Forum Replies Created
-
Posts
-
I think the misinterpretation is coming from the idea that industries should operate at an imagined “full-capacity” simply because you supply them with enough materials.
As far as I’m aware, Train Fever works rather differently. An industry won’t produce more goods than there is demand for those goods. That demand can be fulfilled entirely by a single industry, which will grow much larger over time to accomodate it, or the demand can be fulfilled by several different industries, in which case they’ll balance the demand between them.
The issue you’re likely coming across is that if you’re completely fulfilling demand from a single industry already, and you link a second industry into the network to cover that same demand, the load will balance between them… you won’t just force them to produce twice the output even if the demand for that output doesn’t exist.
The main reason you’d connect up a second industry isn’t to somehow generate a mass of extra demand, but to help with the logistics of meeting the existing demand. You won’t make any more money directly, but you could stand to save a lot of money through added efficiency over time.
If you want to promote demand, you need to do the things that grow cities… supply passengers and goods… and then wait until they create the demand themselves.
As with anything in a simulated economy like Train Fever uses, you won’t see the results instantly… you’ll likely see losses at first. You have to wait and let the results play out over time.
I’ve wondered about the real-world precident for this. Not that Train Fever needs to be slavishly realistic of course, I’m just interested in what ways it could inform gameplay.
Are trucks on real-world freight train routes often reused for different goods in each direction, in order to maximize efficiency, or are factors like cross-contamination too big of an issue?
I’d hazard a guess that raw materials (coal/iron/oil… I guess lumber by virtue of needing specialised log-trucks) tend to stick to dedicated single-use containers, while packaged/dry goods probably end up in mixed-use trucks.
I suppose Train Fever tends to be fairly light on the packaged goods side… that’s all encapsulated as “goods”, so it sort of makes sense that the rest are dedicated use.
One thing I’m wondering about, but haven’t tried yet – I seem to recall loaded/empty truck weight is simulated by the game, in which case, might it be a valid endeavor to attach two different sets of trucks to the same train, one for each direction? Only one set would be heavy at a time, so it wouldn’t slow the train to a crawl, and it might make more effective use of the vehicle overall.
There is absolutely no reason to stop telling us what you would like to see in the game …
Just bear in mind that not everything will be possible, because the wishlist so far would be long enough to spend years on the implementation.
But I will make sure this one makes it on the list that we will consider for the next update!
Tom
Fantastic! Thanks muchly for the response! It’s my hope that these suggestions would not only improve the game for us existing fans, but in the case of expanded water options, would also yield appealing screenshots which would help bring in more of the attention it deserves.
I look forward to seeing what the future brings 🙂
Yes, I agree too, great ideas there @OIFEVT101
I’ve thought the same for a while now, but I’ve struggled to think of an alternative that would potentially make a more interesting system to the one there is currently.
Your suggestions actually sound like a great system that I’d really like to play with. Having industries tied to certain locations is an obvious enough idea, but I like the extension to that where certain secondary industries are tied to specific population sizes/types, so there’s some distinction and necessary interrelation between larger and smaller towns.
Having particular manufactured goods have different effects as the years progress is also a brilliant extension of the idea. That would really give a feeling of progression, and the NEED to progress, rather than just the current continuous expansion, for the sole reason of getting more money, to expand more, to get more money. It loses its novelty quickly. Your way would give the game a much better sense of flow over time I think. You’d end up needing to retire lines to old industries as the world modernised.
In addition to that, passenger requirements could shift a similar way – tourism could increase over time, and I’ve seen someone suggest before that an international airport could spring up at a certain point through the game, or maybe expand from a small airstrip over the game, to the point where it became a massively-high-traffic passenger hub later on.
Of course, the ideal would be to have a vastly fleshed out random map generator *and* a full map editor with height-map importer.
At this point, I’m just trying to push for the simplest suggestions I can think of that would expand the map variety. Given just a couple of tweaks now, I’d enjoy playing the game for a good few months, and hope that more extensive options like those might come along later down the line.
Overall, I’d just like to see the game maintain some popularity, and not fade into obscurity and prove unfeasible for Urban Games to continue supporting… considering what a fantastic job I personally think they’ve already made with this game. All it really needs is a good bit of fleshing-out to keep it interesting.
…another monthly-ish bump for suggesting that what Train Fever really needs above all else is some map variety.
I’m thoroughly enjoying Cities Skylines, but I would much sooner be playing Train Fever… I’m more interested in engineering train lines than I am planning out cities… but well, having built a few train networks over generic rolling hills, I feel like I’ve essentially built them all. Skylines is proving to have rather more replayability at this point tinkering with traffic networks over endlessly customisable and varied maps. But I still find myself checking back here every couple of days for news of any big new features.
What we need are coastlines, options for long, thin maps, variable topology, variable number of towns/industries, etc, etc… basically, what OpenTTD has had for years.
Any word at all on this sort of thing? It feels a little like things are stalling since the release of the USA DLC, and even the DLC pack didn’t really give me any incentive to get back into the game, as the one and only thing that wasn’t touched upon was the map generation.
Ever increasing volumes of trains, cars, buildings, etc, all played out over the same generic maps really isn’t going to do anything to make the game more interesting.
It’s really frustrating me at this point, as at least some of these suggestions seem like almost absurdly simple things to implement. 64 x 4 km wouldn’t require any more memory than a 16 x 16… lower numbers of towns would come down to exposing a single number in the map generation options.
Come on… let’s at least have those two things.
Erm… “Bump”
I really think these suggestions are important!
As I suggested in the “Skylines is out” thread, I feel the terrain in Train Fever is both the core focus of the gameplay (literally, your opponent *is* the terrain), and also the game’s strongest asset.
Yet I feel like after the mammoth work that went into the terrain engine and map generation during development, post-release, it’s been ignored completely in favor of extra vehicle content and such.
At this point, I feel it’s absolutely vital to the future success of the game that Urban Games put the map generation front-and-center. It could add so much variety to a game that currently feels a little samey. Random map generation counts for little when every map is almost indistinguishable from the last, but random map generation has the potential to offer massive variation and customisation options.
Please, please, please, give us those options!
Post if you agree!
My point wasn’t that anyone is comparing the games like-for-like. It’s blindingly obvious that Train Fever is an entirely different game compared to Skylines or even Cities In Motion.
My point was that making comparisons about how well Skylines runs, and how bad Train Fever may perform on your system (which you most definitely did), is a somewhat unfair comparison seeing as Train Fever attempts to push the limits of terrain detail, while Skylines specifically opts to take a conservative approach, and only allow a very low-res, blobby base terrain.
I absolutely think it was the right way for the Skylines devs to go… it allowed them to concentrate on pushing the city scale/population a lot more than if they’d been sharing massive resources with the terrain engine.
But Train Fever has a different focus – at its core, its a game about “you-versus-the-terrain”… especially when it comes to trains, they needed an incredibly detailed terrain engine otherwise building train lines would have felt farcical and simplistic.
Urban Games simply took on a much heavier technical challenge when it comes to performance. It does still need work, sure, but there’s also no point in making an example of another game that took a far more safe and simplistic approach and holding it up as a beacon of how game development *should* be done. I’m very glad Urban took on the extreme challenge they did, given that I have a system powerful enough to handle it without too many hitches.
Before being too critical of Train Fever in comparison with Cities Skylines consider this:
Skylines has an 18x18km map size (10×10 playable, with unlock mod), which has an overall base grid resolution of 1080×1080. Divided down, that means each cell of the map is 16.666m. (When playing, it appears to selectively refine where you build to create minor embankments and such, but the base terrain is still limited to that)
Train Fever by comparison, has a 16x16km map size (fully playable), which has an overall base grid resolution of (I believe) 16000×16000. Divided down, that’s a cell size of 1m. This means that for every single grid cell in Skylines, Train Fever fits in roughly 278x as much terrain detail!
Considering that, I’m still staggered that Train Fever runs as well as it does! A 256-million-square piece of geometry is a hell of thing to get to display in realtime, and that’s not considering all the houses, roads, vehicles, people, etc. I know they use some very clever streaming LOD tech to not literally display that many polys, but all the detail is there to be looked at when you want to, and I very rarely spot any hint of LOD pop-in while it does it.
I still think Train Fever is an incredible achievement, and I suspect I’ll still be coming back to it long after I’ve had my fill of Cities Skylines 🙂
I haven’t found it breaks my old saves for the unmodded original game, I can still load my saves up just fine.
How many developers guarantee backward compatability for save games based on user-installed 3rd party mods? I really don’t think it’s their obligation to offer support for modifications you make to your game, even if they provide (and improve) a system designed to facilitate it conveniently.
I would think it’d be almost a foregone conclusion that later patches will very often break compatility with older mods. That’s really up to the mod-makers themselves to support (or not). How would developers even attempt to maintain compatibility with modifications they didn’t write?
I really want to see at least some broader map generation options, but a heightmap importer would be a great option, especially if it ends up being the only one.
I’m not sure people are quite factoring in how massive a heightmap this will need to be though 🙂 I *think* the terrain grid scale is 1m per square, which means that for a full-resolution heightmap to make a large map, you’ll need a 16384×16384 image! (a 4096×4096 one just for a small map)… add to that what electricmonk2k brings up above – that 8-bit images won’t contain enough colours to define height properly without making your map look like it has farming terraces everywhere – you’ll need a 16-bit 16k image (not so bad as a single-channel greyscale image… conventional RGB would need 48-bits-per-pixel)
There are plenty of free image editing and 3D rendering packages that fully support 16-bit-per-channel colour these days though. Being a vfx artist, I’d really like to try and build some procedural terrain inside SideFX Houdini and render it out to a 16k heightmap image.
In fact… yeah! Give us that heightmap importer Urban dudes! 🙂
I wouldn’t dream of suggesting they make it too much like TTD, as I really feel there’s a whole lot that’s too dated to really enjoy anymore in OpenTTD, and by-and-large, they’re things that I feel the Urban guys have absolutely nailed and brought screaming into the 21st century with Train Fever. I’m certainly appreciative of the deep level of procedural simulation going on, and I absolutely love the focus on mastering the terrain and realistic-feeling line engineering. It always bugged the hell out of me that TTD allowed trains to turn such tight corners 😛
It’s possible the reason you feel the USA map almost feels like a re-textured version of the original is because that is literally what it is. Try using the same seed for both settings… it generates an identical map in both cases, even down to the town layouts and names.
As you mention, all I’m hoping for to make this a complete and eternally-compelling game experience for my taste is the addition of a few ways to spice up the map generation. The important parts are all in place as far as I’m concerned, I just want to be able to take on specific and varied challenges. I want to have no choice but to thread a train line through that map-spanning mountain range… I want to tame that winding coastline… I want to build a massive viaduct across that deep, sheer-sided desert gorge! 🙂
– I’m not really seeing what you’re aiming to contribute here… nobody here is complaining about realism or the simulation/time scale not making sense. My own post was just a list of simple-to-implement suggestions to make map generation more varied and uniquely challenging and thus to make the game more replayable and interesting.
At the internal abstract scale the game is designed to function at (and at which I feel it feel it functions well and is enjoyably balanced), I would nevertheless enjoy having the option to have long, thin maps, as even the large maps currently feel a bit too restricted, while at the same time, feel like they lack focus, as the landscape and town distribution feels too evenly spread for my taste.
I’m saying nothing about how realistic or not the scale is. I’m simply considering what would make the game feel more fun. Transport Tycoon had almost a complete disregard for realism, and that didn’t stop it being fantastic back in the day. I feel Train Fever could easily feel as fun, if only there was a bit more variation.
It really doesn’t bother me in the slightest how far a horse walks in a day. Call them days, call them minutes… it’s just a game design abstraction so that it doesn’t literally take 150 years to play a single game 😛
Additionally, as minor considerations, I’d suggest the following might make for interesting options too:
- An ultra-mountainous option, or perhaps a “no rivers” option, as the current mountainous option still generates a lot of smooth flat lowlands that feel like they partially defeat the challenge.
- Perhaps an option for a “gradient” of terrain severity, so a map will be mountainous (/ultra-mountainous) on one side, and smoothly transition to flat/lowlands by the opposite map edge. This would allow for really interesting long-distance hill-climb routes to get from one side to the other. It could also be implemented as simply a straight continuous incline from one side to the other, added to the existing terrain types, so you could have an overall flat terrain, that still has a general gradual incline from one edge to the other. (Would combine wonderfully with a “coast” option as I mentioned above along the low-edge of the map)
I still think the ideal and completely straightforward way to address this issue is to allow for non-square map sizes.
A current large map, 16 x 16km if I recall correctly, covers 256km^2. A 4 x 64km map would cover the exact same total area, and would allow 64km-long routes to be built.
If we also had control over the number of towns generated on a map, we could customize our experience to allow for maps that forced you to make networks that covered long distances, rather than as it is currently where it’s always easy to just spider out from the middle along the river-banks.
