[OTTD] London to Tokyo Scenario

[cnww]

I'm working on a large-map scenario suitable for a long-distance network that can really take advantage of some of the fancy new maglev and vacuum tube GRFs to move passengers at ludicrous speed over vast distances.

This nascent scenario is based on real-world topography obtained using Bother, which I then rotated (Japan is significantly further south than the UK) and re-scaled in GIMP.

To keep memory requirements down, I'm using an 8:1 aspect ratio (so the map is a long rectangle) with England & Wales at one end and Honshu, Shikoku and Kyushu at the other (along with some of Hokkaido, Scotland and Ireland).

Sample screenshot

screenshot1.jpg (15.16 KiB) Viewed 9136 times

At the moment, I have completed the initial raw heightmap, and created some PNG files suitable for use in OpenTTD. I'm not yet ready to upload them to BaNaNaS (they need manual retouching to fix up a few shorelines, particularly around the English Channel. I may even need to re-generate them from the raw data), but they are in a playable state, and even if I stall-out here they may be of help to somebody else who would like to create a similar scenario.

I've decided to do this in a public forum partly for feedback, but mostly in the hope that it means I'll actually finish the darn thing instead of iterating through 15 versions and then never actually uploading it (which has happened many, many times in the past 28 years). I intend to update this post as I make progress.

The current heightmaps are all scaled versions of the same diagonal strip from the 188563x34993 pixel image that Bother produced. Non-coastal lakes and rivers are absent, as there is no way to encode river tiles into a heightmap, and I didn't want to create a huge crater down to sea level for every lake in the Himalayas. My current plan is to add noteworthy lakes and rivers manually in the scenario editor, or possibly try to implement the extended heightmap concept using a game script.

They are available in 32768x4096, 16384x2048 and 8192x1024. I believe (please correct me if I'm mistaken) that you need a somewhat recent version of OpenTTD to load any of these, and you may need JGRPP to use the 32k map (at some point I intend to install vanilla OpenTTD, test this, and update this text accordingly).

If anybody has actually read this far, I'd love to have some feedback on your preferred map size, climate, and industry set: My current plan is for my first version of the actual scenario to use the 8 or 16k map and ECS vectors with a 1920 start and roughly 200 towns. Ideally, I'd love to find good FIRS variant that requires most or all resources to be delivered in order for factories to produce output (reaching the end-goal of a Steeltown game with just Sand->Engines->Vehicles is both insane and anticlimactic) and would support a 19'th-century start date (with industry introduction dates), but I don't think any such beast exists.

License information: The original terrain data, as a product of a US government agency, is in the public domain. The cleaned-up and modified data used by Bother comes from CIAT, and is free for non-commercial use (see spoiler for exact text). My own contributions are released here under a CC BY-SA license.

CIAT data permission blurb:

[+] Spoiler

Users are prohibited from any commercial, non-free resale, or redistribution
without explicit written permission from CIAT. Users should acknowledge CIAT as
the source used in the creation of any reports, publications, new data sets,
derived products, or services resulting from the use of this data set. CIAT also
request reprints of any publications and notification of any redistributing
efforts. For commercial access to the data, send requests to Andy Jarvis
([personal email address removed]).

Suggested options for Play Heightmap or Scenario Editor:
- Sub-arctic climate to support snowline
- Clockwise rotation to align (sort-of) with compass
- Max height 128

32k x 4k height map attached to the next post, as for some reason I couldn't attach it to this message. It's well under the 25MiB limit, but my attempts to attach it kept failing with no error message.

[cnww]

32k x 4k height map, version 1. No idea why I couldn't attach this to the first post...

[Argus]

XIS might be suitable, it's a bit more demanding, I get the impression.

[JohnFranklin523]

Glad to see Shandong Province is there

[cnww]

XIS might be suitable, it's a bit more demanding, I get the impression.

XIS is (AFAIR) an everything-all-at-once setup rather than having industry introduction dates, which causes problems in 19'th century starts. Its "official successor" AXIS is certainly something I plan to try. Do you know of any particular problems with AXIS that would make XIS preferable?

I like playing with sailing ships and early steam engines, but honestly if I'm making London<->Tokyo express passenger service the overarching goal of this scenario, an 1820s start date isn't ideal even if I have an industry set that supports it well: The high-speed trains you need to connect them aren't available until the late-20'th or early 21'st century depending on the vehicle sets. That may mean that the way to go here is to produce an (A)XIS version with a start date in the 1920-1960 range (depending on how much I want to catch the end of the age of steam).

The other option is to find, or make, an industry set that starts with a simple 19'th century economy and then builds a modern economy by gradually introducing new industries (modern metal production, industrial fertilizer, petrochemical chains, eventually waste and recycling). I've dabbled a bit in NML before, and these days there are loads of open source examples and sprites to use, so it all sounds doable, but I've only got so much time to spend.

As soon as I finish fixing up coastlines on the 16k map, I think I will start a game with AXIS and random towns and see how it goes.

Glad to see Shandong Province is there

I aim to please . Or, more honestly, I included it because of its location, but I will do my best to accurately represent it in the scenario(s).

[Argus]

I didn't notice any significant problems in AXIS when starting in the 19th century. If there are any, just notify the author, after all it's still an alpha version.
Another thing is the lags, but they only appear when the experimental features are turned on.
On the other hand, XIS, even if it also suffers from the Firs paradox, after all, according to my observation, it requires at least a more regular and constant supply, and when only one resource is supplied, it often produces only the minimum.
By the way, food is produced in ECS even without any resources, do you mind? At the same time, Firs does not have it. I even got to the point where I only had to bring a single ingredient if there was a lot of it.
Farms don't produce anything in the winter season, that's OK, but they don't even in the tropics!
I probably don't even need to remind you of the paradox that a reservoir can only be built in villages with more than 500 inhabitants, when in the tropics every small village has a well.
ECS suffers from far more ailments and illogicalities than Firs.

[Kruemelchen]

This scenario looks great!

It's something I would definitely play! The possibility to play a "silk road" scenario alone opens up so many possibilities in transporting goods even with an early game start (I thought about coding camel caravans as RVs sometime when I come to it, and just thinking about how they would shatter any hope to transport wares over months through deserts and mountains even barely profitable on that scale is hilarious )

On a side note; is there a reason you strove away from including south Siberia (for a trans-Siberian railroad option)?

Other than that, I'm looking forward playing this with lakes/rivers, cities and the main industries set up where they historically/geographically belong

[cnww]

This scenario looks great!

It's something I would definitely play! The possibility to play a "silk road" scenario alone opens up so many possibilities in transporting goods even with an early game start (I thought about coding camel caravans as RVs sometime when I come to it, and just thinking about how they would shatter any hope to transport wares over months through deserts and mountains even barely profitable on that scale is hilarious )

On a side note; is there a reason you strove away from including south Siberia (for a trans-Siberian railroad option)?

Other than that, I'm looking forward playing this with lakes/rivers, cities and the main industries set up where they historically/geographically belong

I'm sorry to say that this has been sitting idle for a while, and it's likely to be a while longer before I make significant progress. I lost my original momentum as a result of being distracted by a pair of hurricanes last autumn, and I'm just now getting back into OpenTTD again.

Siberia is missing because of simple geometry: The map is a narrow strip from the southern UK to central Japan. On the (default) map projection I used, Siberia just happens to be too far north. I'd need to double (or quadruple) the total map area if I wanted to include it, and I haven't really got the hardware for that.

Ideally the map would include both the Siberian rail route and the ocean route (via the Suez canal and passing between India and Sri Lanka), but the map would be absolutely huge (I'm guessing 8192x16384 to do it properly), and there is not (yet) any way to have Artic and Tropical climate areas on the same map (IMHO a combined Arctic/Temperate/Tropic climate, with to restrict industry placement based on climate and topography would be lovely, but it's wayyy beyond the scope of what I'm doing here).

What I may do, as I need to recreate the heightmaps anway (see below), is produce two final scenarios: One with a distorted map connecting Scotland to northern Japan via Siberia using the Alpine climate, and a second connecting England to Australia via the middle east, and Indonesia (the early game would be mainly ships, but with map the angles adjusted correctly, players could connect Italy->Egypt, Oman->Indonesia and Indonesia->Australia with huge undersea tunnels in the late game).

---

p.s. As anybody who downloaded my heightmaps will have noticed, there are some major problems with elevation and coastlines in the current files.

I spent a few hours manually fixing up part of the 16k map, and I have a version where England and Wales coastlines are pretty-good, but when I got to the Antwerp/Rotterdam area I realized that the entire thing was going to have to be redrawn by hand (and the Himalayas may be worse) and decided to stop. I do not think will live long enough to fix it all by hand.

I do still plan to finish the scenario(s), but I think I need to start by creating new heightmaps from a data source that's better at placing coastlines correctly (possibly using the same software with different options, I have some experimenting to do).

Of course in the meantime, if you want to play a game with messed-up coastlines, by all means feel free. You can pretend that it's an alternate earth where there has been some weird tectonic activity, if that helps.

[Kruemelchen]

I wish you good luck for your endeavour! Manually editing is a nightmare, I've halted working on much smaller scenario maps because of that

I'm definitely eager to play a game with your map!

And thank you for your explanation

To me, it seems having both north and south "silk route" in one map is a bit far-fetched, not to speak of climate difficulties. I was just wondering whether it would be feasible to have both the route over Afghanistan and over Mongolia in one map. But when thinking about it, it's actually quite nice to stick to only one of the possible routes per scenario