CNCOPTPLUS MANUAL

This document contains a description of the options and assumptions behind the CnC Base Simulator, aka CNCOPTPLUS.

The main goal of the program is to evaluate a layout for its use as a support base.

Evaluation of layouts to be used as a main offense base is not supported at this time.

This program only operates under the assumptions of the New (Balanced) Economy.

The program allows the user to enter a full cncopt of a base they would like to analyze (see restrictions below).

The user can also choose the operation mode for the base:

1) the levels of the construction yard/support weapon (relative to a 3-touch tib silo).

2) the levels of the defense buildings and units (relative to a 3-touch silo).

3) the levels of refineries (2-touch refinery vs 3-touch silo, by ROI).

4) how often the base will make a 20-jump per day (0.1 means 1 jump in 10 days, 0.5 means 1 jump every other day etc).

5) The ranking of the alliance, which determines the poi bonus (the default value of ranked-20 alliance corresponds to roughly zero bonus etc).

These assumptions will be preserved throughout the simulation.

Each of these options has implications for growth. Spending resources (tiberium or power) on upgrading buildings or units that do not produce those resources themselves necessarily slows down the growth of a base. The options allow the user to choose how much of a burden these will be. The default options are to have defense 5 levels behind the base, and other buildings 1 level behind. The default is to move the base every other day. This represents my perception of the normal mode in which people operate. The default of is no alliance bonus. Changing the options allows the user to properly evaluate alternative modes.

The user can also choose whether s/he wants the program to auto-fill the buildings, or to preserve the locations in the cncopt provided by the user. In the case of auto-fill the program will choose several arrangements that follow different development strategies, analyze them all, and report the best one for growth. In the second case, the user can evaluate a particular development strategy that s/he prefers.

Auto-fill order:

1) All the harvesters that can be connected by at least 3-touch silos

2) All 4-touch, 5-touch and 6-touch silos.

3) Different combinations of 3-touch silos (none, only those touching tib, all).

4) Accumulators surrounded by power plants (1 or 2 accumulators, minimum power plants to surround, max crystal touches and max potential refinery touches).

5) The remaining space is filled with refineries and supporting power plants to maximize credit production.

6) 4 key buildings (CY,DF,DFHQ,support).

There are different opinions on whether 3-touch silos need to be used or avoided, and whether 1 or 2 accumulators should be used. The program allows for various combinations of these and evaluates each of them. Different strategies work for different layouts, and there is not a single best strategy here. The best one for growth among those considered - is reported by the program.

*Note: The auto-fill does not consider setups with too many power plants or accumulators spaced too far apart, because that leaves less space for refineries (which are a burden), i.e. artificially increase the growth rate. These setups typically win growth-wise, but having no refineries is not a realistic goal. However, you can always choose to fix your own placement and test a specific setup.

Simulation:

The program starts with the layout of the buildings (chosen or given) with all 40 buildings present and at level 12. The program chooses a target building to upgrade and waits for the time necessary to accumulate enough resources to upgrade that building. After that time passes, it does the upgrade and looks for the next target. The choice of target building is based on modified ROI (time it takes for the building to pay back the costs). Standard ROI assumes that all resources are equally valuable. Modified ROI weights resources (both for costs and for production) and chooses the weights that are best for growth for a particular base setup. The modified ROI also takes into account the losses of production due to move recovery.

There are a lot of different strategies, and it will take forever to try out all the combinations. To save time, the program uses some educated guesses of which combinations to try out (based on the wide range of simulations that have been tested out on various layouts).

Evaluation:

The evaluation criterion is the performance of the base from level 20 to level 43.

These levels were chosen because level 20 is roughly the level at which you get access to all 40 buildings, while level 43 is the usual level of the base at the time you are ready to kill the fortress for the first time in a regularly sized world with guardian bases at level 50. A good support base reaches level 43 around the 180th day on a server, which is the usual timeframe for the first fortress kill.

Performance measures include the growth rate (of total resources accumulated), the time it takes to get to level 43, the production of tiberium, crystal and credits.

The production of tiberium is evaluated based on the growth rate it can sustain.

The production of crystal is evaluated based on the level of offense you can support, conditioned on your total growth rate (if all of your bases had this same layout). Grade A means that the offense level is enough to kill the fort. Lower levels get equispaced lower grades.

The production of credits is evaluated based on the number of MCVs you can conceive, conditioned on your total growth rate (if all of your bases had this same layout). Grade A means that credit production is enough for 8 MCVs (to have a total of 9 bases).

The estimates for both crystal and credits are taken on day 200 of the simulation, or at the end of the simulation, if level 43 is achieved sooner.

The program reports the layout and the levels of all buildings at the end of the simulation in the text window below. This allows the user to see the chosen layout and the relative levels of the buildings in the browser.

In addition, the program reports the weight on power (in the modified ROI) that was used in the best simulation.

The graph on the left also plots the base level, the harvester level and the offense level that can be supported at each point in time during the simulation. Note that the effective offense level drops every time crystals are used to upgrade defense units.

Input layout restrictions:

All layouts are restricted to have exactly 12 resource fields. When the fixed option is chosen the layout needs to have enough buildings for positive production of tiberium, crystal and power. Only production buildings are left, the rest are stripped out, and a construction yard, a defense facility, defense headquarters and a support weapon are added.

If the entry is bad, the program does nothing or if it can identify a specific error it reports "Bad cncopt" at the bottom. At the moment forgotten layouts are not supported. You need to convert them to GDI or NOD layouts and uncover the resource fields. Conversion of forgotten layouts may become available in the future.

Please report all bugs that you notice.

All suggestions regarding operation modes and growth strategies are welcome.

* To install the Modified Base Scanner or the CnCopt+ Button in-game:

1. in Firefox, replace the whole code of the Maelstrom Addon Base Scanner (145168.user.js) or CnCopt Button (131289.user.js) with the new codes

2. in Chrome, when using the script pack, a) in the options menu switch off the Base scanner/Cncopt button, b) install tapermonkey, c) run the script(s) in tapermonkey

Version 2.0. Created by chertosha on January 23, 2014.