[HN Gopher] Cities Skylines 2 runs with 20fps on an Nvidia RTX4090
___________________________________________________________________
Cities Skylines 2 runs with 20fps on an Nvidia RTX4090
Author : segasaturn
Score : 155 points
Date : 2023-10-26 18:08 UTC (4 hours ago)
(HTM) web link (www.dsogaming.com)
(TXT) w3m dump (www.dsogaming.com)
| hotstickyballs wrote:
| I didn't see any mention of cpu used on the article. I'm guessing
| that might be important.
| belinder wrote:
| I think the CPU will play more of a part when there is a lot to
| simulate. But in the article they said this was on an empty map
| mrob wrote:
| There are CPU/GPU meters on the screenshots. Highest CPU load I
| saw is 78% on one core and lower on other cores, but GPU is
| always 100% or close to it.
| apocalyptic0n3 wrote:
| I'm getting ~60fps on my 4090 at 1440p. Haven't had any framerate
| issues. My GPU has been pinned at 100% and my temps are sky high,
| but the game is smooth. They definitely need to fix this, but
| it's not as dire for everyone as it's being made out to be.
| transcriptase wrote:
| Considering the game only runs adequately for you on... the
| best video card for gaming that money can buy...
|
| You can see why the 99% of customers who have much less
| powerful hardware might consider the situation dire.
| dekhn wrote:
| The entire history of video cards has been "games that were
| hard on graphics cards"... leading to... a new generation of
| graphics cards that run the game with ease... leading to... a
| new generation of games. It's an ongoing battle.
| transcriptase wrote:
| There are games that are hard on graphics cards.
|
| Then there are okay-looking city building games that pin
| the best GPU that exists at 100% usage at 1440.
|
| A card that's meant for playing AAA games at 4K Ultra.
|
| Maybe a dev accidentally bundled a crypto miner they were
| running on company hardware after hours?
| wjnc wrote:
| What is new is that 80% of current gamers is playing on a
| GPU on par the with the highest end model of about 7 years
| ago. Moore's law might not be dead, but at least is
| economically forgotten. So older games might have gotten
| over the performance bump in a year if two, now it's more
| like seven. Paradox failed to correct for economics in this
| party.
|
| (My 1060 6GB is still above average on userbenchmark.com.
| That a 2016 card and has never been top of the line.)
| KeplerBoy wrote:
| That was before Moore's law was dead. Now devs need to
| write code that is at least not terrible. (Insert link to
| casey muratori ranting about modern software)
| laverya wrote:
| Yeah, but generally those games that push for new GPUs look
| spectacular, not like SimCity 5 from over a decade ago.
| Stevvo wrote:
| This is not a new generation game, the graphics are at-
| least 10 years behind the rest of rest the industry. (No
| PBR or other modern rendering techniques are used).
| dekhn wrote:
| I didn't make my comment about this specific game.
|
| I made it about Cyberpunk 2077, Crysis, and Doom 3.
| fkyoureadthedoc wrote:
| > 99%
|
| I don't think that's accurate. There's tons of complaints and
| legit issues with performance for sure, but I don't think
| making up numbers (likely grossly exaggerated) is helpful.
|
| https://www.reddit.com/r/CitiesSkylines/comments/17fsdx9/hav.
| ..
| mvdtnz wrote:
| In fact the proportion of gamers with graphics cards less
| powerful than a 4090 is greater than 99%.
|
| https://store.steampowered.com/hwsurvey/videocard/
| stouset wrote:
| 3080 here. The game runs _completely fine_ at 3440x1400 with
| max settings.
| patates wrote:
| > I'm getting ~60fps on my 4090 at 1440p
|
| > it's not as dire for everyone as it's being made out to be
|
| I don't have the game but your first point makes it sound
| rather dire, especially for a city-building game, does it not?
| apocalyptic0n3 wrote:
| I could have worded it better. Basically, we've been
| inundated with a bunch of pre-release players saying the game
| was basically unplayable on any GPU.
|
| Post-launch, it's perfectly playable for a lot of people.
| Even on older cards. Performance is by no means good, but
| it's not a slideshow like everyone was worried it would be
| pre-release.
|
| (fwiw I also tested this on a 2070. It wasn't great at 1440p
| but was performing just as well as my 4090 on 1080p)
| Mistletoe wrote:
| "Let them eat cake!" -apocalyptic0n3
| transcriptase wrote:
| "I don't know why people complain about potholes. I drive a
| $180,000 Mercedes G Wagon and barely feel them."
| merb wrote:
| if they don't have bread, they should eat cakes instead.
| bakugo wrote:
| >I'm getting 60fps on a $2000 card, that means it's not that
| bad!
|
| I hate modern AAA PC gaming so much.
| aranelsurion wrote:
| and on 1440p. You'd assume 4K would become the expectation,
| at least for the top end. It's been a decade.
| scubadude wrote:
| To be fair this is AUD$70 not AUD$110-120 like some recent
| actual AAA games. (They probably plan to add a buttload of
| DLC though ;)
| nullindividual wrote:
| There have been so many Unity games which do not manage to
| throttle GPU usage. Battletech was another one where at the
| menu it would keep the GPU at 100%.
|
| Not sure why devs allow this, but with Battletech it was a
| known issue that was never fully resolved.
| Geee wrote:
| You'll have to enable v-sync to limit framerate. I enable it
| always (or fps limit) because I don't like blasting my GPU at
| 100% which makes it hot and loud.
| harrid wrote:
| 60fps on a 4090 _are_ framerate issues
| piperswe wrote:
| I'm running 4K, 30fps on a 4080 on medium settings (with a decent
| size city). It's not good performance, but it's plenty for a city
| builder.
| nottorp wrote:
| 20 fps, 60 fps, whatever
|
| When was the last time you could enjoy a 4x or city builder on a
| laptop, with integrated graphics, on battery?
| test77777 wrote:
| Yeah it's been proven nobody can actually see the difference in
| anything over -25 frames.
| methodical wrote:
| Ignoring all of the scientific evidence otherwise, there is
| not a chance you're actually parroting this nonsense without
| actually noticing a rather substantial difference when
| viewing something (say a movie) at 30fps versus 60+fps. It's
| literally night and day. Me and a few coworkers noticed the
| difference between 60 and 120 when upgrading phones recently,
| and that difference is much more nuanced than the difference
| between 30 and 60.
| goatking wrote:
| There is a HUGE difference between 30fps and 60fps games.
| Very noticable. I got used to 60fps so much, that I don't
| even touch 30fps anymore.
|
| It's true that your eyes(or brain?) can get used to the 30fps
| after some time. But if you compare them side by side...
| ElectricalUnion wrote:
| Assuming you can generate 24fps of "movie quality perfect
| images", but you can't.
|
| Let's take motion blur as an example: It's a very expensive
| to do it as a real-time VFX - so we cheat and approximate it
| - that doesn't that look good most of the time - because we
| often have to cut too many corners and end up with a poor
| approximation.
|
| You're generating very slow images that a computer believes
| is a facsimile of reality that, unfortunately, don't
| represent motion blur very well, so they might be "24fps",
| but they look wrong compared to reality, and you probably
| can't conscientiously tell what is wrong, just that something
| seems to be sightly off.
|
| So it's easier instead to attack the problem from the other
| side, and generate (or frame generate, that's a thing now)
| and show 480fps "crisp images" and make human persistence of
| vision do actual motion blur for you.
| jandrese wrote:
| Even when viewing a movie instead of playing a game there is
| a huge difference between the blurry mess that fills the
| screen every time there is motion at 24 FPS vs. the sharper
| details of higher framerates.
|
| The crazy thing is people are actually trained to hate it.
| When you see smooth motion your brain goes "oh no, this is
| some cheaply shot camcorder shit, not a quality movie" even
| though what is really happening is the picture looks better
| than you expect.
| kimixa wrote:
| Right now.
|
| You just can't be obsessed with the world "Ultra".
| Arrath wrote:
| Right now? Just load up OpenTTD and have a ball.
| Kiro wrote:
| OpenTTD is not a city builder.
| crop_rotation wrote:
| OpenTTD graphics are just not even remotely comparable. It
| looked so ugly when I tried it that I couldn't bother to
| continue for more than 2 minutes. If graphics don't matter
| then maybe even drop openttd and play a text based graphic
| game. It will run smooth even on a raspberry pi.
| olyjohn wrote:
| The game is originally from 1994, what do you expect? We
| didn't even have 3dfx 3d accelerator cards back then.
| OpenTTD will run smooth on a Pi. It ran smooth on my 486
| when it first came out.
| crop_rotation wrote:
| I don't expect anything, I am just responding to someone
| comparing it to cities skylines ii smoothness.
| segasaturn wrote:
| I personally greatly prefer the 2D, handcrafted, pixel art
| look of games like OpenTTD, SimCity 2000 and Factorio over
| 3D sim games like CS.
| mcpackieh wrote:
| The graphics may be retro but the gameplay is lit. It's one
| of the most relaxing and satisfying games around.
| olyjohn wrote:
| I'm not a huge gamer, but I did spend 14 hours a day for a
| week playing this game once. Never been sucked in to a game
| like this. And even better if you can find the original music
| and graphics. The music is excellent, and will play in your
| dreams.
|
| First time I played Transport Tycoon was on my 486, I got a
| demo of the original game on a floppy disk out of some PC
| game magazine. Looking back, it's amazing how much game play
| fit onto a 1.44MB floppy disk. I ended up purchasing the
| Deluxe version on a CDROM. I ripped the CD and have kept an
| image of it ever since, which came in handy when I found
| OpenTTD and wanted to use the original graphics and content.
| Chris Sawyer and the OpenTTD devs are legendary in my mind.
| filterfiber wrote:
| Not quite a city building but I think possibly comparable -
| factorio you can at least get to endgame on pretty low end
| hardware with decent fps. It even runs on the switch.
|
| I say comparable because of the number of things moving around
| on belts, machines, etc. I think would be a similar workload.
| Although it is 2D...
| jabroni_salad wrote:
| I play stellaris on next to nothing and it runs like a champ,
| even when I zoom in to enjoy the spaceship explosions.
| jandrese wrote:
| I run Stellaris on an aging i5-3750K and a somewhat newer
| Geforce GTX 1070Ti at 1080p and also think that the
| complaints about performace are overblown. I even leave the
| planet generation settings turned up fairly high (I know a
| lot of people online reduce them to like 0.25 for performance
| reasons). That's not to say I never have slowdowns, enormous
| doomstack on doomstack fleet battles in the endgame hurt the
| FPS, but they are fairly uncommon. Fighting the 4x strength
| crisis fleets are the one other time that the game tends to
| chug, but those tend to be fairly late as well.
|
| That said, I don't have a ton of expansions installed and no
| mods. I have a strong suspicion that some mods are harder on
| performance than others, especially the over-the-top ones
| like Gigastructural Engineering.
| mschuster91 wrote:
| I used to play a game called Mobility back in ye day, early
| 00s, it was city building with a scientific background - IIRC
| it came out of a funding grant of the German government.
|
| Gotta look if I have an ISO flying around somewhere.
| elil17 wrote:
| I feel like this is not only a massive technical mess-up but also
| belies a deep misunderstanding of their customers. Most C:S fans
| were more excited for a more realistic traffic AI and mixed use
| zoning. Yet this experience was made unplayable by an attempt to
| make it look prettier.
| sp332 wrote:
| Why would someone crank all the settings up and then complain
| that it runs slowly? Just put the settings where you want them.
| nickthegreek wrote:
| Because they spent $2k on the GPU alone to do such things.
| sp332 wrote:
| The cards have a certain level of performance. They can
| change the settings of the game to match what they bought.
| If they have a $2,000 card, it's going to look good.
| raydev wrote:
| > If they have a $2,000 card, it's going to look good.
|
| But it doesn't look good on the $2000 cards. That's a big
| problem.
| legitster wrote:
| I mean, how playable is the game on minimum graphics settings?
| As you said, I don't care if the game is prettier.
| elil17 wrote:
| For me - not at all. I don't have a computer that meets their
| minimum system requirements. I really feel like the cycle of
| newer games requiring more hardware is pointless and
| wasteful. I wish publishers would focus on good game design
| and make their games accessible to people without high-
| powered gaming hardware.
| sp332 wrote:
| It takes a lot of time and effort to pare down the assets
| and add tricks to make a game look good on lower hardware.
| Targeting current-gen GPUs means the devs get a _better_
| game out in the same time and budget.
| swatcoder wrote:
| > belies a deep misunderstanding of their customers. Most C:S
| fans were more excited for a more realistic traffic AI and
| mixed use zoning. Yet this experience was made unplayable by an
| attempt to make it look prettier.
|
| That's not fair. This is a new engine that they probably expect
| to support for maybe as long as 10 or 15 years. As a AAA
| publisher, Paradox doesn't get to stylize behind indie-style
| aesthetics and needs to keep up with where their peers are
| headed. It's not aiming to be prettier just for the heck of it,
| but because it needs to maintain a certain mark to keep the
| franchise relevant.
|
| Knowing Paradox, more rich gameplay enhancements probably are
| on the update and DLC agenda, and we can assume that their
| designers really care about that kind of stuff. But for AAA
| publishers like them, there are also other factor that matter
| and that may need to take priority.
|
| That said, what a f'd up and premature launch!
| Miraste wrote:
| I don't think that's true at all. Nothing else Paradox
| publishes has AAA graphics, and that's not the target
| audience for their games. The first Cities Skylines didn't
| beat Sim City because it had better graphics. It was because
| of gameplay. Since EA gave up on city sims, they don't even
| have peers, they're only competing with themselves.
| staplers wrote:
| That's not fair.
|
| There is something taught to children about this idea..
| mvdtnz wrote:
| It's not a new engine. They continue to use Unity.
| WillPostForFood wrote:
| As a C:S fan, I want realistic traffic, mixed zoning, and
| expect it to be prettier. I don't find it unplayable at all
| even when it is turning my PC into a space heater.
| Optimizations will come, and I'd rather be waiting for
| optimizations than features.
| tuyiown wrote:
| > Optimizations will come
|
| As someone that has played C:S 8 years ago and last year,
| I'll tell you, not really, at least, not from people like
| paradox.
| labster wrote:
| Definitely not from Paradox. Europa Universalis IV is 10
| years old now, still getting regular DLC releases, and late
| game is still unacceptably laggy on 2023 computers. Like,
| 20 to 0.2 fps.
| brucethemoose2 wrote:
| Mid-late game Stellaris still chugs on my 7800X3D, which
| is massively faster than pretty much any non X3D CPU in
| this particular game.
|
| And I have it overclocked, with CL30 RAM, running a
| medium galaxy. Thats kind of unacceptable.
|
| They fix it some, then it regresses...
| labster wrote:
| DLC feature bloat >>> incentive to optimize
| brucethemoose2 wrote:
| Funny thing is Stellaris has a "custodian" dev team
| specifically to address this.
|
| The custodians fix a bunch of bugs and performance
| issues... then the features/dlc team introduce new ones.
| WillPostForFood wrote:
| Since my post a performance patch has been released!
| _flux wrote:
| The site timeouts on me :/.
|
| One guy at Reddit noticed with the nvidia nsight that apparently
| the cims are drawn without regard to level of detail:
| https://www.reddit.com/r/CitiesSkylines/comments/17gfq13/the...
|
| Vehicles are handled properly, though.
|
| If the analysis is correct (among some other points risen
| regarding shadows), then it seem rather a big oversight with
| fortunately some easy fixes.
| mvdtnz wrote:
| This is armchair game developer sleuthing and I don't buy it
| for a dollar. Mainly because performance issues with this game
| occur even on an empty map. There is absolutely no way the game
| is rendering 10,000+ characters at that level of detail.
|
| You'll be wise to ignore Redditors when they claim to be making
| some kind of breakthrough that eluded professional game
| developers over years of development.
| tyree731 wrote:
| Except if it's this article - https://nee.lv/2021/02/28/How-
| I-cut-GTA-Online-loading-times...
| DevX101 wrote:
| A lot of people who play Skylines ARE developers.
| mvdtnz wrote:
| A lot of people on Reddit are not.
| Aurornis wrote:
| I checked the IGN review for comparison. IGN reviewer also has an
| RTX 4090 but plays with a lower 1440p resolution:
|
| > I have a 13900k, 64GB of RAM, and a RTX 4090, playing on a
| 1440p ultrawide monitor. I got 35fps at the main menu and in game
| on a brand new map w/o building a single thing. Turning off
| motion blur and depth field increased this from 35 to 50fps. Not
| a single other graphics setting changed the performance at all. I
| turned off every single setting I could or set it to the lowest
| possible, and still only got 50fps.
|
| How can a game even get to this point? Everyone in the
| development process must have noticed that it was running poorly.
| Even if all the developers had the most expensive consumer GPU
| available, they still would have seen poor performance.
| brucethemoose2 wrote:
| > Turning off motion blur and depth field increased this from
| 35 to 50fps
|
| Whats more, such simple postprocessing filters should not tank
| FPS so dramatically, even with a severe CPU bottleneck.
|
| That alone is not just unoptimized, its a severe bug.
| mvdtnz wrote:
| And even if such filters are somehow so intensive, they
| should be turned off by default. I am playing Cities Skylines
| 2 and there is so much unnecessary eye candy that is turned
| on even with a mid-range graphics card. The game runs fine
| for me at 1440p when I turn off the really intensive post
| processing.
| fourteenfour wrote:
| They screwed up the defaults. When I first loaded it was
| set to a resolution with a 24Hz refresh rate. Also the game
| looks bad with the default SMAA, it looks great if you
| change the advanced graphics settings to use TAA. I'm on a
| 3080 at 4K and after fiddling with settings it looks
| wonderful and is very playable. Unbelievable that they
| hosed the first impressions so much.
| brucethemoose2 wrote:
| > Also the game looks bad with the default SMAA, it looks
| great if you change the advanced graphics settings to use
| TAA
|
| Well thats very subjective, as I usually can't stand TAA
| and like SMAA.
| WXLCKNO wrote:
| Being set to a default of 24hz sounds like the idea of
| someone forced to ship the game, hoping users wouldn't
| notice.
| brucethemoose2 wrote:
| Its probably just detecting a 24p TV resolution in
| fullscreen, if I were to guess.
| babypuncher wrote:
| But no TV maxes out at 24hz, that is a lower bound meant
| to support things like blu-ray players for judder-free
| movie playback.
|
| 60 hz has always been the bare minimum any display
| supports, so if any game is picking something below that
| then something has gone horribly wrong. But really, games
| should be using the current desktop refresh rate as the
| default, because you know it's supported and makes things
| like alt+tabbing considerably faster even in exclusive
| fullscreen.
| brucethemoose2 wrote:
| > even if such filters are somehow so intensive
|
| They should _never_ be so intense.
|
| Some shaders unintentionally scale nonlinearly with
| resolution, and I'm thinking that may be the case here.
| fyloraspit wrote:
| Severe CPU bottleneck normally implies that your CPU may need
| to be upgraded though? If you are running a 13900k, there
| isn't much room for that.
| epolanski wrote:
| Depth of field is often more nuanced than simply
| postprocessing filtering, it generally also swaps textures
| and models based on distance from camera.
| brucethemoose2 wrote:
| I didn't know this, interesting.
|
| I don't really like DoF anyway :P
| bluescrn wrote:
| More importantly, why the hell does a city-builder game need
| motion blur?
|
| It's usually a pretty obnoxious effect even in action games,
| the first thing to turn off if the option exists...
| 2023throwawayy wrote:
| Oh, they noticed. They complained. They said "we need to fix
| this".
|
| The project manager said "Not high priority for launch", then
| closed the ticket.
| ryanisnan wrote:
| Is this satire? If this is real, I'd love to know more.
| refulgentis wrote:
| It's satire, people are reaching way too far in this thread
| (ex. IGN's review is using 3080 / Ryzen 7 and they were
| happy with performance)
|
| The site none of us can read probably has more info about
| this being a RTX 4090 quirk. And it's not exactly
| surprising the dev team wasn't optimizing for RTX 4090, and
| the quirky reaction here would be justified if they were.
| IntelMiner wrote:
| How would a 4090 perform _worse_ than a 3090? It 's not
| like Nvidia completely rebuilt the GPU between revisions
| dragontamer wrote:
| > It's not like Nvidia completely rebuilt the GPU between
| revisions
|
| NVidia absolutely changes the assembly language rather
| dramatically between iterations. That's why the code is
| in NVidia PTX (or HLSL / Microsoft's DXIL, DirectX
| Intermediate Language), and then recompiled to actual GPU
| assembly each time you install a game or run the code on
| a different system.
| IntelMiner wrote:
| I have not seen an instance where a GPU series that has
| been out for over 12 months has _regressed_ in
| performance prior to an older one
|
| Perhaps back in the very early accelerator days when
| everyone was making GPU-specific hacks of Quake 1 but
| even that was smoothed over by the transition to
| DirectX/OpenGL/GLiDE at the time
| hrydgard wrote:
| It's probably just a very good guess, since that's how it
| always goes.
| tomatotomato37 wrote:
| If you notice how fast the PR and guides for optimizing
| performance has come I'd say not only the developers but
| everyone that's not the top of the top knew the game's pain
| points.
| epolanski wrote:
| I mean if you have time to make guides, you had time to
| implement some presets in the ui...
| epolanski wrote:
| More like c-suite, there's no chance PMs have anything to say
| in this and won't report issues upstairs just to be told "you
| can fix it after launch".
| legitster wrote:
| Is 50 FPS a huge issue?
|
| I play a lot of city builders and so long as the UI is still
| smooth I really don't notice any gameplay difference between
| 30fps and 60fps.
|
| I put 8 hours into Frostpunk before realizing I was limited to
| 30FPS.
| ajaimk wrote:
| 4090 is the fastest available graphics card. That's the
| problem.
| wlesieutre wrote:
| 20 fps is playable for a city builder, but that's on a $2000
| GPU
|
| A bit unfair since most people aren't gaming at 4K (Steam
| hardware survey puts it at 3.29%), but still
| washadjeffmad wrote:
| Ok, so around 4 million users. I wonder what the overlap of
| 4K owners and people who play city builders is?
| dimgl wrote:
| I feel like this is grasping at straws? Plenty of
| unorthodox games go mainstream. Can't expect consumer
| hardware to the stay the same forever. This game is just
| poorly optimized.
| washadjeffmad wrote:
| I think I'm giving people the opposite impression that I
| intended.
|
| Well-compensated people (like engineers) are more likely
| to own a 4K monitor and high-end GPU, so I meant to
| convey that there's probably a large overlap with the
| type of people who play city builders (like engineers),
| so they really should have accounted for that.
|
| It was only after I bought my 4K 120Hz OLED that I found
| out how popular they are for sims, including pinball.
| It's been awesome.
| janosdebugs wrote:
| Anecdotal evidence: I play at 4k if I can manage, and I
| sport a 2060S.
| wackget wrote:
| My personal opinion: yes. As 160+Hz monitors are finally
| becoming commonplace, it's becoming less defensible for games
| to run anywhere below 60fps.
|
| But it's the fact that these figures are being seen with the
| most powerful consumer graphics card (on a brand new
| unpopulated game map) that's the problem.
| legitster wrote:
| City Skylines 1 has been played and supported non-stop for
| over 8 years. Why should they limit the highest possible
| graphics settings only to existing technology?
|
| Again, this is a _simulation_ game. If I ask the game to
| spit out more detail and graphics than can be physically
| supported by today 's technology, I guess I don't
| necessarily see this as a design fault.
| kaibee wrote:
| If it were doing a ton of compute-shader simulation on
| GPU to actually increase the fidelity of the simulation
| that would be one thing, but having a badly implemented
| graphics pipeline isn't that.
| iforgotpassword wrote:
| Somewhat yes, I didn't try it yet, but a few things don't
| add up then:
|
| Why enable those heavy, performance crippling post
| processing filters by default?
|
| From what I can find in this comment section, and I guess
| this has to be taken with a grain of salt, performance
| doesn't scale. It seems we max out at 50fps on a beefy
| 13th gen Intel with the fastest GPU available, but then
| there's a couple comments with mid-range hardware where
| performance is nearly identical.
| danudey wrote:
| > If I ask the game to spit out more detail and graphics
| than can be physically supported by today's technology, I
| guess I don't necessarily see this as a design fault.
|
| To be clear: I had to turn the quality down to much worse
| than I recall Cities Skylines 1 looking in order to get
| worse performance than Cities Skylines 1.
|
| The major complaint here is not "I cannot max the
| settings on my mediocre hardware", but rather "I cannot
| mediocre the settings on my max hardware"; the
| performance is bad at any level, but my hardware, while
| not top-of-the-line, should be able to run a game like
| this at 4K _playable_ , or at 1080p while looking awesome
| and running great. Instead, I get 1080p with a mix of
| medium/low settings in order to get it looking pretty
| okay in most aspects.
| cmovq wrote:
| > Why should they limit the highest possible graphics
| settings only to existing technology?
|
| This assumes Cities Skylines 2 is using some next gen
| graphics technology when in practice it is a Unity game
| without any cutting edge graphics. Another comment
| mentioned they are doing 10k+ draw calls per frame - it's
| just poorly optimized.
| serf wrote:
| this is a trope that is spit out by the developers of
| every poorly optimized sim in recorded history.
|
| Yes, scalability towards the future is good -- but not at
| the detriment of player performance _now_.
|
| If the player base doesn't stick around during our
| current dark-times medieval technology stack, there won't
| be a player base when we have whatever future tech makes
| it playable -- see the problem?
| FirmwareBurner wrote:
| _> As 160+Hz monitors are finally becoming commonplace,
| it's becoming less defensible for games to run anywhere
| below 60fps._
|
| You're forgetting the increasing popularity of the Steam
| Deck where even 60 FPS is a luxury so game devs should
| definitely have that as a target.
| BlueTemplar wrote:
| CS2 devs reportedly had 30 FPS as a target. Which is
| actually fine for that kind of game, but it sounds like
| that is NOT for low-end hardware ?
| adgjlsfhk1 wrote:
| For CPU heavy games it totally makes sense to target
| 30fps, but that better be 30fps at 1080p with integrated
| graphics, (or 30fps 1440p with 1060 class gpu). Add eye
| candy if you want, but it really should run well on any
| remotely modern system.
| giobox wrote:
| > As 160+Hz monitors are finally becoming commonplace
|
| The Steam hardware survey suggests otherwise to my eyes;
| it's often easy to forget working in tech that _a lot_ of
| Steam customers have relatively shit hardware:
|
| https://store.steampowered.com/hwsurvey/Steam-Hardware-
| Softw...
|
| While they don't track monitor peak supported refresh, just
| looking at the rest of the results suggests to me high-
| refresh gaming (over 60FPS) is going to be relatively niche
| at best. Lots of PC gaming specs I take for-granted on my
| own builds or consider "average" are actually not that
| common.
|
| I absolutely agree 60fps should be considered the floor for
| most modern PC games running on "reasonable" hardware
| though, 30fps belongs to the past.
| Iulioh wrote:
| Off topic but 42% of VR headsets are Quest 2
|
| Woahhh
| aeyes wrote:
| Most Steam customers play games like Counter Strike,
| Dota, Team Fortress, PUBG, GTA, Call of Duty which have
| modest hardware requirements.
|
| I wouldn't read too much into these stats.
| giobox wrote:
| Right - you are proving my point... Most customers
| probably do not have 160fps monitors.
| stodor89 wrote:
| 50fps is okay. 50fps on a $1600 GPU is definitely not okay.
| bee_rider wrote:
| It is just a matter of preference, I don't see why people get
| bent out of shape about it.
|
| 30fps is fine for lots of games. Especially for stuff that
| doesn't require super twitchy gameplay.
|
| If somebody really cares about 60fps or 120fps, fine. But the
| developers have some effects they want to implement, and if
| some of them don't fit into 8.3ms, let those of us who don't
| mind 30fps enjoy them, right?
| tomatotomato37 wrote:
| It's not the average that's the problem, is that it has hard
| stutters from when you do something like swing or rotate the
| camera around and force a lot of new things to load. Also
| there's no fps limiter so I can't stop the game from
| attempting to match my monitor's 165 FPS with a overheating
| laptop 3060, causing the stuttering to be even more visible.
| KennyBlanken wrote:
| You can set a frame limit per-program in the NVIDIA control
| panel.
|
| Undervolting also reportedly works really well in laptop
| nvidia chipsets, FYI.
| michaelt wrote:
| 50 FPS certainly doesn't make the game unplayable, it's true.
|
| But if that's with an empty map, it's an ominous sign for
| people who plan to build a city in this city building game,
| as rendering usually slows as polygon count increases.
|
| And if it's with an $1800 GPU, its an ominous sign for the
| 99.3% of gamers who don't have an $1800 GPU.
|
| And if that's at 1440p - well, I'd wager a lot of the folks
| with a $1800 GPU also have a 4k screen.
| antisthenes wrote:
| It's a huge issue, considering the game doesn't look like a
| revolution in graphics AND you're essentially playing it on a
| super-computer by 5-6 year old standards.
|
| > I put 8 hours into Frostpunk before realizing I was limited
| to 30FPS.
|
| Yes, and I have beaten games that rendered improperly at
| sub-30FPS as a kid, because the requirement was 16MB of VRAM,
| while my card only had 8. But that's just as irrelevant to
| the point being made.
| pdpi wrote:
| If they said "The game is locked to 30fps" I'd probably have
| been ok with that. It's not a fast paced game, and capping
| the framerate at a low value is probably a decent enough
| design choice for that particular game. As you say, it's
| enough that you won't really notice it in gameplay.
|
| The problem is that we're talking about the absolute highest
| end consumer card sold by nVidia rendering a blank map,
| without such a framerate lock in place. If it's not locked,
| then that framerate is the product of a performance
| bottleneck somewhere. With a modern gaming PC, "performance
| bottleneck" also means you're sitting next to a moderately
| powerful space heater.
| KennyBlanken wrote:
| Locking a game to 30hz would not have been a "decent design
| choice" given that's half the refresh rate of nearly every
| computer monitor out there, especially for a game that
| involves a lot of scrolling.
|
| Games get mocked for locking to 60hz, 30hz would get them
| laughed at outright.
| BlueTemplar wrote:
| The devs supposedly aimed for 30 FPS (but missed it,
| because that would be for low end hardware?)
|
| In my experience 30 FPS is perfectly fine for a game like
| this, I'm not sure what do you mean about "scrolling" ?
|
| It's the much more fast paced, mostly first person games
| that really need those 60+ FPS...
| serf wrote:
| >In my experience 30 FPS is perfectly fine for a game
| like this, I'm not sure what do you mean about
| "scrolling" ?
|
| A top down city sim 'scrolls' up/down/left/right around a
| map, which is a movement that is strongly associated to
| screen tearing -- which i'm not really sure is a relevant
| thing to bring up given the variety of v-sync options
| available.
|
| Even a game locked at 30fps that 'scrolls' often
| shouldn't experience significant tearing with the options
| out there.
| mcpackieh wrote:
| Even without tearing, a high framerate is very desirable
| for scrolling around a map or web page. For instance,
| iphones are widely praised for excelling in this regard.
| It's just easier on the eyes if things on screen move
| smoothly when you scroll/pan around, it's less fatiguing.
| Aeolun wrote:
| I notice when my mouse updates at 30fps. The movement
| feels noticably choppier.
| kriops wrote:
| It's like playing in black and white, kind of. But yes 50 FPS
| is completely fine if you have not gotten used to better.
| Also helps if the game doesn't rely on high levels of detail
| (which Skylines, unfortunately, does), so that moving the
| camera around doesn't just make everything blurry.
| BlueTemplar wrote:
| If moving the camera around makes things blurry, then they
| must be using some kind of weird post-processing on top,
| can't blame that on the FPS alone !
| xcv123 wrote:
| The issue is that if it runs at 50 FPS on an RTX 4090, it's
| going to run at 5 FPS on an average GPU.
| Kuxe wrote:
| Problem with these kinds of statements is that.. well,
| they're not true. I have RTX 2070 running in 3440x1440. I'm
| at 35fps on low settings with some special stuff turned
| off, e.g GI and fogs. In an small town with 4k inhabitants
| still, so things can get worse for me.
|
| Regardless, the problem is blown out of proportion. I think
| we can give it some weeks and 4090 performance issues will
| be in the past.
| xcv123 wrote:
| > I have RTX 2070 running in 3440x1440. I'm at 35fps on
| low setting
|
| LOL. You just proved my point.
| dimgl wrote:
| Yes, yes it is. I have a 4k TV with 120hz. I want to be able
| to use it. If Anno 1800 can run at 100+ framerates, surely
| this game can too.
| sidewndr46 wrote:
| As someone who can safely say that watching a video game in
| 30 FPS is equivalent to watching a slideshow, yes. It gives
| me intense motion sickness immediately. So I can't play
| something below 60 FPS
| xboxnolifes wrote:
| It's not just 50 fps. It's 50 fps with the best possible
| hardware and lowest possible settings.
|
| But also, yes, even in a city builder low for is very
| annoying when you're used to the smoothness of 100+.
| epolanski wrote:
| It's a 3000$ computer playing at low resolution, of course it
| is.
| nzgrover wrote:
| Heh, in New Zealand, the card alone is $3500NZD
| wlesieutre wrote:
| Display pixel counts for reference, a 1440p ultrawide (assuming
| 21:9 and not 32:9) is about 60% the pixel count of 4K
| 4K 16:9 3840 x 2160 = 8.29 million pixels (1.00x)
| 1440p 21:9 3440 x 1440 = 4.95 million pixels (0.60x)
| 1440p 16:9 2560 x 1440 = 3.69 million pixels (0.44x)
| LegitShady wrote:
| also 5120x1440 (32:9) is 7.37 million pixels (.89x)
| refulgentis wrote:
| This isn't accurate. Review here:
| https://www.ign.com/articles/cities-skylines-2-review
|
| "My Ryzen 7 3700X and RTX 3080 were able to handle it okay on
| just shy of max settings"
| LegitShady wrote:
| it is accurate. check the steam reviews.
| Kye wrote:
| I found this test of Cities Skylines 2 with the somewhat
| faster non-laptop version of my GPU. It's so bad.
| https://www.youtube.com/watch?v=eNHqs6HYI0I
|
| It _seems_ okay around 1080, but then you realize there's
| almost nothing going on and it's only doing ~60FPS. It'll
| choke on anything remotely complex.
| LegitShady wrote:
| its also a 3070ti and playing at 1080p, which is sort of
| ridiculous.
| refulgentis wrote:
| Would me reading Steam reviews make the IGN review use a
| 4090?
| LegitShady wrote:
| Perhaps it was edited after the fact. The claim about
| performance in OP has a link to youtube video recording
| the performance.
| refulgentis wrote:
| (insert my comment here, it's not important, I won't
| waste my time writing one. Let's see how many comments
| you can go before saying whoops. My bet is infinity / you
| never will)
| throw0101c wrote:
| > _How can a game even get to this point?_
|
| Quill18 had a live stream the day of release (Oct 19) and it
| seemed fine (2h10m in, after some Galactic Civilizations 4:
| Supernova) from what I saw:
|
| * https://www.twitch.tv/videos/1954873419
|
| Various other live stream days, as well as a separate Youtube
| series:
|
| * https://www.twitch.tv/quill18/schedule
|
| * https://www.youtube.com/@quill18/videos
| rasz wrote:
| Twitch Streamers stream fullhd and will usually run games at
| fullhd.
| KennyBlanken wrote:
| ...and they're not even remotely objective.
|
| The big ones are paid cash, and before they stream they've
| already had a team working with them, covering what exactly
| they do or don't want shown and talked about - and if there
| are any technical problems, they've been worked out, which
| may even include special builds of the game.
|
| The smaller ones are "partners" who get access to the game
| earlier than purchasers, zero-cost items for giveaways, and
| some access to game publisher staff - but only if they play
| by the (NDA'd) rules of the publisher, and those rules
| usually explicitly say "do not say anything negative about
| the game."
| epolanski wrote:
| I don't think CS2 is exactly the kind of game that lacked
| abundance of streamers waiting for the release to be
| honest.
| nine_k wrote:
| My cynical take is that everybody noticed, but did not fix.
| They shipped ASAP. New updates, where they address more tech
| debt, may fix some if this, claim major performance
| improvements, and thus guaranteed to generate positive press /
| posts / tweets.
| Jensson wrote:
| Their publisher is Paradox so probably. They don't have a
| "only ship when its ready" culture.
| Aeolun wrote:
| I don't think Paradox has a particularly urgent shipping
| culture either? It's whatever their QA department does that
| sucks.
| pseudosavant wrote:
| Perhaps it is limited by something other than the GPU after
| 50fps. It could be CPU bound at that point?
| fyloraspit wrote:
| Same question then, if it is CPU bound on a 13900k
| colechristensen wrote:
| Performance problems like this are usually complex dumb
| things. Like calculating a thing a million times when it
| only needs to be done once, doing one thing and context
| switching instead of batching, dumb locking that leaves
| threads spending most of their time waiting, creating and
| destroying a thing millions of times when it should be
| reused, etc etc.
|
| It will just be lots of these things that need to be found
| and corrected.
| jowea wrote:
| Maybe they're going for the Crysis plan and designing for
| future tech?
| cmovq wrote:
| Except Crysis looked incredible at the time. This is just
| Unity 3D graphics.
| naikrovek wrote:
| > How can a game even get to this point?
|
| lots of ways. It is very easy to do things which affect
| performance negatively.
|
| Once you're sure about what you're doing and the data that
| you're drawing to the screen, then you can start optimizing
| things so that performance goes up.
|
| Colossal chose to release when they did because they promised
| to release on that day. The game isn't done, and they've said
| this.
| danudey wrote:
| Not as powerful of a setup, but I have a 5800X, 32 GB of RAM,
| and an RX 6800 XT on a 4K monitor, and I got 5 FPS on the main
| menu (which was showing an empty grass field).
|
| I had to set my resolution to 1080p in order to get anything
| remotely resembling performance (this change alone took main
| menu FPS from 5 to ~55), plus disabling motion blur, depth of
| field, and volumetrics, and turning model fidelity down. I've
| also heard that disabling vsync can make a huge difference, but
| it didn't affect things for me.
| TylerH wrote:
| I fault the people complaining about performance of a city
| builder running at 1440p more than the developer. You're
| looking at concrete and asphalt and brick textures and shit
| like that. You don't need anything higher than 720p.
| riversflow wrote:
| >You don't need anything higher than 720p.
|
| I don't want the UI to look horrible on my 27" 1440p monitor.
| epolanski wrote:
| I do hope you're being sarcastic.
| dansalvato wrote:
| Dude, we're talking about an RTX 4090. A GPU that can run
| current-gen AAA games on max settings at 4k60fps should not
| slow to a crawl when rendering "concrete and asphalt and
| brick textures and shit", to use your words.
|
| So if I have a budget PC, I should be content with running
| this game at Nintendo 64 resolution? Or I guess it's on me to
| fork over $1,000 for a high-end GPU if I want to run it at
| 2/3 the resolution of a midrange mobile phone, as you
| suggest.
| legitster wrote:
| How is there _no_ mention of CPU?
|
| Simulators, especially ones of this size, are usually much more
| CPU intensive than GPU intensive. So a lot of reviews are
| obsessive about how beefy their GPU card is, but I suspect a lot
| of people have "top-heavy" rigs.
|
| I play a lot of simulator games. Most of them are not paragons of
| optimization (well, except maybe Factorio). But these are not
| action games so FPS is not really something I personally optimize
| for and I don't necessarily understand the importance of
| benchmarking it against, like, Cyberpunk.
|
| I wouldn't be surprised if Paragon's optimization fix will be
| just to nerf the depth-of-field.
| coffeebeqn wrote:
| Sure it uses CPU but desktop CPUs that gamers use are beasts.
| Even if it's a few years old mid level CPU it's better than any
| console or mobile device by a long shot. They have dedicated
| tower or liquid coolers so they can push a lot of power into
| the chips
| dharmab wrote:
| With sim games it's not purely about clock speed or TDP. The
| previous generation 5800X3D outperformed the newest chips in
| simulation games for a while thanks to it's very large cache.
| thesnide wrote:
| I'm wondering about any abuse of GPU for non GFX tasks in
| recent games.
| ElectricalUnion wrote:
| If it has embarrassingly parallel tasks that it can dispatch
| to a massively parallel subsystem dedicated to solving
| embarrassingly parallel tasks, is that abuse or smart use of
| resources?
|
| That being said most simulation games are usually memory-
| latency and memory-bandwidth limited, not compute limited.
| thesnide wrote:
| It looks a perfect match. At first. Then you realize that
| you are not alone.
|
| Very much like using the Java stream parallel API in a
| webserver is doing wonders in dev, but not in production,
| as you have many other threads serving other requests also
| starved for CPU cores
| __coaxialcabal wrote:
| Can you recommend some favorites?
| dragontamer wrote:
| I'm not the guy you're responding to but...
|
| Over the last 10 years, my favorite simulation games have
| been Factorio, Tropico (4 and 6), OpenTTD, and Two-point
| Hospital.
| legitster wrote:
| I'm still an old blowhard and think Tropico 1 is still the
| best. It was so tough that you _had_ to become a fascist
| every time. Which made it a harder game but also a much
| sharper commentary.
| legitster wrote:
| The Tropico games are a perfectly pleasant place to start.
| They get progressively easier and more casual.
|
| The Anno games are my all around favorite. 1404 is my
| favorite, but honestly 1800 is probably the best.
|
| Banished spawned an entire genre unto itself even as it
| hasn't aged gracefully. These survival sims have a lot more
| "bite" to them. Try Planetbase for a more streamlined
| experience. Timberborn if you like physics. Also have heard
| good things about Farthest Frontier.
|
| And then there's Frostpunk which is an all-around amazing
| experience. The theming and mood rivals any first-person
| cinematic shooter.
| nrjames wrote:
| Check out Oxygen Not Included
| amoss wrote:
| Captain of Industry is pretty good fun.
| jdiff wrote:
| If the issue is occuring even on the main menu and a completely
| new, bare level, that possibility can be safely eliminated.
| It's also highly doubtful that someone with a 4090 would have a
| CPU pitiful enough to excuse 20FPS.
| Aeolun wrote:
| Also, the speed of the simulation should not be bound to the
| rendering speed.
| MisterBastahrd wrote:
| Most people who have the money to invest in a 4090 are also
| going to invest in a beefy CPU. It's not a secret that CPUs are
| bottlenecks for games within the gaming community.
| mvdtnz wrote:
| CPU is not a bottleneck in Cities Skylines 2, nor most game
| to be frank.
| zbuttram wrote:
| The issues in question here surface even with an empty city, at
| higher populations the CPU will almost certainly become the
| issue but right now most of the complaints are unrelated to sim
| performance and it even seems like for most people the sim
| performs quite well even at higher populations.
| brucethemoose2 wrote:
| > Most of them are not paragons of optimization (well, except
| maybe Factorio). But these are not action games so FPS is not
| really something I personally optimize for and I don't
| necessarily understand the importance of benchmarking it
| against, like, Cyberpunk.
|
| In many of these games, FPS is correlated with simulation
| speed. So when the fps starts to chug, the simulations starts
| going slower too.
| mvdtnz wrote:
| Not in Cities Skylines 2.
| legitster wrote:
| Uhhh, this doesn't even make sense. Simulation speed is
| usually controlled by a button.
| Tadpole9181 wrote:
| It makes perfect sense. Simulations are done in discrete
| time chunks, as the framerate drops and those time chunks
| grow larger there are two choices: make the simulation bad,
| or make the world slow down.
|
| If you don't adjust simulation rate you start seeing higher
| instances of objects phasing through each other, collisions
| not preserving energy correctly, pathfinding just
| fundamentally breaking.
| brucethemoose2 wrote:
| Well ideally sim "ticks" are completely seperate from
| rendering, but this is not the reality in many games
| where they share the same thread, or where cross
| communication (like UI stuff) blocks enough to slow the
| other component down.
|
| Even in Minecraft, with its completely seperate
| server/client, the rendering can bog down TPS due to JVM
| pauses and other reasons I don't even understand.
| legitster wrote:
| But the simulation can be done as fast the CPU dictates.
| If the GPU has to drop frames to keep up, it doesn't
| impact the simulation.
| trevwilson wrote:
| They probably should have mentioned the CPU specs at least, but
| it's very easy to look at utilization while the game is running
| and see that it's GPU bound even with high-end video cards.
|
| Not to mention that this is on an empty map, so there's very
| little simulation even happening yet, and the big FPS gains
| come from turning down GPU effects like DoF and motion blur.
| Der_Einzige wrote:
| Rimworld, especially after running RimPy and converting
| textures for GPU, is extremely fast/well optimized even with
| hundreds of mods.
| Olreich wrote:
| The speed of your simulation in a game like this should have
| minimal impact on the framerate of the game. You can paint the
| screen 144 times a second, but only tick the simulation 10
| times a second and get a much better experience than painting
| the screen and updating the simulation 20 times a second. Maybe
| your city runs slowly, but you can still look around at the
| various bits moving slowly.
| riversflow wrote:
| Not trying to be rude, but this honestly reads like you don't
| actually play simulation/4X games.
|
| > Maybe your city runs slowly
|
| If the simulation is running slow that has a _much_ more
| detrimental on the quality of game play than jumping from 45
| fps to 75 fps. Sure, it's a simulation, but it's primarily a
| _game_ , not a weather model. And I say this as an early
| adopter of high hertz monitors and a frequent fps player. I
| absolutely need high frames and low input latency in a
| competitive PvP game, but in a strategy game, it's much more
| important that the tick rate is fast enough that it's
| interesting. A slow sim is boring, and it's not just me,
| people complain about this all the time in (late game)
| Stellaris(another Paradox title).
| jakogut wrote:
| Hugged to death?
| ssharp wrote:
| "The agony and the irony are killing me"
| notamy wrote:
| http://archive.today/jkQw9
|
| The actual site times out on my end; did it get hugged to death?
| ehPReth wrote:
| seems so, i just got a 503 from them after a very long delay.
| here's an IA link as well for those using Cloudflare's DNS
| resolvers:
| https://web.archive.org/web/20231026180914/https://www.dsoga...
| zbuttram wrote:
| Seems like this was likely from before the hotfix that was
| released this morning which has improvements for some of the more
| egregious issues mentioned like DOF, LOD, and global
| illumination:
| https://store.steampowered.com/news/app/949230/view/37093367...
|
| Still far from ideal but glad to see movement so quickly from the
| dev team and as has been mentioned the game is certainly playable
| albeit with some setting tweaks.
| aranelsurion wrote:
| Makes one think, if those issues were so quickly fixable, why
| they weren't fixed already to begin with?
| mplewis wrote:
| We know why. The game had to ship on a date, and those
| responsible were not willing to delay the launch date on the
| basis of performance issues.
| evanriley wrote:
| After seeing this happen time and time again, it's kind of
| a wild decision to make. So many negative reviews I see
| these days are about performance issues.
|
| You would think a little more time would be put into
| reaching at least some reasonable performance level.
| bakugo wrote:
| > So many negative reviews I see these days are about
| performance issues.
|
| Unfortunately, negative publicity from bad performance
| doesn't really stop these games from selling well, as
| proven by most AAA releases in the past few years.
| fourteenfour wrote:
| I think they released it without having a discussion about
| how bad an impression poor default settings could make. With
| a few adjustments it looks great and is very playable on my
| 3080 at 4K even before the patch. Really big blunder for
| sure.
| alberth wrote:
| Sounds like an Electron app.
| pelorat wrote:
| Unity, so basically the same but for games.
| ptasci67 wrote:
| I will fully concede that the trend of game makers releasing
| half-baked, poorly optimized games that are buggy and unplayable
| at launch is totally a thing and it is frustrating and we should
| demand better (though we keep buying so why would they stop?).
|
| BUT.... the online game community is so insufferable and this
| Cities Skylines II launch is a great example of it. The game is
| not about 4k 120 fps gameplay. It is a simulation game that runs
| fairly well even on last gen's hardware if you drop SOME of the
| fidelity settings. But that's not the predominant discourse. If
| people can't play it at 4k out of the box on their overpriced
| 4090 then they take straight to the internet to complain (and
| mind you they have tried fiddling with exactly 0 knobs to make it
| runnable).
|
| I am by no means making excuses for game makers who certainly
| share much of the blame for creating an environment of distrust
| among game fans. But the online discourse is just rage baiting
| and looking for anything to hate with minimal evidence or
| sometimes even outright lies. Makes me want to go into a cave and
| play my games without seeing any content or discussion about it.
| wackget wrote:
| > If people can't play it at 4k out of the box on their
| overpriced 4090 then they take straight to the internet to
| complain (and mind you they have tried fiddling with exactly 0
| knobs to make it runnable).
|
| The top comment contains this extract from an IGN review:
|
| > I have a 13900k, 64GB of RAM, and a RTX 4090, playing on a
| 1440p ultrawide monitor. I got 35fps at the main menu and in
| game on a brand new map w/o building a single thing. Turning
| off motion blur and depth field increased this from 35 to
| 50fps. Not a single other graphics setting changed the
| performance at all. I turned off every single setting I could
| or set it to the lowest possible, and still only got 50fps.
| ptasci67 wrote:
| Yes, I was addressing the broader discourse more generally,
| specifically Reddit. But you're right that the article did
| directly address this though I would say the tone and title
| of the article are incongruous with the simple fact that they
| were able to get the game to run well with minor tweaks.
|
| I take issue with "only got 50fps". This is not Counter
| Strike or a game that demands 300fps. 50fps (if your 1% lows
| are within reason) is completely playable.
| fluoridation wrote:
| It's not about the number by itself, it's what the number
| implies. If an RTX 4090 can't get to 60 FPS on an empty
| map, what will happen when the game is running on an RTX
| 4060 and it has to render a complete city?
|
| >I take issue with "only got 50fps". This is not Counter
| Strike or a game that demands 300fps. 50fps (if your 1%
| lows are within reason) is completely playable.
|
| You're complaining about people complaining. If you're
| satisfied with the game being playable then play it, but
| others may have different expectations of quality, and it's
| not wrong for them to voice their opinion when the product
| they paid money for doesn't meet them. I personally don't
| remember participating in a congress of gamers where
| everyone agreed not to complain about a game unless it was
| practically unplayable.
| Kuxe wrote:
| >If you're satisfied with the game being playable then
| play it, but others may have different expectations of
| quality, and it's not wrong for them to voice their
| opinion when the product they paid money for doesn't meet
| them.
|
| I'm all for voicing opinions in a civic and calm manner.
| Most people online voicing their opinions come off as
| know-it-all teens or children throwing tantrums. It's as-
| if they have a _right_ to a CS2 with 120fps. Paradox
| warned about bad performance prior to launch
| (https://forum.paradoxplaza.com/forum/threads/updates-on-
| modd...). Nobody claimed or said performance was gonna be
| great. And still, people act surprised.
|
| It's no surprise that the reviews was down at close to
| 30% a couple of hours into the release and today at 52%.
| Why is there such a massive bias towards large reviews at
| the first hours? Because many gamers loves thrashing
| about. It's much more important than taking a step down
| and calming down.
| fluoridation wrote:
| The people who reviewed the game in the first few days
| were the ones who either pre-ordered it or bought it as
| soon as it came out. They were so excited to play the new
| installment they took a gamble and trusted that the
| developer would produce a polished product, because they
| wanted to be able to play it as soon as possible. When
| they got to play the game, they saw it ran poorly to the
| point that it might have spoiled the experience for them.
| They're right to be angry about it, especially when
| developers and publishers make most of their money during
| the first few weeks since launch. By releasing a half-
| finished product they're treating their most enthusiastic
| users like crap. They didn't _have_ to do that, they
| could have delayed the launch. Be it because of decisions
| made by the publisher or by the developer, they chose to
| release when they did. They made their bed, now they have
| to lie in it. I don 't blame anyone who raves about
| performance, because what was released was well outside
| the realm of what's acceptable for a finished product,
| regardless of what said prior to launch. You don't get to
| sell a car with an asterisk that says "by the way, the
| fuel tank leaks so until we find a way to fix it you'll
| use twice as much fuel as normal".
| Kuxe wrote:
| > You don't get to sell a car with an asterisk that says
| "by the way, the fuel tank leaks so until we find a way
| to fix it you'll use twice as much fuel as normal".
|
| Yeah. But in case of CS2, gamers did buy the leaking car.
| Devs analogously said "by the way, the fuel tank leaks"
| and people just went with "OK" and bought CS2, after
| which the customer started to complain (rave?!) about
| leaking fuel tanks. The car salesman retail store said
| "Well you can have all money back no questions asked
| until you've driven at least 160km". Steam has generous
| refunds. What does the customer do? (S)he still goes onto
| review sites and bitch about bad leaking fuel tanks. It
| is very much in bad faith on the customers part.
|
| I wouldn't rush to Colossal Games defense if customers
| just said "It ran bad for me on my 4090 for some reason
| so I refunded". That's not what's going on with the
| negative reviews though. People act entitled.
| fluoridation wrote:
| You're still not getting it. Yeah, if a car dealership
| had such a generous return policy you could get your
| money back and get a car that does what you need within
| your budget. But these people didn't want just _a_ city
| builder and they happened to buy this one. They wanted to
| play the new version of Cities: Skylines. They 're loyal
| fans and they're treated like beta testers.
|
| Yes, it's entitlement. Customers are _entitled_ to get a
| quality product in exchange for their money. When Paradox
| goes to spend their earnings they 're not going to be
| throttled to do it at 45 cents per second.
| csydas wrote:
| this is why car analogies are dangerous :)
|
| I would argue a dealer telling you about a major defect
| directly before you buy the car is a bit different than a
| post on some forums that the product they're selling is
| not well made.
|
| I would suggest it's not reasonable to expect that
| someone buying a game has to do research on a forum to
| know the game is unfinished -- if it's being sold as a
| finished game it's reasonable to expect it's in a
| playable state. the original post was meaning to say it
| would be unheard of for other products to allow companies
| to sell known unfinished products as finished products,
| even with the promise of completing the product. and
| consumers would similarly balk at such a proposal for
| virtually any other object.
|
| it was more the absurdity of the different way games are
| treated which is anti-consumer.
| albertzeyer wrote:
| I have played a little bit and I'm a bit confused about all the
| negative feedback regarding the performance.
|
| I have an AMD Ryzen 9 3900X and NVIDIA GeForce RTX 3090, so not
| too bad, but also not the best.
|
| And the game runs totally fine for me. I just reduced the screen
| resolution (I think 1920x1024 or so) and left the other settings
| as they were (I think all set to High). And I get very constant
| 45 FPS without hiccups. It feels very smooth and playable. Maybe
| it will get a bit worse when my city gets bigger, but so far, I
| don't see any problems.
|
| This is even before the first patch, which was released today.
| This patch is supposed to optimize the performance more. I will
| try it later.
|
| Note, this is with Linux. I run it on Linux with Proton 8. So
| maybe it actually runs better on Linux than on Windows?
| LegitShady wrote:
| You're playing on an rtx 3090 at 1920x1024 and get 45 fps. I'm
| confused why you're confused about the negative feedback.
| crimsontech wrote:
| I don't get it.I would be fuming, it's a 4k card and it is
| sputtering out 45fps at 1080p.
| sp332 wrote:
| 4k is a resolution, not an amount of work done per frame.
| fermentation wrote:
| You have thousands of dollars in hardware and cannot hit full
| framerate on a small city at a low resolution.
| Wowfunhappy wrote:
| Is this on maximum settings? Does the game still look good when
| the settings are turned down?
|
| If so (!), I see no issue here, and on the contrary I applaud
| them for including options. It gives players more options (the
| person who doesn't mind aliasing can play at 1080p and get
| incredible lighting effects), and makes the game more future
| proof.
| bee_rider wrote:
| The we appear to have slashdotted the site, so I'm just reacting
| to the headline, but is this a big problem? It is a Paradox
| published game, presumably we'll be playing it in 2030 on
| RTX9070's or whatever (And buying expansions).
|
| It seems like a shame if the ultimate quality is limited by the
| hardware available now. Turn down the setting 'till you hit 30fps
| (this isn't a twitch shooter.
| garrettjoecox wrote:
| You do realize a 4090 is currently $1,600 USD right?
| bee_rider wrote:
| Yeah it seems like a ripoff, I'm sure in a couple years more
| capable, cheaper cards will come out (and then people still
| playing this game can enjoy it with all the bells and
| whistles).
| CyberDildonics wrote:
| _Yeah it seems like a ripoff,_
|
| What you wish something would be priced is not the point.
| It is priced relative to its power.
|
| _I'm sure in a couple years more capable, cheaper cards
| will come out_
|
| Maybe they should wait and release their unoptimized game
| then.
| u10 wrote:
| While the game should not have been released in this state (or at
| least with defaults set to broken graphics settings, such as
| motion blur and depth of field), CO released a patch today and
| users are saying that it has dramatically increased performance.
|
| https://old.reddit.com/r/CitiesSkylines/comments/17gxkrf/pat...
| brianflakes wrote:
| I believe that lots of the pushback on the game are due to poor
| defaults. Lots of unnecessary graphics settings are enabled,
| leading to bad first time experiences. After disabling elements
| that most users won't care about (motion blur, volumetric clouds,
| global illumination) and following some tips from people online,
| I find no issue enjoying the game >60fps without my GPU fans
| screaming. (to be fair, using 7900x / 3090)
|
| Yes the graphics need optimization, but they could have rolled
| out the release with good defaults and just admitted that high
| end graphics needed more time... but you can enjoy the game in
| the meantime! Instead, their steam reviews will remain marred.
| fourteenfour wrote:
| Yay, finally a comment that is similar to my experience and
| isn't speculation or kneejerk reaction. I'm enjoying the game
| at 4k and it looks better than CS now that I've adjusted
| things. They totally screwed up the default settings, not sure
| why as it seems like an unforced error. It actually picked a
| resolution with a 24Hz refresh rate when it first loaded for
| me. I think their "virtual texturing" optimization also kicks
| off on first run, making the initial menu experience stuttery
| and slow.
| brucethemoose2 wrote:
| In other words, this is a true Paradox game.
|
| For those that don't play other Paradox published games... Newly
| launched Paradox games are early access. Maybe _perpetual_ early
| access with on-and-off performance woes, like Stellaris and other
| Clausewitz games. As much as I like Paradox games, that and the
| endless DLC is the steep price you pay for their sim games.
| bcrosby95 wrote:
| There's certain problematic graphical settings for CS2 right now.
| If you have a potato GPU and put it on low settings (as you
| should with a low end GPU) the performance is fine.
|
| My card is below the recommended specs and its perfectly playable
| with a city of 50k.
| deanCommie wrote:
| According to Collossal Order:
| https://twitter.com/ColossalOrder/status/1716883884724322795
|
| > If you're having issues with performance, we recommend you
| reduce screen resolution to 1080p, disable Depth of Field and
| Volumetrics, and reduce Global Illumination while we work on
| solving the issues affecting performance.
|
| I tried it and had no problems. Everything is smooth, no visual
| problems. I have a mid-tier Radeon from 3 years ago.
|
| Yeah they fucked up the launch and it's unfortunate they're
| getting such negative press, but it's a city simulator it doesn't
| need 120FPS, what are we talking about?
|
| If you want to just have fun and go build cities, there's a clear
| path for it
| jiggawatts wrote:
| Some people are fine with driving at 10 mph because the
| transmission is broken in their brand new car. "It's not a
| Ferrari and I'm not racing."
| nrjames wrote:
| It's likely they had a financial agreement with their publisher
| that stated that the game would launch prior to November 2023 or
| they would lose some % of the revenue. Publisher contracts often
| also mandate a certain Metacritic score must be met in order to
| receive full payment.
| bryanlarsen wrote:
| Looks like this is the straw that'll finally get me to upgrade my
| 12 year old CPU and 7 year old GPU. Has anybody figured out what
| a usable GPU is with today's patch? My initial research indicated
| that there's a sharp performance hit for 8GB of VRAM or less, so
| I was thinking the 12GB 6700XT might be the cheapest card with
| usable performance. Is that still the case with Friday's patch?
|
| So far the best info I've found is the spreadsheet attached to
| City Planner Play's benchmarking guide:
| https://www.youtube.com/watch?v=wyNiXYC9eoM
| petercooper wrote:
| Standards have certainly changed over the years. This takes me
| straight back to 2003 when SimCity 4 came out, turned out to be
| an absolute resource hog, and I'd have been overjoyed with 20fps.
|
| As the late Henry Petroski said: "The most amazing achievement of
| the computer software industry is its continuing cancellation of
| the steady and staggering gains made by the computer hardware
| industry."
| BenoitP wrote:
| I've heard this version: "what Andy giveth, Bill taketh away"
|
| (Intel vs Microsoft CEOs)
|
| https://en.m.wikipedia.org/wiki/Andy_and_Bill%27s_law
|
| Henry Petroski probably said it first though
| Aperocky wrote:
| I've always had a felt for city building games having played
| since Simcity 4 first came out in 2003. imo the genre does not
| need great graphic - it needs great content, Song of Syx is one
| of my more recent (~2 years) favorite, having a city building
| game unable to run with the best video card is baffling at best.
|
| I still consider Simcity 4 with full mods and addons the best
| ever, but for some reason the industry doesn't seem to make
| similar titles any more, the newer city building games focus on
| details but gives up on scale.
| harrid wrote:
| It also uses 10+k drawcalls per frame. It's amateur hour over
| there, bordering on fraud. Even more frustrating to see how
| successful that model is. And how feverish people defend them
| online.
| naikrovek wrote:
| this is what happens when you focus on content before
| performance. they'll get that number _way_ down over time.
|
| games are data problems, like every other computer program. you
| must know your data before you can optimize how it is moved
| around efficiently. draw calls are a very good benchmark for
| this. as they optimize, it will go way down, if it hasn't
| already.
| tbm57 wrote:
| software optimization as a practice is dead
| jncfhnb wrote:
| People make dumb problems for themselves. Play games on 1080.
| Hardware requirements for otherwise max settings drop like a
| rock.
| xtracto wrote:
| Ooh is thereba new CiSk game? I enjoyed playing the first one
| some years ago I'll have to check it out!
| Geee wrote:
| I'm guessing they're rendering those trees without any LODs, i.e.
| they're always pushing millions of polygons. That's why the
| settings don't make a huge difference.
| Daunk wrote:
| Unity. -\\_(tsu)_/-
| Waterluvian wrote:
| I've been playing Pillars of Eternity and I'm amazed just how
| poorly it runs given what it is.
|
| I'd love a no-judgment, no-blame post mortem on a game like that,
| which tries to dig into the technical reasons why it's so poorly
| optimized and the non-technical reasons for how they get there.
|
| I'd imagine it usually boils down to having a deadline and a
| budget, but I'm quite interested in understanding the in-between
| portion much more intimately.
| TylerE wrote:
| Well, they both use the same engine. That might be a hint. Yes,
| to be clear, I'm talking about Unity.
| Waterluvian wrote:
| That's one of those convenient non-exploratory answers I'm
| hoping to dig past. It might be related, but we don't really
| know based on this alone.
| TylerE wrote:
| Virtually every Unity game ever that does a non-trivial
| level of background simulation has considerable jank.
| There's just a point where you hav wot say - look - Unity
| games consistently have performance issues at some point
| it's legit to just call it a poorly performing platform.
| Waterluvian wrote:
| Cities Skylines 1 performed remarkably well on midrange
| computers, particularly for just how good the simulation
| is and how busy the viewport can get.
| scubadude wrote:
| It's basically an emulator versus native C code, it's
| completely relevant
| pawelduda wrote:
| It will keep happening because people rush to preorder or buy
| shitty, unfinished games on a release day- just look at Jedi
| Survivor from this year. The game was in a laughable state for
| months after release. I thought waiting for so long was enough
| for the patch shitshow to be over.
|
| No incentives to ship good quality games, especially looking at
| larger companies. Luckily there are still sincere reviewers that
| will give you heads up about it
|
| I want to give shout-out to the studio behind Lies of P. Game was
| super well optimized since release and I can't recall a single
| bug during dozens of hours of gameplay. It's a bit sad that it
| went by without a lot of hype. This is the kind of game dev that
| should be praised.
| Bellend wrote:
| It's a Unity game. Why are people genuinely surprised that
| performance is absolute garbage? Both CS2 and KSP2 (kerbal)
| REQUIRE incredibly specific tuning that unity simply does not
| offer. The first being rigid bodies at scale (which is semi
| impossible with a dedicated engine) and the latter being "almost-
| factorio-level" low level cell processing. The graphics of both
| are second tier to that.
|
| But no, lets use Unity...
| poisonarena wrote:
| everytime i used to play a game and the made by unity logo
| popped up i would let out a big sigh...
| Waterluvian wrote:
| How hard is it to debug shaders? I admit I've only written like
| two shaders for WebGL and I found it to feel like a black box I
| feed code into and see what I get out the other end.
| NKosmatos wrote:
| Do modern AAA games support dual (or multiple) GPUs? In my days
| we had SLI and Crossfire, but I don't think there's something
| like it nowadays.
|
| For sure better coded games with proper software driver support
| will be needed, but I think this trend of having bigger and more
| power hungry GPUs needs to de-escalate.
|
| I was reading this article about multi GPU gaming, and perhaps
| this is the way to go: https://www.xda-developers.com/what-
| happened-multi-gpu-gamin...
| Iulioh wrote:
| >Do modern AAA games support dual (or multiple) GPUs?
|
| The problem is that modern GPUs don't support multiple GPUs lol
|
| Iirc the series 30 was the last to do it and only 3070 and up
| TinkersW wrote:
| This game sure is getting a lot of free marketing from all the
| complaints about the hilariously bad performance for such a mid
| looking game.
___________________________________________________________________
(page generated 2023-10-26 23:00 UTC)