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

Civilization VII Performance Benchmark

W1zzard

Administrator
Staff member
Joined
May 14, 2004
Messages
28,656 (3.74/day)
Processor Ryzen 7 5700X
Memory 48 GB
Video Card(s) RTX 4080
Storage 2x HDD RAID 1, 3x M.2 NVMe
Display(s) 30" 2560x1600 + 19" 1280x1024
Software Windows 10 64-bit
Civilization VII is new fodder for lovers of the turn-based strategy genre. In our testing the game ran very well on virtually all graphics cards. In our performance review, we'll look at the game's graphics quality, VRAM consumption, and how it runs across a range of contemporary graphics cards.

Show full review
 
5080 beats 4090 at all three resolutions.
First game title in TPU benchmarks to showcase that (excluding boosted AIB versions).

Begs the question: considering the performance spread between AIB variants of each model, perhaps it makes sense to disclose which manufacturer was benched in these tests?

In this specific example, we could conclude RTX 50 works better in CPU-sensitive titles. But if that's an ROG Astral 5080 or a water-cooled variant, that may be a false conclusion.
 
Last edited:
So this game is essentially a CPU benchmark.
 
Don't get the point of this,
Civ games are usually looking very nice for their respective time, but never been a heavy hitter on the gfx requirements.
In the other hand, there is no CPU bench, which would be interesting to see some Ryzen X vs Ryzen X3D.... aaand intel :D

So this game is essentially a CPU benchmark.
You beat me by seconds :toast:
 
Lock to 120/144 FPS (or even 90 FPS), save power, enjoy the game, regardless of GPU.
 
@Cheeseball
I mean, yeah, that only makes sense. You aren’t really required to pull off sick clutch headshots in a game of Civilization. There is no reasonable, uh, reason to go about 120. As in the vast majority of games that aren’t high precision e-sports titles.
 
perhaps it makes sense to disclose which manufacturer was benched in these tests?
All relevant cards are FE. I think the differences are due to CPU usage, because it's pretty CPU limited at 1080p
 
Watered down mechanics, watered down controls, no Nvidia DLSS support, ah yes another game screwed over by providing parallel support to console plebs. It's The Elder Scrolls effect.
 
Watered down mechanics, watered down controls, no Nvidia RTX support
Brother, this was the trajectory Civ was on long before it went to consoles. Did you memory-hole launch Civ V? Are we pretending it wasn’t a step down from IV in every way? And why, in the name of ever-loving fuck, would you want RT in a goddamn 4X turn-based strategy game?
 
Any chance for a CPU benchmark? I know it's much more intensive to swap CPUs in and out of sockets as opposed to just replacing the GPU, but this would be a very interesting game to get see CPU comparisons in. 9800X3D has an obvious cap of ~338fps, but this is also a game I imagine the 3D cache is very influential in. Even just a bench of the 7700X, Core Ultra 7, maybe the 12700K, and then 5800X would be helpful to compare the different architectures. And maybe throw in the 12100 to see how it performs on a 4-core CPU.

Would be extremely helpful to see if the game got noticeable stutters or slowed down in the endgame on some CPUs.
 
why, in the name of ever-loving fuck, would you want RT in a goddamn 4X turn-based strategy game?
Y'know, it's better to trace a ray and not need it than to need to trace a ray and not do it.

And maybe throw in the 12100 to see how it performs on a 4-core CPU.
And i5-10400/Ryzen 5 3600 to see if old yet still popular 6-core CPUs are fit enough.
 
I heard there was a dev named Luigi that worked on Civ 6 and fixed a lot of UI bugs. Maybe they could bring him back.
 
Am I the only one who prefers the low graphics setting over the high, based on the screenshots? The low settings seem like they make the game easier to play. Buildings and units are larger and less cluttered so you have an easier time seeing what they are at a glance, and everything is just brighter (I guess from less shadows)? The only thing that's worse is less trees.
 
All relevant cards are FE. I think the differences are due to CPU usage, because it's pretty CPU limited at 1080p
It's still very surprising to see the 5080 beat the 4090 every time... I wonder if it's a Driver issue or Game Optimization issue.
 
What's wrong with 4090's results? It's at least 10% behind where it was supposed to be. Other 4000 series cards are fine but 4090 looks like an outlier. Did it have low usage or something? @W1zzard
Thanks for the review.
 
Watered down mechanics, watered down controls, no Nvidia DLSS support, ah yes another game screwed over by providing parallel support to console plebs. It's The Elder Scrolls effect.
I laugh at your comment, what do you want a turn based 500FPS game? Seriously. It’s not a peen measuring contest, it’s a turn based human game.

If it’s a miss for mechanics sure, bringing DLSS and FSR into a game that isn’t focused on graphics but the human aspect is as far from the point as you can get. Go touch some grass.
 
Watered down mechanics, watered down controls, no Nvidia DLSS support, ah yes another game screwed over by providing parallel support to console plebs. It's The Elder Scrolls effect.

No DLSS Support is not a problem since the game can use MSAA 4x which is better than DLSS since it has to use TAA by default, whereas MSAA works with the Native image (with Aliasing).
 
The 4090 looks uncharacteristically weak here. Driver or game bug perhaps? Seems almost like it isn't entering a higher power state since it's so CPU bound.

AMD looks weak across the board, game ready driver may claw a bit of perf. back?
 
I want this game to be good so bad.
 
What's wrong with 4090's results? It's at least 10% behind where it was supposed to be. Other 4000 series cards are fine but 4090 looks like an outlier. Did it have low usage or something? @W1zzard
Thanks for the review.

Maybe the game engine likes using INT32 calculations, and since Blackwell expands INT32 ability to every CUDA core while only half of the cores in Ada have INT32, that explains it? IDK, I'm just shooting in the dark.
 
Maybe the game engine likes using INT32 calculations, and since Blackwell expands INT32 ability to every CUDA core while only half of the cores in Ada have INT32, that explains it? IDK, I'm just shooting in the dark.
I'd understand if it was between 5000-4000 series but 4090 is particularly bad. It only 18% faster than 4080S which is 10% lower than 28% average in W1zzard's latest reviews. I took time to check, this is THE lowest gap in performance between 4080S-4090 (again based on w1z's latest test). Interestingly 5090 is also exactly 60% faster than 4090 which is THE biggest gap in the reviews. So this the worst game for 4090 in against both 4080 Super and 5090. Which I think is really weird considering it does not look like simple problem like CPU bottleneck or something.
If if was just calculations, surely 60% more CUDA on 4090 would help it more against 4080S. It could explain 5000s' performance but not 4000s'.

I mean it's not really that important. GPU performance in this game is completely pointless anyway. Only metric that matters is turn times. I just found it interesting.
 
Last edited:
@Cheeseball
I mean, yeah, that only makes sense. You aren’t really required to pull off sick clutch headshots in a game of Civilization. There is no reasonable, uh, reason to go about 120. As in the vast majority of games that aren’t high precision e-sports titles.
I'm currently on turn 270 on extended and my PC is using a total of 190W locked at 90 FPS on a monitor at 4K 240Hz. If I let it run without a FPS cap, the GPU is scratching at 120 FPS and my total system draw is like 450W+.
 
Am I the only one who prefers the low graphics setting over the high, based on the screenshots? The low settings seem like they make the game easier to play. Buildings and units are larger and less cluttered so you have an easier time seeing what they are at a glance, and everything is just brighter (I guess from less shadows)? The only thing that's worse is less trees.
Yes, the low-settings version is much cleaner. I don't know what you can set up individually. In Hearts of Iron IV I play with shadows turned off because I like it better.
 
Well on 4k if turn down a few settings a little bit can run on 8GB vram, that seems decent at least.
 
Back
Top