 |
|
|
|
|
|
||
|
|
||
|
|
||
|
|
|
|
|
|
||
|
|
||
|
|
||
|
|
|
|
|
|||||||||||||||||
Add Forum to Favorites | Send Topic To a Friend | View Forum FAQ | Track this topic |
Last Thread Next Thread |
| Algorithmic Forays Part 1 |
|
 |
randomZ Member since: 8/17/2002 From: Germany |
|||
 Posted - 3/8/2004 1:52:09 PM |
||||
| Nice article. For how long do you plan to keep this series up, and what topics do you plan to cover? --- Just trying to be helpful. Sebastian Beschke Just some student from Germany http://mitglied.lycos.de/xplosiff |
||||
|
||||
|
grahamr Member since: 12/13/2000 From: Middlesbrough, United Kingdom |
|||
|
Posted - 3/8/2004 7:16:42 PM |
||||
| Where is the rest of it? It just started to get going and stopped. Stay Lucky, Graham "Mournblade" Reeds, ICQ: 30514803 http://homepage.dtn.ntl.com/grahamr/ |
||||
|
||||
|
spur4444 Member since: 3/9/2004 From: Haifa, Holy Land |
|||
|
Posted - 3/9/2004 7:39:39 AM |
||||
| grahamr: As I said, patience for the first part, since it's only introductory. In the next article, things will start getting more interesting. randomZ: Thanks ! I plan to keep the series up for as long as I feel fun doing it. This depends on the topics, the amount of my free time and reader feedback. The topic of state machines is vast, and it's possible to cover a lot of interesting algorithms/data structures/techniques related to it. After that, I don't know. Maybe reader feedback will help me decide. -- Eli Bendersky [edited by - spur4444 on March 9, 2004 7:42:41 AM] |
||||
|
||||
|
NatasDM Member since: 8/8/2002 From: Tampa, USA |
|||
 Posted - 3/9/2004 2:23:56 PM |
||||
| Before this article was posted, I was thinking (and even started) on an article about iterative finite state machines (decomposing an algorithm with loops into a finite state machine). If your interested in covering this topic, say something to keep me from using any more time. (maybe I will send you the prototype doc and some sample code) Cheers |
||||
|
||||
|
luskan Member since: 3/9/2004 From: Warszawa, Poland |
|||
|
Posted - 3/9/2004 4:58:49 PM |
||||
hi, great article  - I would really like to read about making big complicated FSMs more readable and easily changable. I mean what are the alternatives to masive functions composed of switches and ifs??thanx |
||||
|
||||
|
_DarkWIng_ Member since: 1/11/2001 From: Duplje, Slovenia |
|||
|
Posted - 3/9/2004 5:04:36 PM |
||||
quote: Jump tables or function pointer arrays. You also might come up with something using a bit of template magic. You should never let your fears become the boundaries of your dreams. |
||||
|
||||
|
DakeDesu Member since: 7/12/2001 From: Edmonton, Canada |
|||
 Posted - 3/10/2004 6:41:09 AM |
||||
| Er... I know it is only an introduction... but it is still quite short. I certainly hope part two comes out soon. As per other things you could have put in, would have been more on the history of FSM. Such as, I think it was, the Turing Machine, which Alan Turing designed, that was one of the greatest FSM designed... though, it has been forever since I've read anything on that, so my memory might be weak. Not too bad of an article, but much too short. Unless you plan to be putting out parts bang bang bang. |
||||
|
||||
|
randomZ Member since: 8/17/2002 From: Germany |
|||
|
Posted - 3/10/2004 3:43:32 PM |
||||
| AFAIK, a Turing machine is not an FSM. What I learned in CS classes: 1. Finite State Machines. Have an input, may have an output. States change depending on input. 2. Stack Machines. Have an input, may have an output (I think). States change depending on input and topmost stack element. On state change, a stack operation (push, pop, noop) is performed. 3. Turing Machine. Input and output both take place on an infinite tape. A change performs, depending on the element at the current tape position, a state change, writing an element to the tape and moving left or right on the tape. --- Just trying to be helpful. Sebastian Beschke Just some student from Germany http://mitglied.lycos.de/xplosiff |
||||
|
||||
|
Zefrieg Member since: 3/27/2002 From: Anchorage, AK, United States |
|||
|
Posted - 3/11/2004 12:18:57 AM |
||||
| Let us just hope the author has the skill to implement a finite-state machine container. They are very useful when dealing with game elements such as NPC dialog trees that are created externally from the application. |
||||
|
||||
|
spur4444 Member since: 3/9/2004 From: Haifa, Holy Land |
|||
|
Posted - 3/11/2004 12:33:18 PM |
||||
| Thanks for all the feedback - keep it coming as it will help me improve the articles in the future. The next couple of articles are already ready, the next will be released by GameDev in a couple of weeks, so just be patient. One more thing: coding a class for a FSM is not my goal. If this is what interests you most, there are tons of information online with implementations in every language you could dream of. My goal is more theoretic - I want to explain interesting algorithms and techniques, the coding is only an implementation of these algorithms. Naturally I try to make it robust and efficient, but surely it's not the best out there. -- Eli Bendersky [edited by - spur4444 on March 11, 2004 12:33:53 PM] |
||||
|
||||
| All times are ET (US) |
Last Thread Next Thread |
|
