SRL Archive - Discussions and Debates

Well, we have a maze thread, and personally I think this one is more common and probably easier to solve.

My opinion:

Pull lever. Make fairly generic DTM's of each animal while on the lever screen while it's straight (it rotates, make DTM's of them while they are perfectly level). Run a loop, wait etc until find one of the DTM's. Have the script take that picture, then use CopyCanvas to bring it down to the size of one of the balloon animals walking around. Then, find the bitmap, click, go back, etc.

This is hard to explain because I don't have pictures, but I'll try to go through it one more time.

First:

Get the random, pull the lever, make DTM's of each animal while it's perfectly level.

Second:

GetClientCanvas, CopyCanvas to make it as big as the animals are walking around.

Third:

Camera angle down, find bitmap, click, talk to guy, repeat.

It would look something like this,

procedure DoScapeRune;
begin
  if (FindScapeRune) then
  begin
    LoadBalloonDTMs; // Loads the saved DTM's of each animal on the screen
    if (FindObj3(fx, fy, 'ever', LevelColor, MSX1, MSY1, MSX2, MSY2, 15)) then
    begin
      Mouse(fx, fy, 2, 2, true);
      Flag;
      repeat
        Wait(100);
        for BalloonDTMArrayNum := 0 to HoweverManyAnimalsThereAre do
        begin
          if (FindDTM(BalloonDTM[BalloonDTMArrayNumArrayNum], fx, fy, X1, Y1, X2, Y2)) then Break;
          if (not (LoggedIn)) then Exit;
        end;
        if (not (LoggedIn)) then Exit;
      until (false);
      SafeCopyCanvas(GetClientCanvas, GetBitmapCanvas(BalloonBitmap), X1, Y1, X2, Y2, SmallX1, SmallY1, SmallX2, SmallY2);
      CloseWindow;
      LowestAngle;
      MarkTime(Marker);
      repeat
        if (FindDeformed(fx, fy, BalloonBitmap, MSX1, MSY1, MSX2, MSY2)) then Break;
        KeyDown(VK_RIGHT);
        Wait(50 + Random(200));
        if (TimeFromMark(Marker) > 20000) then Logout;
        if (not (LoggedIn)) then Exit;
      until(false);
      Mouse(fx, fy, 2, 2, true);
      GoBackToPrisonPete;
    end;
  end;
end;

Keep in mind I may have done this random 5 times, if that. If anybody gets the random and is decent with canvases (optional, I guess) then they can try to match up how much bigger the picture is in the first screen compared to on the main screen. It's possible that it changes and this idea won't work at all but it looks decent drawn up, IMO.

Discuss, suggest, etc.

Nice idea.


But what about once it finds the DTM on the level screen, have it look for a bitmap instead.

Hmmm, it finds the DTM, then makes a bitmap.

I don't understand what you were asking. Bitmaps would be huge if you used them for those pictures.

I think he means if it finds the DTM, say of Dog1, then it would look for the bitmap of Dog1 on the ground. Obviously the bitmap of dog1 on the ground would be diferent to the dtm, but this is just a way of linking the two together if you know what I mean.

No whoops. sorry… wrong picture…

When you pull the lever, it will look for a DTM of a animal picture… THEN depending on the DTM, look for a bitmap… It may have to constantly rotate the sreen to get it the right way.



EDIT: Claw can understand me! YAY! But I was shamedd when I was beaten…

Alright, I see. You're saying once it finds the DTM of what dog it is then it loads the bitmap of that dog and looks for it on the ground. The only difference between that and what I did is that you don't have to make bitmaps, you only have to find the difference in size between the first picture and the second that you are looking for. Either one would probably work.

There is a large possibilty that it will click the wrong balloon… bitmap masks are probably the way to go (of the purple color)…

Also make 4 or 5 bitmap masks of each animal, on the ground, at different camera rotations (all lowest camera angle though).

This would largely increase the chance of finding it.

Also we have to account for the distance we are away from the balloon… the size increases/decreases depending on distance.

they will find a way, no bad idea though, pancakes, your idea sounds cool too..

Is there a SCAR function to shrink stuff like DTM's Bitmaps?

Cool Idea.

I heard some talk of people working on this before, and they were searching just for the tails, rather than the whole animal…..$.02

~RAM

Still, I havn't seen this in awhile, but I thought the changes were that some had antler shaped things and then some other changes in the legs. I don't remember :O, it might have been tails and ears. . .

This is hard to solve legitly :eek: