Possible "OpenSC4" ?

[cydonianmystery 96]

I apologize in advance if this has been covered before, but I'm genuinely curious.

With other older classic games such as Morrowind and Rollercoaster Tycoon, the modding communities addressed their games' older engines and limitations by making new and open source engines for the game; OpenRCT2 and OpenMW. Both are quite amazing for what they do, and OpenMW addresses a lot of the engine-level bugs and issues TES IV had. OpenRCT2 also has things like modern resolution options, which are not available in base RCT2. So I'm wondering if the community has ever considered a potential "OpenSC4" or similar to address this game's various engine-level issues like single-core limitations, no passenger/el train "cabooses", and other issues I've seen brought up over the years that just don't seem to have any possible solution.

I decided to make this a new topic in case this might lead to further discussion.

[Cyclone Boom 16,854]

I recall this topic from a few years ago on the same subject:

(It's fine how you've made a new topic too, but just in case the older thread is insightful.)

[cydonianmystery 96]

So looking at the linked thread, the issues seem to mainly boil down to a fear of EA's legal team (very understandable), using existing assets (this is usually not an issue with open source engines, but I don't know enough about SC4's engine to comment), and that it'd probably be a better idea to just make an entirely new, open source simulator that is SC4-like.

And to the latter point, I actually think that'd be a good idea. In spite of its age, SimCity 4 offers a lot of major advantages over more "modern", 3D games like Cities Skylines. CSL blew SC13 out of the water, for sure; but it's still a far cry from the region-based gameplay SC4 offers and the sheer size of maps. The isometric style of SC4 just allows you to do more stuff than full 3D rendering.

And (I'm much more knowledgeable of the RPG side of video games than sim games) we've seen a resurgence in isometric RPGs in the last few years (Shadowrun Returns, Pillars of Eternity, Pathfinder Kingmaker) because like SC4, RPGs can just do more when they ditch the "everything must be 3D" approach. So if anything, it's proof that something like this could not only work, but be popular.

[Lazarou Monkey Terror 1,819]

Someday we'll liberate SimCity4 for the Masses 

[HayleyRenaNemeth 465]

That would be really nice. SimCity 4 is older than half of the people on the internet. SimCity 4 still has so many issues that can cause it to be unstable and crash.. It would be nice if we as a community would be able to fix those issues like they did with OpenRCT2. 

[Naomi57 2,985]

Hi @cydonianmystery,

I, too, would love to see an OpenCities project.

Coming from a development background with large scale systems, I can see there are multiple reasons why this OpenCities hasn't happened:

• Fear of EA's legal team.
• People play SC4 for different reasons.  Reconciling all those different reasons is trickier than it looks.
• Maxis SC4 is a carefully balanced large scale simulation (in gaming terms), more and more impressive the deeper you dig into it, even though it has its limitations and is showing its age.  Replicating that careful balance is difficult with an open source project, requiring assertive team leads who have the energy and are in it for the long haul.
• There is a truly enormous amount of 3rd party content, and that content works around and exploits and conforms to SC4 quirks.  Building an automated porting tool for that content is a big project in itself, and it's not a silver bullet.  Each piece of 3rd party content would need to be converted, with porting finished by hand, then tested, and beta tested ... and permission sought from the 3rd party creator.

It's huge, and our community isn't.  When it happens, it would need to carefully piggy-back on existing open source gaming platforms.  Attempting to reinvent the wheel would be a seal of doom on the project.

It's a darned shame that the original Maxis development team was disbanded.  A SimCity 4.x series could have had a long and glorious life.    

I think someday, several someones will pull together an OpenCities project.  So many possibilities:

1. Configurable keybindings and menus.
2. Multi-threaded CPU and modern graphics hardware support.
3. True grade-separated networks, with detailed tunnel levels.
4. Subways that build as an underground RRW.
5. Network/building overlaps.
6. JSON or XML data definition for mod files, using a NoSQL database platform for fast/scalable/modular interchange.
     https://en.wikipedia.org/wiki/NoSQL
7. Automated dependency management.
8. Improved inter-city region play.
9. Data interchange for neighbouring multi-region play shared over the internet.
10. And much more ...

My career focus being data technology, I'll happily volunteer to contribute on data definition and data interchange, but I don't do any C++ programming these days.  That ship left long ago.

[Naomi57 2,985]

On 31/08/2021 at 6:58 AM, HayleyRenaNemeth said:

SimCity 4 still has so many issues that can cause it to be unstable and crash.

Incidentally, just checking, have you installed the .DLL fixes to improve SC4 stability?

• Official Maxis patches:  https://www.NAM.sc4devotion.com/index.php?title=SimCity_4_version_1.1.638
• https://community.simtropolis.com/files/file/30883-sc4fix-third-party-patches-for-sc4/
• https://www.NAM.sc4devotion.com/index.php?title=NTCore_4GB_Patch

There's configuration issues to improve stability, too.  -CPUCount:1 is an easy one.  Graphics configuration is another common source of stability problems.

[Leo -- 186]

20 minutes ago, Naomi57 said:

Hi @cydonianmystery,

I, too, would love to see an OpenCities project.

Coming from a development background with large scale systems, I can see there are multiple reasons why this OpenCities hasn't happened:

• Fear of EA's legal team.
• People play SC4 for different reasons.  Reconciling all those different reasons is trickier than it looks.
• Maxis SC4 is a carefully balanced large scale simulation (in gaming terms), more and more impressive the deeper you dig into it, even though it has it's limitations and is showing it's age.  Replicating that careful balance is difficult with an open source project, requiring assertive team leads who have the energy and are in it for the long haul.
• There is a truly enormous amount of 3rd party content, and that content works around and exploits and conforms to SC4 quirks.  Building an automated porting tool for that content is a big project in itself, and it's not a silver bullet.  Each piece of 3rd party content would need to be converted, with porting finished by hand, then tested, and beta tested ... and permission sought from the 3rd party creator.

Its also (from my understating) currently impossible to decompile C++ code (the language that SC4 is written in) while it is possible to decompile C (the language that TTD and RCT1&2 was written in from my understanding) which will make a full open source rewritten extremely difficult if not impossible so you are just better off with a rewrite.

Tbh on the subject SC4's source code I'm surprised that it hasn't been leaked by a disgruntled Maxis employee.

[Naomi57 2,985]

31 minutes ago, Leo -- said:

Its also (from my understating) currently impossible to decompile C++ code (the language that SC4 is written in) while it is possible to decompile C (the language that TTD and RCT1&2 was written in from my understanding) which will make a full open source rewritten extremely difficult if not impossible so you are just better off with a rewrite.

Tbh on the subject SC4's source code I'm surprised that it hasn't been leaked by a disgruntled Maxis employee.

The use of decompiled source code is not as easy as it sounds.  With comments, constant definitions, function names, and variable names, all stripped from the decompiled source code, all those loses conspire to make decompiled source difficult to decipher, and that's assuming the company didn't take measures to protect the source code more deliberately.  The time spent by a talented programmer, puzzling over the original source code, is often time better spent by said talented programmer making something completely new, especially 20 years later.

Even a project's original source code, complete with source code comments and full documentation, is often close to worthless unless you have members of the original development team, too.  Software developers are not interchangeable, and the original Maxis development team have knowledge that cannot be readily regained.

Aside from all that, decompiling to create a derivative product would be illegal, and EA has a very capable and proactive legal department.   

[Leo -- 186]

21 minutes ago, Naomi57 said:

The use of decompiled source code is not as easy as it sounds.  With comments, constant definitions, function names, and variable names, all stripped from the decompiled source code, all those loses conspire to make decompiled source difficult to decipher, and that's assuming the company didn't take measures to protect the source code more deliberately.  The time spent by a talented programmer, puzzling over the original source code, is often time better spent by said talented programmer making something completely new, especially 20 years later.

Even a project's original source code, complete with source code comments and full documentation, is often close to worthless unless you have members of the original development team, too.  Software developers are not interchangeable, and the original Maxis development team have knowledge that cannot be readily regained.

Aside from all that, decompiling to create a derivative product would be illegal, and EA has a very capable and proactive legal department.   

I'm just pointing IMO a glaring thing that you missed which made projects like OpenTDD possible (Yes I know they used clean-room reverse engineering just like ReactOS is doing now but windows is also primarily written in C but the point I'm making is that C++ is a lot more complex and harder to do any reverse engineering legal or not and rewrite then C based programs)  

[Naomi57 2,985]

2 hours ago, Leo -- said:

the point I'm making is that C++ is a lot more complex and harder to do any reverse engineering legal or not and rewrite then C based programs)  

Agreed.  I was just making the point that reverse engineering is not some kind of miracle silver bullet, even ignoring the legal issues.

In my own case, I was one of those "members of the original development team" of a Windows application suite, back in the late 90's, transferring my employment from defunct software house to multinational corporate client.  The legal issues were covered by an escrow agreement, but the multinational corporation would have been backed into a nasty corner without me and my colleagues coming across to help them out.

I really don't like to see good projects die, so I helped them out and finished the project, before moving on to greener pastures.

Going over decompiled source code, is like fossicking for gold.  Occasionally you strike a rich vein of reusable algorithm or data definition, but most of it is thrown out.  In terms of SC4 functionality, it's probably quicker and more effective to fossick through existing open source for pieces to reuse or re-purpose, rather using decompiled bespoke source code that is 20 years old.  One fruitful area would be GIS systems and there's likely to be dozens of route-finding open source projects lying around somewhere, some of them even related to real-world city planning or logistics applications.

There's surely a whole heap of 3D modelling open source out there, too.  Attracting programming talent and modelling talent to an OpenCities project would be one of the key factors in deciding which open source projects to choose.

Fossicking for these open source pieces, would be one of the first steps to a real OpenCities project.  Much more fruitful starting place than the old Maxis source code.  An OpenCities project might even attract open source talent focused on real-life simulation rather than gaming?

[cydonianmystery 96]

Sorry for the necro, but apparently a small project exploring an OpenSC4 is being looked at on Reddit:

 

[Ulisse Wolf 4,406]

43 minutes ago, cydonianmystery said:

Sorry for the necro, but apparently a small project exploring an OpenSC4 is being looked at on Reddit:

 

I am very perplexed about this project that is ongoing. 

How do you implement the game?
Will the mods be compatible with this version?
Does it include those features hidden in the source code?

But the most important question is this

What is EA's reaction to this project?

[Cyclone Boom 16,854]

Indeed the quandary I always have with any such project, is whether upon completion it could even potentially be an advancement SC4 is as we know. Or whether the aim is to be an alternative offering which could enhance the gameplay in certain ways. Certainly from the modding side, it would need the DBPF framework implementing which is the proprietary structure that EA/Maxis developed. I know many of us are content with how SC4 is and the creative possibilities it still provides.

To me it seems logical any advancement on SC4's capabilities would add to what we have already, as opposed to building something entirely new from the ground up. That otherwise seems more like a new game entirely, inspired by the ideals we're used to. So there certainly are unknowns over anything like this.

Still those who do wish to explore such projects and have the motivation, all the very best with whatever they come up with.

[TogaMasterJohn 703]

At first when I read the title of "OpenSC4", I was wondering if someone was working on a clean-room implementation of SimCity 4 (which would be neat). I was entirely wrong! 

I wish them the best of luck, and I'll try to keep an eye on the project as well. I'm not familiar with GDScript, so it'll be interesting to see what they'll be able to pull off.

[wheeeze 90]

Hi,

I'm the OP from Reddit. Despite nearly two decades of playing SimCity 4, this is the first time I use this forum, so please excuse me if I'm breaking any rule.

@ulisse, the answers to your questions:

Quote

How do you implement the game?

Within Godot. First thing I'm doing is loading and interpreting the game files. For more technical details, the answer is, I'm not sure yet. I'm still discovering how the files are arranged in SimCity 4, the limitations of GDscript, and adjusting the code based on that. Getting a functional region view is the first goal/milestone, it should not be too hard. For the city view/actual gameplay, Godot is powerful enough to take care of UI and rendering with as little hassle as possible. The automata simulation will most likely require me (or someone else brave enough) to use Godot's capabilities to run C++. I don't reject using another language, it's just that C++ is the one I happen to be the most comfortable with, in the realm of "fast" languages that is. If, by then, I feel confident with Rust, or find that C is a better target, or C# is good enough, then so be it.

Quote

Will the mods be compatible with this version?

I wish it to be so, but it is low on my bucket list, and I'm not quite familiar yet with the technical aspects of modding. If, as I understand it, modding consists of DBPF files, there is no reason they shouldn't be compatibles. Patches that alter the executable cannot be compatible. I'd also like a modding API to exist, that could give a lot of freedom on how the simulation works, but it's not an immediate concern.

Quote

Does it include those features hidden in the source code?

They're hidden enough that I don't know that, so they can't be included

Quote

What is EA's reaction to this project?

Hopefuly none negative. I'm rather positive the law is on my side, but I'm not a lawyer, and even if I were one, my budget and ability to fight in court is somewhat lower than EA's. If they want to crush me, they'll do it. I hope they stay either neutral or lend a hand (let a man dream).

 

@Cyclone Boom

Quote

the quandary I always have with any such project, is whether upon completion it could even potentially be an advancement SC4 is as we know

First goal is a 1:1 implementation. Altering the gameplay, implementing what I think are "improvements" (besides obvious technical ones like higher resolutions support) is not a goal of the project in itself. An exception to this, I'd like to add a multiplayer capability in the form of a shared region where everyone works on their own city. I remember such a tool was made several years ago (around 2017 if my memory is right). But, again, this is long term and not my focus right now.

@TogaMasterJohn

Quote

I was wondering if someone was working on a clean-room implementation of SimCity 4 (which would be neat). I was entirely wrong!

I'm absolutely not an expert on the topic, but my understanding was that a clean-room implementation meant the developer, who hasn't seen the sources, worked from a specification written by someone who obtained knowledge of the proprietary techniques and sources. Since I work from the SC4 wiki specs and have never even decompiled the game, I think I'm, for now, OK on that aspect. After all, there are already many tools out there who are distributed by people other than EA, and which work on their proprietary formats. OpenSC4, in a way, would just be such a tool. Feel free to correct me if there's something I misunderstood about the concept of clean-room development. Although, as I said above, if EA thinks this project must not go on, it likely won't go on, regardless of whether the law is theoretically on my side or not.

 

Regards and thank you all for the support you've expressed

[TogaMasterJohn 703]

I'm not a lawyer either, but I don't see how EA could kill the project. There doesn't seem to be any violations of the EULA at all, and the user would need to have a copy of the game (most likely the fully patched game) for the assets. If anything, it could potentially make them more money. But that's another rabbit hole and a whole other deeper conversation.

I hope to start learning C# this summer as I find it interesting and if I like the language, that might be the next college class or two that I will take. If you decide to start using C# for anything, I would for sure love to take a look at it even more! Perhaps then I would be able to help out, too. 

If modding would be easily feasible, 'quality of life' improvements/standards like NAM and SPAM would be neat to see, along with any fixes for game bugs or annoyances. Then again, I guess a lot of the bugs could potentially be fixed in this project. 

[CorinaMarie 29,578]

4 hours ago, wheeeze said:

... this is the first time I use this forum, so please excuse me if I'm breaking any rule.

No rules broken at all and welcome to posting.

Feel free to post any linkys to your Reddit stuff if that's helpful for anything you discuss here.

And just a quick how to tip you can use: When you want to quote someone, you can highlight the selected parts with your mouse and most times within a moment it'll pop up a little Quote This thinger. Click that and it adds in who it was you are quoting with the time stamp too and that also beeps (sends a notification to them) that you have quoted them once you submit the reply.

Note: Sometimes that Quote This doesn't pop up and I find re-highlighting the text usually solves it. Another little bug is sometime the reply editor box won't open and for that a normal refresh of the page fixes it.

[wheeeze 90]

3 hours ago, CorinaMarie said:

you can highlight the selected parts with your mouse and most times within a moment it'll pop up a little Quote This

It works, thank you

 

[nos.17 5,112]

On 5/12/2022 at 4:55 PM, wheeeze said:

First thing I'm doing is loading and interpreting the game files. For more technical details, the answer is, I'm not sure yet. I'm still discovering how the files are arranged in SimCity 4, the limitations of GDscript, and adjusting the code based on that.

Looks promising! I'll keep tabs on the project, and if you need any advice on the structure of DBPF files, I'll be happy to offer up what I know.

[Pakistani890 122]

hey @wheeeze is godot able to handle all of this SIMULATION, why not unreal or unity

[wheeeze 90]

On 5/25/2022 at 8:04 PM, Pakistani890 said:

hey @wheeeze is godot able to handle all of this SIMULATION, why not unreal or unity

Unreal and Unity aren't free software

[Pakistani890 122]

5 hours ago, wheeeze said:

Unreal and Unity aren't free software

THEY ARE

[wheeeze 90]

55 minutes ago, Pakistani890 said:

THEY ARE

They're free as in free beer, but not free as in freedom, i.e., they're proprietary. Besides, they have good graphic capabilities, but to my knowledge, they're no better than anything else for the simulation part.

[Pepe The Great 227]

8 hours ago, wheeeze said:

They're free as in free beer, but not free as in freedom, i.e., they're proprietary. Besides, they have good graphic capabilities, but to my knowledge, they're no better than anything else for the simulation part.

wow this is very exciting, please let me know when this project will start its "go fund me" account. i would love to pitch in to have a SimCity 4 that can take full advantage of the model computer hardware. thanks for being here and giving us details.

[rsc204 21,806]

On 25/05/2022 at 8:04 PM, Pakistani890 said:

why not unreal or unity

Because as game engines they are woefully bad for supporting this type of game (a simulation).

[Pakistani890 122]

4 hours ago, rsc204 said:

Because as game engines they are woefully bad for supporting this type of game (a simulation)

Ummm why?? isn't city skylines made in unity and highrise city to

[rsc204 21,806]

14 hours ago, Pakistani890 said:

Ummm why?? isn't city skylines made in unity and highrise city to

Yes Cities Skylines does use Unity and it's worse off for having done so, it's extremely inefficient since the engine wasn't designed to handle complex simulations. In other words you need a much more powerful PC than you would if it were optimised and there are other restrictions it's use brings too.

Without getting into a load of technical stuff, game engines are usually built for a specific type or genre of game, because the code to make a Racing Sim would be very different to that needed to make a First Person Shooter. A game engine is essentially reusable code that you can licence, in other words a developer doesn't need to write their own, instead they can concentrate of creating the assets and other design parts. Whilst you can modify the game engine or otherwise add new APIs to handle new functions, the game engine just wasn't built with these genres in mind and as such a devoid of the kind of optimisation a proper dedicated game engine has.

Consider GlassBox, the engine made for SimCity (2013), graphically speaking the game looks way better than CS and it runs on much lesser hardware just fine using less resources. Because a dedicated game engine will always beat a generic one, the problem is that it's very expensive to make your own engine, hence the use of imperfect solutions. Don't get me wrong, the Unreal engine is the perfect tool to make an FPS, because at it's core that's what it was created to be, it also works fine for many action/adventure type games, usually with a focus on shooting enemies.

Think of this like using a Truck as your daily-use vehicle, sure in many ways it's like a car and can probably do all the things you'd need, it's just the wrong tool for the job and a woefully inefficient method of getting around.

[TheMurderousCricket 5,048]

Hopes for OpenSC4 would be much higher if Maxis was the only stakeholder apart from us. They have already proved many times that they are indeed on our side and were willing to help as much as they could for as long as it wasn't in contradiction with some legal mumbo-jumbo.

EA is a publisher. This status causes problems and they don't care. Nonetheless, I consider EA to still be a little bit more friendly than other publishers out there who are simply scum...

On a humorous note, EA encourages us to *whispers* "Challenge everything", so why not give OpenSC4 a try...? 

[Pakistani890 122]

8 hours ago, rsc204 said:

Yes Cities Skylines does use Unity and it's worse off for having done so, it's extremely inefficient since the engine wasn't designed to handle complex simulations. In other words you need a much more powerful PC than you would if it were optimised and there are other restrictions it's use brings too.

Without getting into a load of technical stuff, game engines are usually built for a specific type or genre of game, because the code to make a Racing Sim would be very different to that needed to make a First Person Shooter. A game engine is essentially reusable code that you can licence, in other words a developer doesn't need to write their own, instead they can concentrate of creating the assets and other design parts. Whilst you can modify the game engine or otherwise add new APIs to handle new functions, the game engine just wasn't built with these genres in mind and as such a devoid of the kind of optimisation a proper dedicated game engine has.

Consider GlassBox, the engine made for SimCity (2013), graphically speaking the game looks way better than CS and it runs on much lesser hardware just fine using less resources. Because a dedicated game engine will always beat a generic one, the problem is that it's very expensive to make your own engine, hence the use of imperfect solutions. Don't get me wrong, the Unreal engine is the perfect tool to make an FPS, because at it's core that's what it was created to be, it also works fine for many action/adventure type games, usually with a focus on shooting enemies.

Think of this like using a Truck as your daily-use vehicle, sure in many ways it's like a car and can probably do all the things you'd need, it's just the wrong tool for the job and a woefully inefficient method of getting around.

hmm ok that explanation was life changing and what about unreal and also the C:S dev didnt even think of optimisng the game what about that and one more thing i am very supportive of Open sc4 i am just giving suggestions that might help or ease the process