Announcement

Collapse
No announcement yet.

Calling all journal gurus

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

    Calling all journal gurus

    So I am making progress on the RST AUBench, but getting the render to work is really causing me grief.
    I created partial journals for both RAC & RST, and added lines one at a time, re-running the journal each time, just to make sure I was getting exactly the same results out of both. I managed to get to the point where I have RAC showing Floor Plan 6 and RST showing Structural Plan 6 exactly the same, ready for camera placement.
    And yet, when I use the 7 lines below for the camera I get radically different results. RAC looks right, but RST places the camera much too far to the lower right, and while the target is in the right direction, it isn’t far enough towards the upper left, and stops well short of the building. Now it looked like the difference was about a 1:4 ratio for both the camera position and the target location relative to that. So I divided all the numbers by 4 for RST, and low and behold the camera position was now right. BUT, the target position was now seemingly half what it should be. So I tried just diving by two for the last two lines of the camera creation part. The target is still way too close to the camera, and is looking almost directly to the right of the target, not left and up at 45 degrees. Arg! I also tried leaving the target numbers as from the RAC test. Nope, distance from camera is better, but not right, and direction is almost 180 off, but not exactly.

    So then I tried just cutting off the last two lines, that define the target point. When the RST journal was finished the camera was placed, and I manually picked the appropriate location for the target. The resulting view looked about right. So then I reviewed the journal that was created from that run, and the value of my manual pick point was 707, 387. No rational relationship to the other numbers, but whatever. I copied those lines into the partial journal and ran it and... The damn target is now down and right from the camera! WTH? But the numbers are right, because when I run the journal that was created from the combination of running the partial journal and the manual pick, it WORKS! but use those same numbers in the partial journal and it doesn’t? WTH again?

    So any journal gurus out there that can help me understand what these numbers really mean? Is this relative to the Revit origin? Relative to the Revit screen? Relative to the last mouse down point? And why would numbers that worked when selected manually suddenly not work when copied into the journal?

    The RAC version looks like this:
    Jrn.Command "Menu" , "Create a 3D view by placing camera and focus target , ID_VIEW_NEW_3DVIEW"
    Jrn.MouseMove 0 , 713 , 566
    Jrn.LButtonDown 1 , 713 , 566
    Jrn.MouseMove 1 , 713 , 566
    Jrn.LButtonUp 0 , 713 , 566
    Jrn.MouseMove 0 , 435 , 289
    Jrn.LButtonDown 1 , 435 , 289

    The first four similar lines place the camera, and if the values are divided by 4 the camera seems to go in in the right place. The last two lines place the target, and nothing seems to be working there.

    If anyone just knows off hand what these numbers means, that would be great. Otherwise, I can post the revised RST version of the journal tonight, and anyone interested can see what they can make happen.

    Any help is greatly appreciated. Thanks!
    Gordon
    Pragmatic Praxis

    #2
    Originally posted by Gordon Price View Post
    So any journal gurus out there that can help me understand what these numbers really mean? Is this relative to the Revit origin? Relative to the Revit screen? Relative to the last mouse down point? And why would numbers that worked when selected manually suddenly not work when copied into the journal?
    Gordon
    Jrn.MouseMove 0 , 713 , 566

    0: mouse button state (1 = mouse button is down)
    713: x coordinate measured from origin (upper left corner of the display window)
    566: y coordinate

    During journal playback, the display is scaled using the transforms defined in GlobalToPage and ProjectToPage. When you're splicing journal snippets from another session, include those transforms for any new picks and it should work. You may have to re-establish the display by redefining the original transforms after your splice to maintain downstream picks.

    You'll probably need to include the screen resolution directive if it's different than the original journal
    Jrn.Directive "ScreenResolution", 240, 1050
    Last edited by Scott B; February 28, 2011, 06:16 PM. Reason: screen resolution

    Comment


      #3
      Upper left corner? Good lord, no wonder all the numbers made no sense. I think just about everything uses a lower left corner origin. Will have to take a look at the transforms and see what is going on. I assume actually that they are not included in the original journal, as that is what allows the journal to run consistently no matter what screen configuration a particular session is using?
      Now the next question. When applied to a Structural Plan the exact same values behave differently compared to a Floor Plan. Any thoughts on exactly what the delta is? My hope is to get a render view that si exactly the same as RAC, otherwise the numbers are useless for comparison.

      Thanks!
      Gordon
      Pragmatic Praxis

      Comment


        #4
        The journal will run despite a difference in res. During playback Revit reads the pick coordinates from the journal so your picks will still pick even if they are off the visible portion of the screen.

        A possible difference between RST Structural Plans and the RAC Floor plans is the work plane. Also, when creating elements in structural plans, e.g. Walls, they are typically created below the active level whereas in RAC, the elements are created from the active Level up to the next level.

        Comment


          #5
          Scott, I reran everything, and it sure seems like the issue has something to do with the GlobalToPage and ProjectToProject stuff. In the journal logged as the AUBench journal is run, I get a bunch of
          Code:
          Jrn.LButtonUp	0 ,	713 ,	566
          ' 0:< Picking (GView) directives insufficiently specified ' -- picks would be unstable 'View = [Project2] Structural Plan: Level 6 'Missing directives = GlobalToProj
          But the difference between the AUBench journals is verified to be just the reference to the different views.
          All in more detail in the attached PDF, and all journal files included in the zip.

          Hope that is enough for you to easily point me in the right direction. You can download the AUBench stuff from this forum if you want to run the actual journals.

          Thanks!
          Gordon
          Attached Files
          Pragmatic Praxis

          Comment


            #6
            Gordo .. I'm late to the game .. what kind of progress have you made here , have you overcome this hurdle? I did notice you were ready to release the latest AUBench, so I assumed you are done?
            bim cad tech com

            Comment


              #7
              Hee hee. No, not done. I actually abandoned the idea of a single journal file. RST just wouldn't play nice. I am now working on a more modular approach. Shooting for a new RFOBenchmark in the near future. Depends on how some testing and journal making goes tonight.

              Gordon
              Pragmatic Praxis

              Comment

              Related Topics

              Collapse

              Working...
              X