Quote:
Suicide Junkie said:
The basic armors will have to have two dimensions at least though... the light->heavy->(advanced armor tech area) and the beefing up of armor components in each class.
Though I suppose the light->heavy won't look much like a dimension of a tech grid after the operation.
|
I have been playing around with a version that eliminates the "Armor manufacture" dimension, and it does seem to work a lot better. The AI does not stop and refit ships to gain the advantages of Heavy Armor Class 6C versus 6B, which is good. The steps from class 5 to class 6 are far enough apart that ships can conduct missions before being recalled for refits.
I have finished the basic armaments that everyone has, and tried a game using just them (no specialization). The AI was building much better ships. I then added a few advanced weapons for the Minbari, Narn, EA, and Centauri, and that game is in progress.
I also completely abandoned the (at least) four-times- modified AI_Construction_Vehicle file, and inserted the original one from SE4 1.91, and lo and behold, the AI is kicking *** again. In my current game, I have managed to get back to 11th place after 100 turns (after starting in 4th and slipping as low as 17th out of 20). This is with medium AI player difficulty and medium AI advantage - I don't think you can make the game competitive at less, and want to reserve the higher settings for players wanting more challenge.
Some of the things I have done, and am still testing, are:
1. I have generally decreased RoF and increased range. Missiles, for instance, now generally have a reload time of 3 turns, dropping to 2 turns with advanced technologies (vice 2 turns dropping to 1), but have ranges of 10 with light missiles, 14 with medium missiles, and 18 with heavy missiles. This should get rid of some of the real jumps that the EA, for instance, used to get with advanced ballistic weapons technology. It alos eliminates tradeoffs between payload and range. The tradeoff is now between hill space and range. Lasers are now longer-ranged but less potent.
2. I have toyed with the whole concept of what it means to get more advanced versions of the same weapon, and decided that the tonnage cost and structure benefit will remain constant over the life cycle of the weapon (vice decreasing cost and increasing structure as in the Val version). The assumption here is that the basic components remain the same, but power systems become more efficient, maybe ammo improves, and that further experience with the weapons allow them to be used more efficiently. This will have a beneficial impact on the total hull/component change that occurs with a new version of an existing weapon (as you will not add more armor to compensate for the reduced tonnage of more advanced weapons). If you want to see hull space saved, you will need to get an improved or advanced version of the same weapon type.
3. I have inserted "improved" as a new weapons type intermediate between the basic weapon and the advanced version of the weapon. Sometimes "improved" means a higher ROF, sometimes it means a more powerful volley with an existing ROF, but it almost always means a better range and less loss over range.
I also made a change in the gain per level for weapons. I was using the Timstone method of maybe a year ago, which was that the damage at range 1 for each successive mod was equal to the damage at range 1 for the level 1 weapon, multiplied by 1 1 /x, where X was 11-level (so that at level 2, for instance, range 1 damage was 1 1 /9 of the damage at range 1 for the level 1 weapon). The problem with this approach was that the level 9 weapons was 50% (1 11/9= 1.5)better than the level 1 weapon, but the level 10 weapon was 100% (1 (11-10)=2.0) better. Half of the change between level 1 and level 10 occurred between levels 9 and 10!
Now, the increases in levels should not be arithmetic (i.e. constant change per level) because the cost is not constant. I ended up splitting the difference, and using the average between the Timstone method (which is kida cool, actually) and the constant change method (which is easy to calculate, and so appealed to me!
) I like the product, but wanted the purists to be warned that change was a'coming. It isn't possible to scale the increased damage to the research cost, because the research cost delta is changeable in game setup.
The real "character change' that the mod still needs is making the facilities race-specific. I have done nothing whatever in this regard - every facility is still generic. Generic facilities work, and some generic-ness is necessary to keep the AI able to compete, but I am looking for ideas in two main areas:
1. Differences between homeworld hubs that don't duplicate racial differences; and
2. System modification facilities that we can both use as human players and get the AI to use (or else facilities that it does not hurt the AI if it does not use them).
Any thoughts on this are welcome. I hope to start putting the various data files here in the next couple of weeks, with explanations of each, so we can discuss them separately and you can offer suggestions on each.
A big thanks to Timstone for his work on the weapons (and a prayer that he will send an unlocked version of his weapons spreadsheet soon), to pathfinder for creating a version of the mod that worked (and for helping me understand the AI files), to IF and SJ for the whole armor/sensor/ECM concept that adds so much flavor to the game (I am not sure anymore who originally proposed what), and, of course, Val for getting this started. If I left anyone's contribution out, feel free to say so.
Faith manages.
Asteroids Champion!
Re: Babylon 5 Mod
Linear Mode
