Staredit Network > Starcraft I Database > File: Menu System

Sunken Treasure
(Reload)
Menu System
Download
Filename: Menu System vBeta.scx
Filesize: 169.45KB

Thumbnail

Rating
Not enough votes.
Log in to rate

File Information
General
Uploaded on:Aug 8 2010, 11:23 pm
Uploaded by:jhuni
Author:Not Given.
Downloads:293

Map-Data
Tileset:Ice World
Dimension:64x64

Forces
Force 1
Human (Zerg)
Force 2
Computer (Terran)
Computer (Protoss)
Computer (Zerg)
Computer (Terran)
Computer (Protoss)
Computer (Zerg)
Computer (Terran)
Force 3
Force 4

Map-Statistics
Units:254
Triggers:1085
Locations:190
Strings:789
File Description
This demonstrates a new menu system that should by far more user friendly and useful then the traditional dropship/gateway based selection systems. This menu system is the result of three components:

1. A location grid system that will move a 7x7 grid of locations to any coordinate in the map. Each location is 8 pixels by 8 pixels so the area covered is 56x56 pixels.

2. A font system that will draw any standard ASCII letter using burrowed zerglings in the 7x7 area, death counter arithmetic is used for new lines and the ETX character is used to indicate the final character in the string. Each death counter stores 4 characters since death counters are 32-bit and characters are 7-bit.

Ascii Table

3. A selection system which uses zerglings and burrow to select options in the menu-system.

All Images

Comments
Apos wrote on Dec 14 2010, 4:35 pm:
Very clever

payne wrote on Aug 10 2010, 3:17 pm:
Interesting.

[01:28 pm]
NimoStar -- Unless specifically logged and retrained. But that wasn't in the original version. It just played itself
[01:28 pm]
NimoStar -- it has a static training, which is converted into weights. it doesn't actually learn from new games.
[01:16 pm]
NudeRaider -- but it would explain why that bot needs cheats to win
[01:05 pm]
NudeRaider -- Oh_Man
Oh_Man shouted: No training is very computational expensivr it doesnt train in real time tht culd actually be used against it
oh but alphastar training was *very* cpu intensive.
[12:16 pm]
Oh_Man -- inference vs backpropogation
[11:31 am]
Oh_Man -- No training is very computational expensivr it doesnt train in real time tht culd actually be used against it
[11:11 am]
NudeRaider -- Oh_Man
Oh_Man shouted: Hope that makes sense. Also artosiscast just dropped a 1 hour vid he analayses a bunch of its games. The TLDR is the maphack is responsible for a large percentage of its wins because its just hard countering zerg rushes
People keep mentioning it's a bot using reinforced self-learning (or w/e the correct term is) but I wonder, does it still actively learn now that it is deployed? How does that work? Or was it trained on like 100k games playing against itself and now training is finished and its capabilties are now static?
[09:53 am]
NimoStar -- ban the bot :))
[05:17 am]
Oh_Man -- Like it doesnt even bother to build a comsat unless it sees a lurker or dt, etc
[05:15 am]
Oh_Man -- Hope that makes sense. Also artosiscast just dropped a 1 hour vid he analayses a bunch of its games. The TLDR is the maphack is responsible for a large percentage of its wins because its just hard countering zerg rushes
Please log in to shout.


Members Online: NimoStar, NudeRaider