SAMAn Animated 3 D Programming language Christian Geiger

SAM-An Animated 3 D Programming language Christian Geiger, Wolfgang Mueller, Waldemar Rosenbach C-LAB Reviewed By: Johnny Sia, csia 005 Allen Wang, awan 015 Li Li, lli 057 Hui Zhang, hzha 113

SAM Introduction (1, 3, 3. 1, 3. 2) • Solid Agents in Motion • Domain: • Expected type of user: • Metaphors used: • A SAM program is a set of agents which communicate by exchanging messages • General purpose programming such as 3 D games, virtual reality, CAD • Define programs in 3 D and additionally animate their execution • Any programmer, as it can be used for general purpose programming • Agents with ports, messages and rules Johnny

SAM Execution & Animation Li (3. 3, 3. 4) First Phase: 1. the 1. selected rule check (rule with true condition) grows size all agents the conditions of their rulestointhe parallel. of the agent until the condition overlaps the arrived Note: The order by which rules are elaborated is not defined Messages. in SAM. Why? 1 st In 3 D, any intended order of the enclosed rules is hard to Phase 2. the scanner moves over the slices of the tube which indicates determine. the execution of the individual actions. 2. Produced message is kept at the output port of the agent until 3. some actions generate messages at the output port. the beginning of the second phase. 3. All other actions like moving, rotating, coloring, deleting, and creating agents are immediately executed. 4. after having executed all actions the messages generated Second Phase: at the output move to their destination. Execution Cycle 1. Start when all agents have completed the execution of their actions. 2 nd 2. All messages move to their specified destination then. Phase 5. in parallel to (4), the scanner returns to its initial position (Frame 6) and the selected rule returns to its initial 3. After their arrival the first phase triggers again the pattern size and location. Matched message(s) at the input(s) is/ matching for selecting rules, etc. are consumed when rescaling the rule (Frame 7 -8)

Sam Examples (4, 4. 1, 4. 2) Producer-Consumer Example In Agent execution phase: ENVIRONMENT prepares “hay” message. COW prepares “milk” message and consumes “hay” received in last cycle. CAT consumes “milk” received in last cycle from COW In Message Exchange phase: ENVIRONMENT send “hay” to COW send “milk” to CAT Abstract 3 D and Textual Form Allen

Conclusions ► Type: a general purposed visual language ► End user: programmers, need to know C++ ► Terse: using a small set of notations, color to describe actions ► Reduced viscosity: each agent can be simply changed by assigning an external Open. Inventor model to it ► Support: needs extensions in various directions e. g. recursive behavior and loop constructs ► Further work: concepts for dynamic rule sets with priorities