Sekiro: Shadows Die Twice was one of the most anticipated games of 2019. Developers by the creators of the Dark Souls series, Sekiro uses an enhanced version of the engine that was used in Dark Souls 3. The game has just been released so it’s time now to benchmark it and see how it performs on the PC platform.
For this PC Performance Analysis, we used an Intel i7 4930K (overclocked at 4.2Ghz) with 16GB of DDR3 RAM at 2133Mhz, AMD’s Radeon RX580 and RX Vega 64, NVIDIA’s RTX 2080Ti, GTX980Ti and GTX690, Windows 10 64-bit, GeForce driver 419.35 and the Radeon Software Adrenalin 2019 Edition 19.3.3. NVIDIA has not included any SLI profile for this title in its latest drivers, meaning that our GTX690 behaved similarly to a single GTX680.
FromSoftware has added a respectable amount of graphics settings to tweak. PC gamers can adjust the quality of Textures, SSAO, Depth of Field, Motion Blur, Shadows, Lighting, Effects, Volumetric, Reflections, Water Surface and Shaders. Furthermore, and contrary to Dark Souls 3, Sekiro: Shadows Die Twice also features proper K&M on-screen prompts.
Let’s start this time with our GPU benchmarks. Sekiro: Shadows Die Twice is locked at 60fps and most of our graphics cards were able to offer a smooth 60fps experience at 1920×1080 on Max settings. While some users reported various frame pacing and/or stuttering issues, we did not experience any (with the exception of some stutters during the in-engine cut-scenes whenever the camera was changing).
Now while the game is easy on the GPU side, it appears to require more than four CPU threads in order to be properly enjoyed. FromSoftware lists an Intel Core i3-2100 as a minimum CPU and an Intel Core i5-2500K as a recommended CPU. However, those with these particular CPUs will not be able to enjoy the game.
In order to find out how the game performs on a variety of CPUs, we simulated a dual-core and a quad-core CPU. We’ve also unlocked the framerate in order to get a better idea of the performance difference between these setups (from what we’ve seen the “unlock fps” mod does not introduce new issues). Without Hyper Threading, our simulated dual-core system was unable to run the game as it was stuttering like crazy. With Hyper Threading enabled, we had a minimum of 8fps and an average of 19fps in the first area at 1920×1080 on Max settings.
But what about our simulated quad-core system? Well, here is something really interesting. Without Hyper Threading, the game was unable to hit 60fps as it was running with a minimum of 35fps and an average of 40fps (with all four CPU cores being used to their fullest). When we enabled Hyper Threading, however, our performance skyrocketed to a minimum of 112fps and an average of 131fps. That’s a HUGE performance difference. Modder Kaldaien claimed that the game uses multi-threaded rendering with 16 queues, something that could explain the performance gains when we enabled Hyper Threading on our simulated quad-core system. As such, we strongly suggest enabling Hyper Threading for this particular title.
At 1920×1080 on Max settings (and with an unlocked framerate), we saw our GTX980Ti offering performance similar to that of the AMD Radeon RX Vega 64. We’ve said it before and we’ll say it again; AMD needs to step up its game as its DX11 drivers are not that efficient. At 2560×1440, our AMD Radeon RX Vega 64 offered better performance than the GTX980Ti, though both of them were able to run the game with 60fps. And at 4K, the only GPU that was able to offer a smooth experience was our NVIDIA GeForce RTX2080Ti.
Graphics wise, Sekiro: Shadows Die Twice looks better than Dark Souls 3 but did not really impress us. There are some cool scenes (like the grass scene in the intro level) and there is limited environmental destruction, however the game’s visuals appear washed out. We are pretty sure that Reshade will do wonders to this game, however the vanilla version is not that great. Sekiro looks better than Dark Souls 3, however it isn’t close to the visual standards that have been set by other current-gen triple-A third-person games.
All in all, Sekiro: Shadows Die Twice performs well on the PC, provided you have a modern-day quad-core CPU that supports Hyper Threading. Those with weaker CPUs will definitely encounter performance issues. Furthermore, there is no official support for 21:9, there is no VSync in-game option and there isn’t support for Borderless window. The mouse movement also seemed/felt a bit weird. On the other hand, the game does not require a high-end GPU for 60fps at 1080p on Max settings and it displays proper keyboard and mouse on-screen prompts. What this ultimately means is that while Sekiro is not one of the most optimized PC games we’ve seen in 2019, it’s way, way, way better than Dark Souls 3, Nioh or most of KOEI Tecmo’s latest releases!
John is the founder and Editor in Chief at DSOGaming. He is a PC gaming fan and highly supports the modding and indie communities. Before creating DSOGaming, John worked on numerous gaming websites. While he is a die-hard PC gamer, his gaming roots can be found on consoles. John loved – and still does – the 16-bit consoles, and considers SNES to be one of the best consoles. Still, the PC platform won him over consoles. That was mainly due to 3DFX and its iconic dedicated 3D accelerator graphics card, Voodoo 2. John has also written a higher degree thesis on the “The Evolution of PC graphics cards.”
Contact: Email
pirate it, its a meme game
What? the game is really good!what are you talking about?
Only 4.2Ghz??? On an arch over 5 years old??! DDR3 at 2133Ghz???!!!
Yeah, their PC is really old
The data is used to see if there’s a difference between core and thread count, and what the engine can actually make use of. Everyone knows that higher clock speeds and Gen mean more FPS so it’s really useless data showing it, especially as we’re moving further away from GHZ as games are being played at 2k & 4K where the CPU becomes useless.
Didnt you look at the charts? The game is completely CPU bound at 1440p. They desperately need an update and there’s no getting around that.
How is it CPU bound? Can you please explain? The cpu limitations only come into play when gaming at 1080p which pushes most of the work load between the CPU and GPU, the higher you go the less and less it matters, we’re talking single digit differences which is useless data and that isn’t what they’re trying to show.
A game is considered CPU bound when you lower resolution and the framerate no longer increases. There is no difference (margin of error) between 1080p and 1440p for the higher end GPUs, so the game is completely bound by the CPU’s performance.
You are correct that traditionally, CPU bottlenecks only show up at 1080p, but with the latest Turing cards, this is definitely no longer the case. There are plenty games in my own PC with 1080Ti & 8700K where even the 8700K can’t keep up at 1440p. Most of the time, the game engine, or Denuvo, is to blame for this.
I wish DSOG had a 8700K or 9900K to do GPU testing, that would boost the framerate by at least 30% compared to their current i7 4930K.
If you have any questions feel free to ask 🙂
Seems like mods are fairly easy to make for this too. Something I hoped would be the case way back when it was announced.
My i5 6500, Aorus GTX 1060 6GB OC, 16GB DDR4 2133mhz can play this game at locked 60fps
At 1920×1080 on Max settings.
just few drop in cutscene but rarely.
This game is well optimized and the Graphic are Awesome.
i5 6600k @4.3 RX480 8gb and 16GB ram @3200. Max settings 1080p 60fps without drops. The game run fine with 4 cores without threads if you have a modern CPU.
Of course it does. But since john has no idea about what he’s talking about nor hardware to test this stuff, he uses his half decade old cpu for every single god damn game and issues statements about what sort of cpu you need or dont need in order to play game X. All that from his one cpu. Its mind boggling how tech illiterate the guy is and he keeps trying to sound as if he makes tech analysis of games, jesus
It’s far better than it, but far far harder. It’s pretty much Feudal Japan-Soul with just as bleak of a game world!
No idea what the f*ck the person who called it a “Meme game” meant.
so this is the hardest souls game?
No at least it’s not harder than DS1, but you have to get used to controls specially when everyones likes to block your attacks and specially bosses. also no rolling, only dodge and you will get hit while dodging so no iframe either and add jump to the mix, it becomes confusing at start.
First DS was punishing, even after death you could have carried the curse. but yeah, they all like that but difficulity is one side, the great level design in their games is what makes them fresh.
The basic combat mechanics are more complex, you have different ways of avoiding attacks that are specific to each type of attack, you can’t just dodge everything for the most part.
IMO the bosses require more skill than Dark Souls, and Sekiro doesn’t have summons or grinding for stats to make the combat easier, so you have less ways to cheese difficult fights.
Outside of bosses and minibosses, it’s more of a mixed bag. Some fights are easier, some are harder but it’s overall much easier to escape fights and the introduction of stealth gives you ways to take out a lot of enemies without even having to fight.
Overall I’d say it has a wider variance in difficulty than Dark Souls, with the peaks being noticeably higher.
It’s not “far better” than Dark Souls, that’s just stupid. It’s a different game, and whether it’s harder depends if you find its more reflex-oriented combat harder. They have implemented a lot of cool ideas, the controls are tight, and the stealth works, but all the variety and depth that were part for the course with the RPG elements of Souls are gone. Apples and oranges, but it’s yet another great game from From ^^
graphics look pretty disappointing for a 2019 title. And what’s up with the 60fps locked frame rate? This is 2019 for heavens sake! Also, I hope mods will roll out soon to significantly improve the textures and add some shaders.
Dark souls looked 500x worst when it come out…… .
It’s all about the gameplay buddy 😉
Yeah you can say that about Pong too
Haha, Got ‘Eeeeem… I tell ya, publishers/developers have the best PR teams for free via these gamers that just make constant excuses for them.
New video from Overlord Gaming.
https://www.youtube.com/watch?v=Jt_B1kat1nQ
In other news Game doesnt use Denuvo. Cracked on Day one and yet it’s the highest sold game, year to date on Steam. On March 22nd there was 93k people playing this game on steam. I’ve never seen numbers that high ever. The highest I’ve seen was RE2: Remake and that was 68k people playing at once and a that game uses Denuvo. So what does Sekiro NOT using Denuvo, cracked at Day 1 and is the highest sold game on Steam (year to date) tells us?… ?
Overlord Gaming has a new video on Youtube. I think you’ll be interested 🙂
Sweet, love those guys. Thanks for the heads up, gonna go check it out now. ?
It doesn’t tell us anything actually.
Scientifically, to know if it was related you would have to launch the game with Denuvo in a separate universe to even compare them.
So, yeah, aside from making a wild assumption of correlation it doesn’t say anything aside from people wanted to play it.
The average person doesn’t give AF about Denuvo or the inclusion of it.
To me, this simply states that From makes excellent games, and people are excited to finally be able to play a real game after all the fast food that’s been pushed lately.
It also shows that if you make a good game, people will buy it. Anti piracy software doesn’t help with anything.
I agree with the first half of your statement.
The second half, isn’t true. People will steal if they can. Trust me, I work in the public. I know this for a fact. A lot of people just suck and will only buy if that is their only option, especially if there are no consequences.
Please don’t make stuff up, also piracy is not stealing. Trust me i work in super public.
Thank you for saving me the time to reply exactly that. 😉
That’s only partly true. Sure there will always be those people. But plenty games without DRM have proven that if a game is good, it will sell, despite the piracy.
The point is you shouldn’t ruin the experience for everyone with Denuvo because of a handful of people that pirate.
I would agree with you, but I’m yet to have Denuvo ruin anything for me in any way, nor do I know anyone personally who has either. If it wasn’t talked about online, I would never even know it was implemented.
Overlord Gaming
sub to his Youtube channel
Not of a fan of that channel. I’ve come across it before and it is obviosly highly opinionated and absolutely full of bias. That kind of content is only good for fluffing ones own pillow.
I prefer to look at everything from a more analytical standpoint, without cherry picking what fits what I want to believe.
He presents data, can you show any case where his opinions arent factual?
No, because it’s against this guy’s narrative. he is a well known shill both for epic and denuvo.
There are plenty. He leaves out data that doesn’t pad his opinion. It is why I’m not a fan. Either present the whole story, or nothing at all.
I’m fine when people disagree with me, or show data that I didn’t expect, but when it is incomplete it is bothersome.
What data, please give me an example
Cause afaik, he always shows footage to back up his claims. He is opinionated sure but contrary to many others he takes the time and effort to prove his statements.
Of course you’re not a fan of that channel because you’re one of the most disgenious person I’ve seen commenting on this site. I’ve sat and watched all your replies on this topic and I wouldn’t even bother arguing with you because something is severely wrong with the Schitt that comes out of your brain. You are so disingenuous.
You don’t seem to grasp what disingenuous means as it seems to directly contradict what you’re actually trying to falsely claim.
You don’t seem to grasp you’re the one making false claims here. I’m still waiting.
hard games sell on pc.
that it would’ve sold even higher if it hasn’t been cracked day 1?
part of the reason why I bought at full price.
With my old but still kicking FX8350 4.2Ghz, 16GB RAM 1866Mhz, Asus Strix GTX970 OC, @1080p all settings maxed, rock solid 60FPS.
It runs a little bit better than Dark Souls 3 for me but I don’t think it looks better or has the complexity. Physics destruction and ragdoll physics are more simple than in Dark Souls.
The only thing that is better is the lighting and probably the character design, yet no customisation at all in that part.
The game has actual shadows now, not just baked in shadow maps. That’s a big plus.
They also baked in a really nice AA implementation and proper motion blur. I’m digging it.
Has anyone figured out how to use ds4windows with this game? my only xbox360 controller has a dead bumper and the ps4 controller is just much better anyway.
The PS4 controller works with this game. Go insteam Steam, Big picture, and turn on controller support for the DS4. Should work fine.
Works perfectly now after fps unlock.. also works with fullscreen after you unlock refresh rate with another mod, fullscreen forces 60hz by default.
can’t really agree with the hyperthreading
Using an old i5 2500k so no hyperthreading and not even manual oc at 3.7ghz and the game runs smooth at a rock solid 60 fps without any frame issues whatsoever at 1080p, using a 6gb 1060 with a half year old driver.
I don’t agree. The visuals are really nice on my PC. Detailed and clean visuals with great art direction and mostly free of things like pop in, jaggies and noise. It looks better than most similar games such as Dark Souls 3, Bloodborne, Nioh, and even Devil May Cry 5. So I don’t know what he’s comparing it to. Devil May Cry 5 is the most recent somewhat comparable action game and aside from its nice character models in cutscenes I find most of the environments somewhat bland compared to Sekiro.