1OpenTTDOpenTTD Cuba Scenario

In this post I provide documentation and support for my OpenTTD Cuba scenario

So, I’ve finally got back to OpenTTD and got back to scenario making.  I’ve spent a few weeks tinkering with different ways of generating the best heightmaps and have settled on a cubic function.  The basic problem is that OpenTTD only supports 16 different heights.  If you scale this evenly to your overall terrain then you get rather large increases in height before you actually go up a level in the OpenTTD world (e.g. – the UK’s highest point is Ben Nevis at 1344m.  Divided into 15 this is 90m per height increase).  This doesn’t produce a terrain that seems realistic (e.g. the pennines between Manchester and Sheffield are only 5 OpenTTD tiles high at maximum – it is easy to build a rail line right over this!).  The first solution is to use more appropriate height increases for each level, but this results in the top of all hills and mountains appearing “carved off” which is far from ideal.  I realised that as you get highest, the amount of land at that height decreases dramatically.  E.g. there is a lot of land between 0m and 50m above sea-level, but very very little between 1000m above and 1344m above.  Giving each height equal importance isn’t necesary.  I realised I could use a scaling factor which has the effect of reducing this sharp peak effect.  Basically what this means is that the heightmap is more sensitive to real height changes at low altitudes – so we get lots of detail at low altitude where most of the land actually is.  We still take account of higher peaks though and don’t carve the top off anything.

The real heights which results in an OpenTTD level increase in the cuba heightmap are

20, 70, 120, 170, 230, 290, 340, 410, 490, 570, 670, 780, 900, 1090, 1610

As you can see, 5 levels cover the first 250m real meters, 4 levels cover the next 250m, 4 levels cover another 500m and the final 2 levels are for heights above 1km.  From both a visual and gameplay perspective this seems to provide a optimal solution

Hopefully feedback for this scenario will be positive (and so will my own playing of it) – if so I will be use this improved system for a new release of the UK scenario


Release 1.00, 3rd April 2012

  • Licence – CC-BY-NC-SA v3.0
  • Populations 1/250 of real figure
  • Uses enhanced (cube function) heightmap from SRTM4 data
  • Compatible with OpenTTD 1.1.0 upwards
  • Requires ECS Vectors
  • No other NewGRFs enabled
  • Approx 80 cities
  • 4 Banks – in the only towns with big enough populations
  • 5 Breweries – from RateBeer.com
  • 0 coal mines
  • 3 Oil refineries – http://en.wikipedia.org/wiki/List_of_oil_refineries#Cuba (Havana, Cienfuegos, Santiago de Cuba)
  • 1 Oil rig – off north coast – http://www.bbc.co.uk/news/world-latin-america-16642286
  • 0 oil wells on land
  • 3 gold mines (placetas, santiago de cuba, Pinar del Rio)
  • 1 steel mill (las tunas)
  • 0 iron mines
  • lots of farms / fruit plantations – 30% of land area used and 20% of population work in this
  • 3 shipyards (havana, santiago de cuba, guantanamo)
  • 2 tourist centers – varadero beach, cienfuegos
  • 0 vehicle factories
  • 0 power stations
  • Animal Farms?
  • Chemical Plants?
  • Factories?
  • Fishing grounds?
  • Forest?
  • Furniture factories?
  • Glass works?
  • Paper Mills?
  • Printing Works?
  • Sand Pits?
  • Sawmills?
  • Textile Mills?
  • Tinning factories?
  • Water Supplies?

Please send us your thoughts by commenting below! If you would like to subscribe please use the subscribe link on the menu at the top right. You can also share this with your friends by using the social links below. Cheers.

Leave a Reply

One Comment