Indy is a Bolo autopilot, otherwise known as a brain. Indy is also a 'borg. When activated Indy will play Bolo automatically. If the caps lock key is down Indy will be in 'borg mode and will assist you while you play rather than playing by itself.
• What does Indy require?
Indy needs about 750k to run. You must increase the Bolo application heap by this amount to provide enough space for Indy. Therefore the application heap should be set to about 1850k (you can reduce the amount of memory Bolo wants by removing the sound file). If you select Indy from the menu and it does not turn on then you need to increase the amount of memory dedicated to Bolo.
Indy also needs a good deal of processing power. In general you need at least a IIcx class machine (16MHz 68030) or higher to run one Indy. A general rule of thumb is about one Indy per 10MHz on an '030 and one per 5MHz on a '040. Any more than this and Indy will not get enough CPU to be effective.
• How do I use Indy?
Place the Indy brain file in the "Brains" folder that is located in the same folder as Bolo. Load up Bolo and set up the game, make sure that you have "allow brains" checked. Select Indy from the "Tank Control" menu and Indy will begin playing. To use several Indy's on one machine make sure that you have AppleTalk turned on (even if you are not connected to a network). Start up several copies of the Bolo application. Each copy of Bolo will be for an Indy (one copy may be for yourself if you wish to play on the same machine as the Indy's). Ally the Indy's appropriately to give yourself a challenge. Two Indy's are a challenge for a human player, though this may vary depending on your skill.
• How do I use the 'borg mode, and what does it do?
Press the caps lock key so it is locked on. When Indy is in the foreground and the caps lock key is down it will enter 'borg mode. Any Indy's that are in the background will be unaffected by the caps lock key (ie. the will play automatically). Use the Bolo default keys to drive around and combat as you usually do.
When in the 'borg mode Indy will send your builder on tasks and will also communicate with any Indy's you are allied with. Indy will build road underneath the tank when on poor terrain. Indy will repair pillboxes that are damaged when you are next to them. Indy will build blocks between you and a pillbox that is spiking your base if you are on that base. Indy will automatically shoot an enemy builder that wanders into gun range if you do not press any keys. Indy will tell all the other Indy's that are allied with you about pillboxes and bases that you see.
• Why should I use the Indy 2.0 'borg over other 'borgs?
Indy is not the best 'borg out there. However it communicates with allied Indy's, this is a major advantage. If you attack a pillbox for example, the Indy 'borg will tell your Indy allies about it. If you walk away from the computer for a moment, you can turn Indy on by just pressing the caps lock key (Indy will already know the map and all the pillboxes since it records those in 'borg mode). Indy is very good to your builder (Indy may get your builder killed, but it's fairly rare).
• What is different about Indy 2.0 over older versions?
Indy 2.0 uses a more advanced movement algorithm that previous versions. Indy can now navigate on maps that would have stumped previous versions. Indy is now much faster at driving from point A to point B. The new movement algorithm is courtesy of Daniel Kallander, thanks Daniel!
Indy 2.0 also communicates with other allied Indy's. Indy will tell others about pillboxes and bases. Indy informs others when it discovers a new pillbox or base or the information about a pillbox or base changes (for example if Indy shoots up a pillbox it will tell it's allies about the new strength of that pillbox).
Indy has a more advanced decision making algorithm. This means Indy will make much better choices when playing and in general be a tougher opponent.
Many small changes and fixes. This is actually the biggest reason why Indy is so good at playing. Indy has so many small directives and instructions for minor occurrences that they together make for intelligent actions (well, usually).
• What are the messages that Indy sends?
The messages that are from "Autopilot" are messages that Indy sends to itself. This is an information stream that tells the user what Indy wants to do at that time. The messages are all fairly self explanatory (things like Refuel, Fix Pill, etc.)
The messages from other allied Indy's are information about pillboxes or bases. A sample pillbox message may be : /PILL 8 128 130 E 6. This means that Pillbox 8 is at location 128,130 (X,Y where the origin is at the top left of a map and the map is 256x256 in size), is an enemy pillbox and has a strength of 6 (from a range of 0 to 15). A sample base message may be: /BASE 7 166 117 E 90 90 90. This means that Base 7 is at location 166,117 is enemy and has 90 units of bullets, mines and armour (note that Indy cannot tell information about enemy base armour, mines or bullets and does not send messages about friendly bases. Therefore the last three entries are only for future use.)
• What are Indy's strengths?
Navigation is definitely a strength. Indy can make it's way through a maze much faster that a human. Combat with other tanks is also a strength (Indy is a nasty fighter, though there are tricks you can use). Indy is fairly good at anything that requires precision, like shooting your builder (after all it's a computer).
• What are Indy's weaknesses?
Anything that involves pattern recognization (such as realizing that a checkerboard mine field is in front of you after blowing up several). Any strategies that are obscure or are specific for a map will throw Indy off. In general anything that involves higher level strategy is a weakness (got a simple algorithm for general strategy that wouldn't bring a Mac to a crawl? We'd love to hear from you! :^) Indy also doesn't have a sense of time, it see's the world as disconnected snapshots. Indy is far from perfect and will fail time to time, but 2.0 is better than 1.4.
• Why isn't there a better user interface with customizability?
Because we haven't got around to writing it yet. It will be there soon, but we would rather work on other more critical things.
• What future plans do you have for Indy?
Make it better. Seriously there are so many things to do that it's hard to select out one specific thing that we are going to do. In general however, we will try to make Indy communicate better than it does right now (perhaps with leaders designating tasks and everything). We will also add many more little things that hopefully will add up to a big difference. We may make Indy into an application that functions as a server to multiple bolo applications.
• I'm writing a brain, can I have some/all of your code?
It depends. We will not give out the source to the top of the line Indy until we are sick of Indy and Bolo and never want to hear of either again (may take a while :^). We've put a helluva lot of work into Indy (especially John) and want to maintain ownership of the code. Still we may give you snippets of code if you ask us (trading is always nice too). We have been considering releasing the 1.4 source code but it is in a bit of a disarray and would take some work.
• Who are you?
The person who wrote this document is Gersham Meharg, who is about to graduate from Simon Fraser University with a degree in Geography (rocks, rivers n'stuff). John Lawrie is the major author of Indy. John is a computer science undergrad at Simon Fraser University.
• History
Independent Brain - The first version of Indy (where the name came from). Wasn't very good and had several bugs.
Indy 1.0 - The version where we changed the name to Indy (shorter and snappier, no it has nothing to do with Indiana Jones). The movement algorithm wasn't very good (it just looked 1 square around the tank, and used a route find to unstick itself).
Indy 1.1 - Bug fixes on 1.0.
Indy 1.2 - Improved Movement algorithm.
Indy 1.3 - Very buggy release, essentially a 1.4 beta in disguise.
Indy 1.4 - Most popular and capable release. Version 1.4 was quite stable and incorporated many features making Indy reasonably robust. Indy would still get stumped on many maps. This is the version I recommend for people who do not have the CPU to run 2.0. This version is more capable than any other brain I've seen (excluding 2.0 of course).
Indy 1.5b1, b2, b3 - Beta versions of 2.0.
Indy 2.0 - The big jump in version number and capability. We decided on 2.0 over 1.5 since the changes were so lengthy over version 1.4.
Indy 2.0.1 - Small bugfix in movement algorithm over 2.0.
Indy 2.0.2 - Fixes movement argorithm. Indy should not get stuck anymore unless..
- Has more reslove when attacking pills through walls.
• Indy is Freeware.
That means you can copy Indy as much as you like. However you may not distribute modified copies. Indy is copyright 1993-1994 by Gersham Meharg and John Lawrie. We make no warranty whatsoever about this code. (it really can't do anything harmful except crash your computer, and this is fairly rare).
Even though Indy is freeware it would be nice if you sent us $5 or $10 if you really enjoy Indy and would like to encourage us (However, pay your Bolo registration fee first). Failing cash, a postcard would be nice (I'm a geographer, I love to see stuff from around the world!). My address is: Gersham Meharg, 22 Escola Bay, Port Moody, B.C., Canada, V3H-3Z1.
Send us some email with your comments and suggestions!
