can i make server?

[shs]

if the gamer doesn't need to communicate i won't need a "common" server?

[rvk]

if the gamer doesn't need to communicate i won't need a "common" server?

Why do you call the game "online" then?

I asked you what you exactly meant with "online game".

[shs]

i want to make application game and i want people to connect to it and play

[rvk]

i want to make application game and i want people to connect to it and play

Bingo... connect to what ???

Where do you "host" the game?

[shs]

in the application?

[rvk]

in the application?

And where is your application running?
On a hoster somewhere or on your own computer/server?

If it's on your own computer/server you would need to open up a port again for the clients to be able to reach it.

[shs]

oh okay but if i want to use the local network i don't have to right?

[jamie]

While thinking about it, you may want to use UDP packets. They are faster but can suffer
from lost data.
  If the data you plan on  exchanging is things like player moves then you can create your own
 error checking for the packets.

 Doing it the TCP way will cause the network to do hand shaking and that slows things down
for quick responses and not needed if you can evaluate the data you receive as a whole to
determine if it's a useable block of data.

[rvk]

oh okay but if i want to use the local network i don't have to right?

Yes. You should understand that by now since you already made a message program that worked locally.

[shs]

but how can i make a game with that?
for example in the server game, if i'm wanna make the shape move and if the client connects to the server, the client can move his shape. how can it read the movement?
for messaging they had recv and sendstring

[rvk]

In your Button1Click in the server you have a repeat until false.
This is an never ending loop.
You never break out in that loop.

So when you click the close button, the program is still executing that loop.

You need to create a break-mechanisme to break out of that loop.

(You could set a boolean WeNeedToTerminate to true in FormClose and change the repeat to "repeat until WeNeedToTerminate".)

[shs]

is it better to use timer than the repeat method?

[rvk]

You could use a second timer for the ListenerSocket to listen to incoming connections, yes.

Better would be threads but that's fat more complex.
(try the Timer2 first)

[shs]

Code: Pascal  [Select][+][-]

1. procedure TForm1.Timer2Timer(Sender: TObject);
2. begin
3.     if ListenerSocket.canread(100) then
4.     begin
5.       edit3.Enabled := True;
6.       button2.Enabled := True;
7.       SetLength(Connections, Length(Connections) + 1);
8.       Connections[high(Connections)].Sock := TTCPBlockSocket.Create;
9.       Connections[high(Connections)].Sock.Socket := ListenerSocket.Accept;
10.       Timer1.Enabled := True;
11.       //memo1.Lines.add('Client ' + IntToStr(length(connections)) + ' is connected');
12.       if length(connections) = 1 then
13.       begin
14.         memo1.Lines.add(FormatDateTime('DDDD, dd MMMM YYYY HH:MM:SS', Now));
15.         memo1.Lines.add(
16.           '------------------------------------------------------------------------');
17.         memo1.Lines.add('Say Something');
18.       end;
19.     end;
20. end;    

yes it works
i made this 'disconnect' but i get sigsegv error message when i click that button

Code: Pascal  [Select][+][-]

1. procedure TForm1.Button3Click(Sender: TObject);
2. begin
3.   timer1.enabled:=false;
4.   button3.enabled:=false;
5.   button1.enabled:=true;
6.   ListenerSocket.closesocket;
7.   listenersocket.free;
8. end;      

[rvk]

Also disable Timer2 in Button3Click because if you free the ListenerSocket you can't use that anymore in Timer2 until you click button1 again (when it is recreated).