So with W10 effectively obsolete, and despite the fact that Microsoft can’t seem to update Windows 10 without bricking it to pieces it makes sense that they would effectively put lipstick on a pig and usher out a whole “new” operating system that’s visually a reskin of their previous.

Now you can just go about anywhere else for some faggot to tell you how good their oversimplified UI looks and functions, I’m personally disgusted by it.

But that’s not we’re here for today at all, you see in all the kerfuffle regarding the new task bar, start menus and rounded tabs (which I actually like) nobody has seemed to bother with some performance comparison between this leaked Windows 11 versus Windows 10.

Now when it comes to using Windows 10 I strongly recommend either going for those older LTSB/LTSC versions simply for stability, functionality and of course being bloat free. If however you desire performance or the features that have come with the major release updates for Windows 10 in the absence since v1809 I hardly look no further than a custom third party “SUPERLITE” edition of Windows 10 curtesy of Ghost Spectre.

Barebones, with the bloat simply scrapped on the latest version of Windows? Whether or not you wish to put your trust in something like that is entirely up to you however I simply cannot overlook the fact that Ghost Spectre releases increase my performance whether in games or synthetics comparative to a traditional install of Windows 10.

So should you actually bother to wait to upgrade to Windows 11 in the future when a fully functioning “stable” OS is actually released? Or rather should you just set and forget a Windows 10 install mostly bloat free from the likes of Ghost Spectre when in the pursuit of performance?

Because in a pure apples to oranges comparison we’re going to find out which operating system users should be pirating for the best performance possible, come the time of Windows 11’s release.

Now because I’m quite lazy I’m only going to showcase whatever marginal differences in performance there happen to be across Cinebench alongside two actual games in the forms of Forza Horizon 4 and Borderlands 3.

Why such a small sample size you may ask? As mentioned above I’m lazy, I don’t get paid anything to do this sort of shit and honestly something is better than nothing. At the very least I hope that the takeaway from this will be that Windows 11…. or rather Windows 10.1 is shown to be as garbage as it looks, and more rather that you should be installing your very own “superlite” copy of Windows 10 instead.

So all in all we’ll be comparing Cinebench R15, R20 and R23 with a couple of DirectX 12 titles with a fresh install of the leaked Windows 11 along with a fresh install of a “SUPERLITE” Windows 10 copy by Ghost Spectre and I’ll be comparing them both to a rather bloated retail install of Windows 10 20H2 to act as a rough baseline.

Since the previous bout of performance testing from the H.A.G.S article, my system specifications have changed rather drastically.

CPU

Ryzen 9 5900X

GPU

MSI Radeon RX 5700XT Gaming X (w/ ETH mining BIOS) @ 2070/1792

RAM

32GB – 3733MHz @ 14-16-16-32 1T

 

Starting things off with the old faithful Cinebench R15 we’re seeing pretty much consistent figures apart from the standard Windows 10 falling behind that of Windows 11 and the skimmed SUPERLITE which scores 271 CB for its single core and 3922 CB for multi core.

With the Ghost installation taking the top score albeit marginally over Windows 11 by 1.49% in single threading and 0.20% in multi threading which may as well be within margin of error.

However a standard install of Windows 10 falls behind the gutted Ghost SUPERLITE by 2.26% in its single core score and 2.10% in multi-threading.

When moving onwards to Cinebench R20 things do change however in favor of the retail Windows 10 install no less.

With the standard 20H2 install of Windows 10 taking the top score in the much more preferable single core score of 634 CB while both the Ghost SUPERLITE and Windows 11 practically throw out identical scores for both the single core and multi core portion. But once more when it comes to the single core score both W11 and Ghost fall short of a standard install by 0.63% once again within margin of error.

And we see the same sort of thing with the multi core scores between all three operating systems with the top score coming by the SUPERLITE install at 9054 CB which is miraculously only 0.05% above that of the Windows 11 score while being 0.25% higher than that of the standard Windows 10 install.

Now for the last results for Cinebench, that being Cinebench R23 which is above all else a glorified benchmark purpose built for crApple’s M1 processor in my own personal opinion, whereas R20 was a pitiful attempt to claw back some points for the Intel camp considering that it utilizes AVX.

Glad to report that AMD’s IPC and multi threading capability is just that strong that Intel still can’t cope with these sorts of benchmarks.

But that aside we see a rather mixed bag with Cinebench R23 where this time around the traditional Windows 10 installation leads in both single and multi threading. With scores of 1609 CB for the single core and 22,842 CB for the multi core it’s rather dominate with a 2.22% margin over a SUPERLITE install and 3.6% over Windows 11 in the single core score.

Things are a little more mellow when it comes to their multi core scores with the 20H2 install edging out both the Ghost and Windows 11 by lower margins, 0.71% over Windows 11 and 0.92% over the SUPERLITE installation.

Finally getting into some games we have Forza Horizon 4 which is certainly a title that runs as good as it looks.

Now when it comes to games there are those that favor testing with higher resolutions where the GPU is the restricting component, essentially what we all should really be doing however for as long as I can remember people have adored Intel for their ability to once crush the competition when it comes to low resolution testing as a means of raw CPU performance of which provides less of a bottleneck.

This instance however has Forza Horizon 4 at 4K resolution with the Ultra graphics quality preset selected and utilizing the in-game benchmark sequence to see if there’s any change in overall performance where quite honestly there really shouldn’t be any.

Unsurprisingly both the standard Windows 10 installation combined with Windows 11 provide merely identical performance figures with 89 frames per second achieved with an FPS average within 0.1% of each other.

Shockingly though a Ghost SUPERLITE install simply doesn’t give a fuck about logic and reason for it had achieved 90 frames per second (a 1.12% increase) with an FPS average of 92.4.

And if a installation of the gutted SUPERLITE manages to provide a performance boost in a GPU bound scenario on Forza Horizon 4 then surely enough it ought to rip and tear through a low resolution run at 720p.

And surely enough it does just that taking a rather dominate lead of 1.71% over a standard W10 installation and 3.19% over a rather disappointing score on Windows 11.

Honestly I have no idea exactly what was going on with the performance on Windows 11 at 720p resolution, figures were all over the shop with runs sometimes going down as low as an achieved FPS of 340. Before adjusting memory timings and giving my 5900X a much needed kick up the arse I easily managed to produce 345 FPS at the very least, no matter what however Windows 11 just didn’t want to play along by giving me the numbers that it probably should have.

Not only however did the Ghost SUPERLITE thrash the other installs when it came to the FPS achieved but also in the average CPU render statistic with 1.85% over Windows 10 and 3.72% over Windows 11.

Finally onto our last game comparison which happens to be the biggest disappointment in gaming for me personally, I knew it was going to be complete utter woke trash but even I managed to still be disappointed of it. With the fourth installment of the franchise confirming that I may in fact have a fetish for cel shading sadly graphic quality aside the only pozzitive aspect about Borderlands 3 was its gun mechanic.

Now because I honestly couldn’t bring myself into actually playing the game instead rather I opted for the in-game benchmark feature. With this test being conducted at UWQHD resolution (3440×1440) which is honestly underrepresented in benchmarking figures. Utilizing the High preset showcases that a mere Radeon RX 5700 XT is probably not the ideal solution for such a high resolution but I digress.

Compared to Forza Horizon 4 we see a rather different story with both the Ghost SUPERLITE and Windows 11 taking charge over the standard Windows 10 installation with the overall victor being the Ghost install with a 2.11% margin over Windows 10 and a marginal increase of 0.22% over Windows 11.

And it’s a rather different story completely when lowering the quality and resolution down to 1280×720 with the Very Low graphic preset where we continue to see the gaming prowess of the Ghost SUPERLITE with a marginal increase of 0.16% over a standard Windows 10 install, however Windows 11 continues to flip flop given it’s at a 2.49% deficit to the Ghost install with an overall average FPS of 281.65 comparative to 288.87.

Overall across every single test shown above the surprising victor happens to be a standard installation of Windows 10 (20H2) which retails a marginal lead of 0.37% lead over Windows 11 but only 0.3% over the Ghost SUPERLITE.

Now, excluding Cinebench and focusing entirely on gaming performance the Ghost SUPERLITE install provides the best overall performance.

With a much more noticeable performance gap of 1.32% over a standard Windows 10 install and a whopping 2.72% over Windows 11 which in all honesty for the gaming portions has been a big let down. So much so that the standard copy of Windows 10 actually proves to be 1.38% faster than it when it comes to the gaming figures alone.

Conclusion

So overall what do I think of Windows 11?

I don’t like it. Microsoft seem incapable of releasing updates for Windows 10 without breaking something major so in all honesty why even bother with releasing another one after stating that Windows 10 would be their last? Because Windows 11 resembles its predecessor greatly albeit with a large amount of minor changes and some major ones at that as well. Fundamentally I hate its mobile-esk oversimplified start menu.

As said above Microsoft are putting lipstick on a pig. However upon using the leaked copy of Windows 11 the desktop experience minus the cancerous UI honestly felt much faster than that of Windows 10 though this very well could just be a placebo. The only aspect I actually liked about Windows 11 was the fact that tabs now have rounded edges, that’s pretty nifty.

But the conclusion to the overall performance between operating systems has gone in favor of the Ghost SUPERLITE which does way with the cancerous telemetry hidden beneath Windows 10 alongside removing many of the garbage programs that you’re unfortunately forced to have as standard.

Windows 10 in its current form is still a buggy broken mess, so why exactly would you even want to upgrade to Windows 11 as it will most likely release by the end of the year? You’d be setting yourself up into being a BETA tester for Microsoft’s latest and greatest telemetry and datamining operating system which without a doubt upon release will have even more telemetry and unique ways of collecting your information.

And much like how Windows 10 was pushed down our throats by Microsoft to garner as many installations as possible particularly from users on Windows 8/8.1 and primarily those on a Windows 7 operating system. Microsoft will more than likely push Windows 11 onto us enticing us with free installs and the like, but don’t be fooled.

This is pretty much why those concerned regarding having their data gathered by soulless corporations such as Microsoft they instead opt for installing a LTSC/LTSB copy of Windows 10 instead, of which considering the age of such versions you’d be far better off garnering the features and performance benefits from an equally gutted copy of a much newer Windows 10 version like those packaged by Ghost Spectre.

Performance across synthetics has proven to be within margin for the most part regardless of operating system however there’s no denying that a scrapped Windows 10 installation is by far your best choice of option if you truly seek the best gaming performance possible. Eventually however as the years tick by and LITE versions of custom Windows 11 ISO’s start to emerge perhaps it might be ideal to consider making the shift then.

blog comments powered by Disqus