Results 1 to 12 of 12

Thread: Make all Interface

  1. #1
    Join Date
    Nov 2010
    Location
    Australia
    Posts
    1,472
    Mentioned
    0 Post(s)
    Quoted
    8 Post(s)

    Default Make all Interface

    Hi guys, I need some help regarding the new(not really) update made to the make all interface. The picture is shown below.



    Now, I need to know the command(s) for clicking the button, because im not sure if the uptext works on this new text.(I don't want to know the dtm/bmp from string method of finding the image and then clicking it, since its not very efficient.)

  2. #2
    Join Date
    Jan 2010
    Posts
    5,227
    Mentioned
    6 Post(s)
    Quoted
    60 Post(s)

    Default

    How are bitmaps not efficient? :x You could use FindColor/FindColorTolerance/GetColor/CountColors or whatever if you don't wanna use bitmaps.

  3. #3
    Join Date
    Nov 2010
    Location
    Australia
    Posts
    1,472
    Mentioned
    0 Post(s)
    Quoted
    8 Post(s)

    Default

    Quote Originally Posted by i luffs yeww View Post
    How are bitmaps not efficient? :x You could use FindColor/FindColorTolerance/GetColor/CountColors or whatever if you don't wanna use bitmaps.
    Well, anything to do with images(complete or not) takes processor power. Thats what i meant by efficient, and not to mention you have to free them. I already know the functions above, I just wanted to know if there was something else similar to uptext and the likes.

  4. #4
    Join Date
    Jan 2010
    Posts
    5,227
    Mentioned
    6 Post(s)
    Quoted
    60 Post(s)

    Default

    Well UpText checks are always good. And there's like.. FindTextTPA and stuff, which could be used here, if the charset is available.

    I wouldn't worry about processor power.. Unless you have a computer that came pre-loaded with 95, you're fine.

  5. #5
    Join Date
    Feb 2009
    Location
    Irvine, CA
    Posts
    2,873
    Mentioned
    8 Post(s)
    Quoted
    138 Post(s)

    Default

    If you really are concerned with processing power, for something as static as this mousebox works great! Along with an uptext check, of course.

    But that's if your computer came pre-loaded with 95, as luffs said.

  6. #6
    Join Date
    Nov 2010
    Location
    Australia
    Posts
    1,472
    Mentioned
    0 Post(s)
    Quoted
    8 Post(s)

    Default

    Speaking about processor power, which takes more, bitmaps or DTM's?

  7. #7
    Join Date
    Mar 2010
    Posts
    119
    Mentioned
    0 Post(s)
    Quoted
    0 Post(s)

    Default

    Quote Originally Posted by kongking View Post
    Speaking about processor power, which takes more, bitmaps or DTM's?
    In the worst case, checking a location for a map could require checking all pixels of the bitmap, or all the points in the DTM, for every candidate point. The DTM would be significantly faster in this case, due to its small size (maybe 5 spread out pts) as opposed to the bitmap (possibly 5x5 square = 25 pts).

    Realistically however, these checks short circuit very very soon. Most points on your screen will fail to match either the DTM or the bitmap after even the first or second pixel checked. DTMs are especially good for this, and bitmaps are slightly worse because you may have locally similar regions in false positives.

    So, DTMs have an advantage. However, the practical difference is not too significant. I certainly (and I suppose this is just personal opinion) dislike people who harshly criticize bitmaps. Maybe they had bad experiences in the past from large bitmaps with high false positive rates (the biggest bottlenecks in bitmaps). Sure they're slower in theory, sometimes even practice, but the difference is not necessarily practically significant.

    Now, I need to know the command(s) for clicking the button, because im not sure if the uptext works on this new text.(I don't want to know the dtm/bmp from string method of finding the image and then clicking it, since its not very efficient.)
    First of all, what everyone above me said. The location is constant. If you know the interface is up, then it's quite safe just to do a mousebox. Of course, you should verify that it is indeed up through some combination of visual and uptext checking, and that you didn't either lag or accidentally miss some previous clicks.

    If you are interested in finding its location, keep in mind that first of all it's clearly in the chatbox region, and second of all it looks like pretty much nothing else. A strategically chosen small bitmap or DTM which focuses on certain elements of the image (say, a bit of the cake dough combined with a bit of the cake pan rim) is going to be very specific.

    If the string you want to find onscreen is known, and you need it to find it very often, you may benefit from a premade bitmap. This is the sort of thing you see in, for example, autominers for slightly faster dropping. You just cut out a small section of the text you're interested in as a bitmap. It doesn't have to be full height, and certainly not full width, as the words in their entirety. Because the words have very specific line shapes that nothing else will really share.
    Wow. I've been gone a very long time indeed. So much has changed.

  8. #8
    Join Date
    May 2007
    Location
    Everywhere
    Posts
    1,428
    Mentioned
    0 Post(s)
    Quoted
    2 Post(s)

    Default

    You can use reflection, find the root point of the interface (I can do this for you if you want), and then just call the GetRelativeX and Y points and width/height, and a more-or-less perfect box for you to click in will be created.

  9. #9
    Join Date
    Jan 2010
    Posts
    5,227
    Mentioned
    6 Post(s)
    Quoted
    60 Post(s)

    Default

    Although reflection, I assume, is slower than bmp/dtm/color. Especially a simple GetColor.

    (Not that I care, just that he seems to care. )

  10. #10
    Join Date
    May 2007
    Location
    Everywhere
    Posts
    1,428
    Mentioned
    0 Post(s)
    Quoted
    2 Post(s)

    Default

    Quote Originally Posted by i luffs yeww View Post
    Although reflection, I assume, is slower than bmp/dtm/color. Especially a simple GetColor.

    (Not that I care, just that he seems to care. )
    Based on his case, if he wants uber-efficiency, I'd do what you said GetColor (or GetColorTolerance?) with the orange.

    What is defined as inefficient though (@ kongking)? Back in the old days I used a 700 line BMP and it found it very fast :s -- though I'd never do that now.

  11. #11
    Join Date
    Nov 2010
    Location
    Australia
    Posts
    1,472
    Mentioned
    0 Post(s)
    Quoted
    8 Post(s)

    Default

    Thanks for the help so far guys, but now even the bitmap/dtm way won't work. It just can't find it. Has anyone else tried to click the button?

  12. #12
    Join Date
    May 2007
    Location
    Everywhere
    Posts
    1,428
    Mentioned
    0 Post(s)
    Quoted
    2 Post(s)

    Default

    Quote Originally Posted by kongking View Post
    Thanks for the help so far guys, but now even the bitmap/dtm way won't work. It just can't find it. Has anyone else tried to click the button?
    If you use BMP, try using tolerance. I personally recommend doing "FindColorTolerance" with the orange thing.

    Or save yourself time and use Reflection x-)

Thread Information

Users Browsing this Thread

There are currently 1 users browsing this thread. (0 members and 1 guests)

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •