Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Why Cities: Skylines 2 performs poorly

blog.paavo.me

Post reply on HN
Print view View original

521–530 of 593 posts

Re: Why Cities: Skylines 2 performs poorly

#521

Post by johnnyanmac »

solardev wrote: post #91
dvaletin wrote: post #79

Earlier quoted context omitted.

Which only incentivize companies to publish unfinished products with “will fix it later” ideology.

Is that a big deal? That's easy to avoid if you don't pre-order games and just wait for the day 1 reviews. Even if you did end up with a shitty situation, Steam lets you refund the games with minimal hassle. On the other hand, there are players who'd rather have the game earlier (like me) than a few months later, despite its launch issues. The alternative approach -- Baldur's Gate 3 being in Early Access forever -- i…

large companies don't have much incentive to use EA. The usual incentive for EA is to make a game cheaper for early users and to get important playtesting done.a big game with a big ad budget, QA testing studio, and internal playteting doesn't need to worry about that.

Re: Why Cities: Skylines 2 performs poorly

#522

Post by _w1tm »

torginus wrote: post #516
psunavy03 wrote: post #405

Earlier quoted context omitted.

Agile done properly is literally the opposite of this. Big Important Exec spouting Agile terms they don't understand and cluelessly forcing top-down crap on the teams, more likely. Normalize your story points, prole! Daddy needs metrics!

Personally, I've never seen 'Agile' done well. Every project where we made good progress was engineer driven. We talked to customers about what they wanted so see and we created tasks and implemented them. Every project managed by a dedicated servant-leader Six Sigma Kanban certified Scrum ninja-coach project has been a bureaucratic shitshow, with constant cargo culting to agile priniciples and forced ceremonies.

> Personally, I've never seen 'Agile' done well. Every project where we made good progress was engineer driven. We talked to customers about what they wanted so see and we created tasks and implemented them.

> Every project managed by a dedicated servant-leader Six Sigma Kanban certified Scrum ninja-coach project has been a bureaucratic shitshow, with constant cargo culting to agile priniciples and forced ceremonies.

The engineer-led project sounds more agile than the latter Agile insanity.

Re: Why Cities: Skylines 2 performs poorly

#523

Post by kamray23 »

johnnyanmac wrote: post #514
kamray23 wrote: post #513

Earlier quoted context omitted.

random quirks do. this is not random quirks. it's a systematic and expected issue of underoptimisation caused by releasing a product before it was even slated to be ready. one bad mesh is not the issue, it's never the issue. we're talking of thousands of terrible meshes and a near-total lack of basic optimisations applied at the last stage of development to most games. the manpower was not enough to release within th…

I'd call 4 examples of "this needed LODs" random quirks in the grand scheme of things. It's not like every single mesh is 100k vertices. Grossly underoptimized, yes. But the devs pre-empting their announcement with "we're not satisfied" tells me they were too busy slaying dragons to worry about the annoying barking Chihuahua in the room. It was expected, yes. It does not mean they weren't trying to fix it in the 11th…

It's pretty clear that with the performance that the game has it's a lot more than four models with bad LODs. They would probably have identified that issue within days. There's likely a more fundamental issue, not only with the technology but also with LODs for most detail models. These are only examples since you can't list every model out of hundreds being rendered.

I really think that the performance and scale indie developers can squeeze out puts AAA developers to shame nowadays, and I really hope that that'll continue to happen. All it takes is time, organisation, and a lot of time.

Re: Why Cities: Skylines 2 performs poorly

#524

Post by DonHopkins »

DonHopkins wrote: post #493
jcranmer wrote: post #443

Earlier quoted context omitted.

That's not entirely true. SimCity 4 was the last of statistical city simulators. That is to say, if you put down something like a distressed Ong Condos, the game would go "okay, here's 20,778 people in this node in the transportation network, do some network flow and figure out where they work." Newer titles, including the ill-fated SimCity (2013), are agent-based, where they essentially drop in a person and tell the…

>SimCity doesn't lend itself well to multiplayer functionality You make a lot of great and accurate points (including blaming EA instead of Maxis). But I disagree that SimCity doesn't lend itself well to multiplayer functionality. (And perhaps EA isn't the right company to pull it off properly, being more focused on competition and violence than collaboration and education.) I agree that it would be difficult to adap…

Here's a proposal I wrote to Maxis in 2002 asking to license the rights to develop a commercial educational version of SimCity for Linux, based on the multi player Unix version, with the support of Columbia University. They didn't bite at the time, but Will Wright and I eventually talked them into actually making it open source in 2008:

https://web.archive.org/web/20110611185928/http://www.donhop...

>Educational Multi Player SimCity for Linux Proposal

>Submitted by dhopkins on Sat, 2004-02-07 14:27. C Cellular Automata Pie Menu Applications SimCity TCL Game Design

>Back in March 2002, Maxis told me they were interested in supporting the educational use of products like SimCity. Earlier, I had developed a multi player version of SimCity, which runs on Linux/X11, and was scriptable in TCL. Educators and researchers from Columbia University, MIT, IBM, Xerox and other educational and commercial institutions were excited about gaining access to this version of SimCity, and adapting it to teach and stimulate students' interest in urban planning, computer simulation and game programming.

>So I wrote this proposal and presented it to Maxis, but nothing ever became of it. But recently, Will Wright has been pushing EA to relicense SimCity under the GPL, so the OLPC project can use it. So it may eventually see the light of day!

[...]

More SimCity stuff from my old Drupal blog:

https://web.archive.org/web/20110611162810/http://www.donhop...

https://web.archive.org/web/20111109132608/http://www.donhop...

Re: Why Cities: Skylines 2 performs poorly

#525

Post by johnnyanmac »

kamray23 wrote: post #523
johnnyanmac wrote: post #514

Earlier quoted context omitted.

I'd call 4 examples of "this needed LODs" random quirks in the grand scheme of things. It's not like every single mesh is 100k vertices. Grossly underoptimized, yes. But the devs pre-empting their announcement with "we're not satisfied" tells me they were too busy slaying dragons to worry about the annoying barking Chihuahua in the room. It was expected, yes. It does not mean they weren't trying to fix it in the 11th…

It's pretty clear that with the performance that the game has it's a lot more than four models with bad LODs. They would probably have identified that issue within days. There's likely a more fundamental issue, not only with the technology but also with LODs for most detail models. These are only examples since you can't list every model out of hundreds being rendered. I really think that the performance and scale in…

Sure, more than 4, less than however many assets there are in the game. I mostly want to emphasize that a systematic issue implies that this was an acceptable development pipeline, which I doubt any engineer on the team would say.

>I really think that the performance and scale indie developers can squeeze out puts AAA developers to shame nowadays, and I really hope that that'll continue to happen. All it takes is time, organisation, and a lot of time.

Yup, I agree. Indies don't tend to have money, so the budget comes from. Elsewhere. We can already utilize some amazing tools to cut down the time of scaling up environments. Not as much with actor assets. But I don't think it's too far off (more just locked off in acedemics white papers).

Re: Why Cities: Skylines 2 performs poorly

#526

Post by johnnyanmac »

mvdtnz wrote: post #276
TonyTrapp wrote: post #220

Earlier quoted context omitted.

This is rarely developers' fault. You can bet they wanted to deliver the best product possible, but were not given the time needed to do that by upper management.

I didn't blame the developers. I have been involved in projects that I'm embarrassed by even though the worst decisions were the ones made by upper management (hell I worked on the new Jira front end, a continuing source of humiliation).

I just become apathetic. Shame implies that I have control, I'm the face of the game, and that the game's rep directly affects my day to day.

I'm a cog and even if I break the machine has plenty of redundancy. Nothing to be ashamed about. I got my experiences, expanded network, compensaion, and war stories, they got my X years of labor. It's all a transaction towards each others' true end goals.

Re: Why Cities: Skylines 2 performs poorly

#527

Post by johnnyanmac »

hipadev23 wrote: post #236
TonyTrapp wrote: post #220

Earlier quoted context omitted.

This is rarely developers' fault. You can bet they wanted to deliver the best product possible, but were not given the time needed to do that by upper management.

Why is it impossible that maybe Cities Skylines simply has shitty developers?

Not impossible, but it's an shallow dismissal often used by consumers who don't know how the sausage is made. Large claims require large proof.

IME, the games industry isn't like other pieces of the tech industry where, say, 30% of devs clock in and clock out, collecting a paycheck. it's much rarer (but not impossible) to find someone still in the industry after 5+ years and not passionate. The churn is enourmous and the kinds of management in there are always pushing to get more out of you.

You don't stick with that because you are low quality or lazy. You do it for the people you work with or the product you work on. Naively, perhaps. But there is certainly fulfillment you get out of it.

Re: Why Cities: Skylines 2 performs poorly

#528

Post by animal531 »

Sakos wrote: post #375
thrillgore wrote: post #246

"And the reason why the game has its own culling implementation instead of using Unity’s built in solution because Colossal Order had to implement quite a lot of the graphics side themselves because Unity’s integration between DOTS and HDRP is still very much a work in progress and arguably unsuitable for most actual games." This sadly tracks with my own experiences with Unity's tooling, where DOTS did ship but its i…

I don't understand how Unity burns through up to a billion in revenue every single year, yet their engine still feels so half-baked and unpolished. Where's all that money going?

Their worst problem (compared to Epic) is that they don't finish projects.

They'll want a new feature, for example new scriptable graphics pipelines which will allow them to modify the core rendering a lot easier. Ok that's fine, but then instead of implementing one they try to implement two, while at the same time completely ignoring the existing pipeline which everyone is currently using. The time to get the new pipelines working properly and onboard everyone is counted in years in this case.

Those features they actually "finished", but there are so many experimental packages that just sit there forever, or work in a half-baked manner.

Re: Why Cities: Skylines 2 performs poorly

#529

Post by geon »

fulafel wrote: post #494
heywhatupboys wrote: post #482

Earlier quoted context omitted.

Unity is older than the Core runtime. They had to use Mono as their base.

Assuming they had C# or .NET as a constraint. I'd wager not, they just wanted something that would work for "scripting".

C# and their own custom clr language was the only supported languages, so yes.

Re: Why Cities: Skylines 2 performs poorly

#530

Post by johnnyanmac »

Pannoniae wrote: post #37
xyzzy_plugh wrote: post #32

It's kind of stunning that a game of this magnitude is able to go out the door without model LOD. I suppose the fact that it runs at all is stunning -- surely you could not get away with this a decade or two ago -- but perhaps it speaks to the incredible capabilities of modern hardware. This feels a bit similar to the Electron criticism, where convenience ultimately trumps performance, and users ultimately don't care…

Don't forget the part where they use web tech and waste draw calls like crazy on the UI. These things should literally be banned. edit: not web tech should be banned, but releasing a game with horrible optimisation like this, either by the store selling the game or by the law

Ehh, I've had ideas to try and bridge the web stack to game development. As much as I hate JS, I've hated every other GUI system in games even more, except IMGUI (which isn't necessarily flexible enough for production UI). So why not utilize the decades of knowledge from an industry who's entire paradigm is improving UI workflows.

Of course, it's a butt ton easier said than done, but if the industry could utilize flash for years after it died, I see a glimmer of hope.

>releasing a game with horrible optimisation like this, either by the store selling the game or by the law

you really want the government to regulate video games because you can't run a frame in 16ms? Talk about a slippery slope.

Post reply on HN