i''m wondering - how did you get these values? they are VERY useful to know.

your speeds are Teh Wrong :D, ive measured rotate slow 4,5deg per call, and 360deg in 2secs using no userinput-source, fast rotate is exactly 2times faster and ive also got 40calls per second, maybe wrong, maybe not fixed?


edit: now iam confused, got 20calls per second (still 80calls to rotate ~360deg in slow), hmm i think i need a bit sleep


T2k

[edited by - T2k on August 2, 2003 6:29:04 PM]

it''d be nice to have a RotateBy(Degrees) function...

I did the math and also got the results you got, except for your strafing value. Mine came out to be 0.42 units per second. Can anyone else confirm the strafe speed?

---
Brent Gunning | My Site

I got the same results as ToohrVyk.

Is it possible for ToohrVyk to have the right angles and Khawk to be mistaken?

I just checked this... the FPS seems to be 45.
Try multiplying ToohrVyk's numbers by 45 to get units/second. They come out to be nice numbers.

Angular speed :
Slow : 0.034906 rad/update or 2 degrees/update (90 degrees/second)
Fast : 0.069045 rad/update or 4 degrees/update (180 degrees/second)

Forward speed :
Walking : 0.4444 units/update (20 units/second)
Running : 0.8888 units/update (40 units/second)
Reverse : 0.5555 units/update (25 units/second)

Strafing speed :
Normal : 0.3333 units/update (15 units/second)

I'm not trying to bite the hand that is feeding us, figuratively, but these numbers appear to be right.

[edited by - void2357 on August 3, 2003 3:40:54 PM]

As far as I can tell, the values have nothing to do with the fps or how long you take per cycle. I tested it by setting the release flag to turn off the timing check, then spent about a quarter of a second each frame writing huge amounts of random data to text files to slow down the system - with a cycles per second of about 4, the values I got were still exactly the same as the ones I got with cycles per scond of about 30, and they''re approximately the same values that ToohrVyk got.

However, the values don''t seem to be quite precise... I tested by having the bot stare directly at an object, or as close as it could get, then calling turnleft for 180 cycles. The direction of the object was usually similar to the original direction (within about ~0.03 degrees), but every so often it would be several degrees off. Once again I tried slowing down the system to see if it was somehow timing based, but it still almost always took 180 cycles to be facing directly at the object again.

My best guess would be the KHawk added a random factor just to make our lives difficult