Arrow Key Movement System?

I haven't experimented with the TFT arrow key movement alot, but from what I've heard here on the forums it has a few problems, namely that units given move orders simply don't respond quickly enough to arrow key presses.

Has anybody come up with a way that works really well?

Ya know, its just hard to think of another way to do it other than checking for the person trying to move the screen, and relocking it every .5 seconds, and then ordering the unit to move that direction in small incriments. Ive found that they just simply to do not respond as well...I guess you could try using triggers for every time that somene presses W, the unit moves up a higher increment..or somethign...what the heck am i talking about. But the arrow keys i think is waht messes it up, because normally that moves the screen and the game has to relock the camera and then move the unit, so who knows.

If you lock the camera, would it still try to move the camera or would the delay be reduced? Who knows... bah, I'm just babbling.

The camera has nothing to do with the delay. The old RoC method of key movement was to have the controller next to a shop that had the varias items that triggered movement. This seemed to be the quickest way to do it, yet lots of problems arise with this doing. Just look at the Death Sheep scenario and play it. The response is quite quick.

To make a very good keymovement make 3 boolan variables 1 for each key. Then if a player presses a key make the boolan true if he releases the key set it to false.
Then make a trigger:
event - every xxxx second
action - if boolan left = true then order unit to move to position of it self ofset by xxxx using units facing + x els do nothing
then make a right if action and a up if action

I think you can figure out how to change the units facing when it stands still, if not post a reply

the key press event is slow I dont know why. My mod has a trigger that goes every .03 secs to check for key press boolean. It must be a time delay inbeween time wc3 engin sees an key press and the and the time jass sees it. My arrow key moevent triggers has cut the delay time to about .5 of sec wich is pretty good from what i have heard.

The .05 second delay method juz doesn't work for me cos the units start to jerk. As far as I know, the most non-jerky timing is 0.2 seconds.

But as for using the TFT arrow key system to create very sensitive orders, I think I have a method that makes it quite sensitive (I for one, can't tell the difference).

For those who have trouble understanding juz open the file "kekeke" and look at the "Movement' section. It's a map I started on a few days back, and I'm just posting it for the movement. Don't bother playing it.

For those who dun wanna download the map, here's how it works.

Trigger 1
Event - Every 0.2 seconds
Conditions - Forward = 1, Back = 0, Left = 0, Right = 0
Action - Order unit to move to Any point with polar offset.

Trigger 2
Event - Player presses up arrow key
Action - Set Forward = 1

Now we all know that. That's really simple and all... but many people forget this.

Action - Run Trigger 1 <Checking conditions>

If you happen to have triggers for diagonal movement (That's what variables left and right are for) then juz run those triggers too, checking condition.

The trigger will stop running once the arrow key is released if u put a terminating trigger (Release, Forward = 0).

Running the the triggers but checking conditions means that you don't have to wait for the 0.2 second cycle to start again.

I have a small flying game which uses the arrow keys to detect horizontal and vertical changes and I haven't really noticed any serious delay problems with it. I think people are just expecting fps movement reactions from an rts game.

~kharma

well, sometimes people want to make slight little adjustments to their movement. And getting faster response times makes it much much more of a pleasant experience for the fussy gamer.

Here's mine...

Works pretty smoothly, just tweak the waits to your liking.

as far as i'm aware it just has to do with the fact that you can't reduce the latency in network games

if you compare, side by side, the speed at which a key-press order takes place and the speed at which a click order takes place, they are roughly the same

from what i've tested, i've noted that
A: single player key-pressing reacts the fastest,
B: host multiplayer key-pressing reacts second fastest,
and C: client multiplayer key-pressing reacts the slowest

uh, these maps post here that claim to be fast, have they been tested as multiplayer client?