Messages

[eval(string command)]

Hm. Well as long as others are asking, I also thought an eval(string command) (as JScript would put it) would be nice.

Of course it is possible to build your own inside AGS (up to a point), but it would be something of a coding abomination based on "if(){}else if(){}else if(){}else ... " and a whole heap of string manipulation (or a custom input language/method).
That would probably be good enough for less than 20 functions.

I was thinking about function pointers too at one point, but I'm not sure they'd be worth it.

I had a similar problem which came down to the event based room script for the interaction being out of sync with the global script.

Try displaying a message within the interaction script to see what the settings are, and if they are correct then look for another process that may be resetting them.

You may have to import a function from the global script to the global header that tells the global script to behave differently, and call it at the end of the interaction script.

Ha! I was just about to update my question, but you were too quick for me.

Thanks
I just discovered GetGameParameter().
Since I started using the page links instead of trusting the 'Contents' panel, I've been finding new things.

For baselines I was hoping to retrieve existing values, as might be set in the room editor, rather than adding extra parameters to everything. Avoiding explicit per-object data entry is one reason that I want the value in the first place.
But of course, the info you provided is sufficient to at least make it work, so thanks anyway.

[edit: damn this forum ignoring where I put the caret on the first click.]

I don't know how you intend to test adjacent rooms, but I don't enjoy switching cursor modes just to travel around. Doing this at room boundaries also draws attention to the division of what we pretend is continuous.

I have two ideas. When there's a bot in the next room, either
a) Display a (simple, symbolic) graphic overlay at the appropriate edge or door, indicating that Ray has seen or heard something.
or b) Change the cursor graphic (not mode) when over the appropriate position. The graphic reverts automatically to it's default otherwise.

I prefer (a), but (b) would force the player to continually scan the area, even behind them, if you wanted to keep the pressure on.

Ideally, this would work for scrolling rooms too, but that would require more complicated coding to get relative positions in the same room.

Is there a function for this? And what about sprite dimensions? The alternatives seem a bit excessive.

On a related topic (since the above would solve this too), how does one find the depth ordering of overlapping objects/characters.

That's a very nice idea you have there.

Apart from suggesting that there may be remapping or transcoding issues, I can't think of anything that'd help. I know that green flash is quite a common symptom (why only green? I don't know).

Anyway, it'd still be worth doing even if the problem persists in my opinion.

Not yet, I must admit. I'm still at the stage of ensuring that I (a) use the AGS core effectively, and (b) finish something.

I don't really have a rich enough code base down to put it through its paces yet anyway.

I understand its potential, but I must resist the temptation to overreach.
Although my next game ...  :

Suggestion.

Code: ags

int useVideo;

function game_start(){
//...
useVideo = FileOpen("1.dat", FILE_READ);
if(useVideo){FileClose(useVideo);useVideo=1;}
//...
}

useVideo is then available for choosing between code branches when the need arises.
The last statement isn't necessary, just more conventional. You may even wish to set a bit for each video, but I assume that all or none is the usual case.

Or, you could include the videos in the zip if you don't want them to be optional.

You may be right about the walking sound, but the hard drive/neighbourhood 'ambience' isn't very conducive to gameplay either. The should be near constant sound from the game if only as a distraction from real world noise.

I found the tutorials very useful for sorting out what was relevant to putting together a working basis, and getting a feel for how things are organised, after which it was much easier to investigate the details.

I had to edit the hrefs for offline use, though.
I'm sure I'm not the only one who would prefer that it use relative rather than absolute URLs.

Another excellent extension.
Sophisticated game design becomes even more accessible. Thanks.

Spoiler

Ok, 'scuse me, I forgot one detail about the detective I.D. The answer doesn't become available until you get it from someone else.

[close]

Spoiler

What- you just strolled into someone's place and rummaged through their stuff without even talking to them first? ;p

[close]

Rachel,

Spoiler

Improvise. There's another item that may be instrumental re: the bars

[close]

No. Not if you have any consideration for the game writers or other players.

The game title is a suffiently descriptive subject for this forum.

If you must tell everyone how clever you've been so far, there's this whole big message area to do it in.

I was thinking that the intended use might only involve rendering something on the GUI while open, based on some variable game properties, without actually saving the image.

Thanks again.

Did you even include that video file in the zip? I have the DivX codec and everything but it doesn't work. I couldn't see a file named 3.dat in the folder!

Heh. Nor me, but I figured maybe the dat file is supposed to be generated by the exe?

Hey, don't worry about me. I'm just providing the info in case you find it useful.

(Personally, I suspect the .swf )

[once only]

Gilbot: Yes, the call is made once only each time a walking character stops.
I omitted that line because it is no longer required for the basic mechanism, only to (optionally) indicate whether we are interested in the event, which depends on the game design. If we wanted to signal disinterest, it would probably make more sense to do so in "stuff()".

If we always used it, we would simply put -

Code: ags

function on_event(e,d){
if(e==CHAR_STOP){
if(d==EGO){
stuff();
}}}

Or if applying the same code to different characters -

Code: ags

function on_event(e,d){
if(e==CHAR_STOP){
stuff(d);
}}

[...]However, the issue is that if a blocking script was running when the character stopped, on_event wouldn't be able to run and therefore the event could get missed.[...]

Hmm. I hadn't considered blocking scripts losing events. I'll have to be careful about that.

Just so. I should really put something like the actual code required in case of confusion -

Code: ags

int watchEGO;
function repeatedly_execute(){
  if(watchEGO){
    if(character[EGO].walking==0){
      watchEGO=0;
      stuff();
}}}

Of course, in practise we would still have a flag to signify our interest in the event, so the amount of text is about the same -

Code: ags

int watchEGO;
function on_event(e,d){
  if(e==CHAR_STOP){
    if((d==EGO)&&(watchEGO)){
      stuff();
}}}