Transport Tycoon Forums

I've always felt that the OpenTTD scoring system could do with modernising; most of the objectives are so easy that any company will acheive them, and the minimum 10k vehicle profit often isn't a fun challenge to acheive as it means selling all my heritage trains, lots of road vehicles, and any unlucky boats that only make a profit every other year. With the new social integration features, I started to think about the score a bit more, since I feel like these new features could be used to setup a proper set of leaderboards for the game, with a new scoring system. The existing system wouldn't work well for a public leaderboard since it would just be a sea of 1000s and 900s for the first several thousand entries, so a replacement system would need to have no maximum cap for most categories.

I think a lot of the existing categories are good, (e.g. types of cargo, number of recently serviced stations) so by removing the maximum scores for them and just making it so every station or cargo type increases the final score by a set amount. I would probably change the minimum profit and maximum profit categories into an average profit category, and I think it could be worth adding a couple of new categories to act as score multipliers, e.g. Percentage of industries that are serviced, percentage of towns that are serviced, average percentage of cargo carried per source, and average journey delay. (due to inefficient junctions and traffic) The other main factor with a leaderboard is you'd need some way of "normalising" the score, if the score has no cap someone could just make a ludicruosly huge map and give themselves 400 years to build a vast empire. The time taken, map size and number of towns and industries would need to be multipliers as well, e.g. if you have 200 years instead of the default 100, your score is reduced by 50% (this value might need to be adjusted after testing). A lot of settings would need to be compensated for as well, e.g. the cargo delivered category should be reduced to 1/5th of the normal score if you set cargo production to 500% (aside from compensating for day length), and time taken should be modified based on what you set the day length to.

I can see how this would be a complex thing to impliment but I think it would be a really cool mechanic and would make a rather useless carry-over from the original TTD into a fun, competitive aspect. People often ask for ways to make OpenTTD feel more like a game and having a genuinely good scoring system where you can compete with other players or yourself to get a new highscore at the end of each game would be a good method of doing that.

The other idea I had is the "Goals" section of the title. In games such as Sid Meier's Railroads, each map has a set of goals, usually something like "Be the first to connect Pittsburgh and Philidephia by 1889" or "Transport 10,000 tons of munitions to Las Vegas before 1945" (I made these up as examples.), and I think it would be cool if the game (optionally of course) could examine a newly generated map, and choose a few goals in this style, (along with any other ideas that would work, e.g. "Connect X amount of Industries of X type", or "Have X amount of railway tiles") using things such as the size of cities, distance to travel, and the availability of cargos at that date to make the decisions. I can see how this could be difficult with different NewGRF combos, especially stuff like AIRS that only introduces some industries later on, but I'm sure it's possible. Even if it couldn't be done for randomly generated maps, it would be great to give scenario developers the ability to write in goals manually so people who play the scenario can try to achieve them. I feel like some of this has been done with game scripts, but I'm not really an expert on that.

Good news: Version 13 introduced Scriptable Company League tables as a feature. So you can write a Game Script that replaces the default score calculation and company ratings with whatever you want.

Yeah, that's not what he wanted. Can't solve a global issue ( outdated system) with a grf. If you want to compare companies in general can't expect to everyone have the same grfs. And I agree that the company performance score is outdated and only meant to the og 256*256 maps. Idk, but have a suspension it's not the most used map size (but even if it is).

It's meaningless to compare companies "in general". In city-builder they need to be compared by town population, in money goal by bank balance, in cargo goal by cargo amount. As for "maximize some random junk score" goal current performance rating works just as well. At least it has an excuse of being true to the original game.

_dp_ wrote: 30 Apr 2024 20:04 It's meaningless to compare companies "in general". In city-builder they need to be compared by town population, in money goal by bank balance, in cargo goal by cargo amount. As for "maximize some random junk score" goal current performance rating works just as well. At least it has an excuse of being true to the original game.

These are the kind of factors that I think should go into the score. In 2050, (or whenever set), your company should be scored on elements like that. But specifically I think it should have no cap, allowing you to get a higher score by transporting more or improving rating, and so on.

jfs wrote: 29 Apr 2024 06:35 Good news: Version 13 introduced Scriptable Company League tables as a feature. So you can write a Game Script that replaces the default score calculation and company ratings with whatever you want.

This all goes well over my head but if it means it's possible to make custom scoring parameters that's really cool, and I hope someone more skilled can figure that out.

Gilgames wrote: 30 Apr 2024 12:03 Yeah, that's not what he wanted. Can't solve a global issue ( outdated system) with a grf. If you want to compare companies in general can't expect to everyone have the same grfs. And I agree that the company performance score is outdated and only meant to the og 256*256 maps. Idk, but have a suspension it's not the most used map size (but even if it is).

it does seem like it would do what I wanted but I don't have the skills to reprogram company league tables myself. I still think it would be nice to have it built in to the base game though if a leaderboard was to be made, but that's kinda a longshot anyway, an improved scoring game script is a good idea.

JustMoreMemes wrote: 02 May 2024 18:23 These are the kind of factors that I think should go into the score. In 2050, (or whenever set), your company should be scored on elements like that. But specifically I think it should have no cap, allowing you to get a higher score by transporting more or improving rating, and so on.

These kinds of factors often don't even exist in a vanilla game. It has no town claiming so no population. No victory points, no vehicles cargo (Steeltown end cargo), no machinery (Yeti end cargo) and so on. Companies should be scored on the progress toward the game goal. Vanilla game has no goal so should have no score. It makes no sense and only serves to confuse players that are forced into doing things they don't like (e.g. upping min profit) to be on top of some random leader board. But if someone want to make a game with a goal of maximizing multi-factor score that's exactly what mods are for.

It is a truth universally acknowledged, that the metrics which make up the performance rating make no sense whatsoever. The most offending one is the "Min. profit", because it punishes players which use transfer vehicles.

It is certainly possible to design a more reasonable set of metrics, which at the same time don't depend on things that only exist in add-ons. For example, I like the metrics used by this game script.

_dp_ wrote: 02 May 2024 19:59 Vanilla game has no goal so should have no score.

This is one of the common objections to OTTD - this game has no purpose. That's why many players see it in this table, but well, the current scoring method is quite limited mainly by the minimum profit criterion. In addition, for someone who already knows this game a little, achieving the maximum score becomes quite a boring formality.

Scripted company ratings are not a solution here, because they allow you to compare companies only in the current game, without the possibility of comparing the best results.

I don't know if minor changes to the current rating system would be acceptable. Perhaps the solution would be a new improved rating system available under a toggle in the best score window.