Announcement

Collapse
No announcement yet.

MT Predictor

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

  • #16
    I have no idea what the problem is in this case. I was just thinking that if Adobe Photoshop was crashing, I would call Adobe to solve the problem and not Microsoft. Even though Photoshop makes calls to the Microsoft Win32 API.

    The Desktop API is not written in VB. It is C++ I believe. Although maybe you meant MT Predictor was VB (I've never used it)? The API is robust enough for client apps, assuming your client app is written with the knowledge of the limitations of the Desktop API. I have an app that I use for trading that connects to the Desktop API. Zero problems and it works everytime I run it. Zero crashes. That said, the Desktop API does indeed have limitations, as acknowledged many times by Robi. They know that and they do offer more "powerful" APIs.

    Cheers... George

    Comment


    • #17
      Obi-One, which is one of my nicknames, you are correct that the API used is not the same as your other applications. We are working with MT Predictor to "upgrade" their API to one that is more robust and hopefully they will complete that conversion shortly. As I've said before and defended by Genspoo, the API is a much cheaper product and has innate limitations (Microsoft's specifically). The business case of what the software does that usually determines which one is developed against.

      Jan upgraded his system reported that things worked better - 256 MB of RAM was not enough. It works without a hiccup on my desktop, but I am also not really using that hard.

      Genspoo brings up a good point. There are many instances when the API works exceptionally well for the application it is intended for, but there are times it does not. I would say 9 times out of 10, I end up finding not the most efficient use of the
      API, which leads to it crashing. By changing the logic of how it is used, it creates more stability and there are numerous commercial applications where the API is great fit and works as designed.

      Comment


      • #18
        MT Predictor

        Dear Mr Papp,

        Upgrading to 512 MR Ram unfortunately did not help with MT Predictor. I stopped using the trial today. Having to start up MTP and/or eSignal 5 to 10 times a day is too much for me.

        One of the error messages I receive:

        at IESignal.HooksClass.ReleaseHistory(Int32 lHandle)
        at MTPEod.Form1.tbTop_ButtonClick(Object sender, ToolBarButtonClickEventArgs e)
        at System.Windows.Forms.ToolBar.OnButtonClick(ToolBar ButtonClickEventArgs e)
        at System.Windows.Forms.ToolBar.WmReflectCommand(Mess age& m)
        at System.Windows.Forms.ToolBar.WndProc(Message& m)
        at System.Windows.Forms.ControlNativeWindow.OnMessage (Message& m)
        at System.Windows.Forms.ControlNativeWindow.WndProc(M essage& m)
        at System.Windows.Forms.NativeWindow.Callback(IntPtr hWnd, Int32 msg, IntPtr wparam, IntPtr lparam)

        My "Problem log" is apparently too big to attach to this forum - it consists of 10 pages of problems over the last two weeks and I did not even record all of the probems and error messages.

        Just to mention: my computer otherwise works fine. No virusses, no ads, no spyware. A thorough check with an extensive diagnostic program did not show any problems.

        When the software runs properly, I would be interested to try it again.

        I would like to thank the other forum participants for their contribution.

        With kind regards

        Jan

        Comment


        • #19
          Unfortunately, the limitations of the Desktop API were too great on certain workstations for MT Predictor to work as expected. The way the Desktop API works, it allocates a lot of RAM, and the way MT Predictor makes its calculations, where the requests are in spaced out in specific intervals can pose problems. Hopefully, the memory upgrade was not too costly. I think you will consider that money well spent in the long run. I have 1 GB RAM on my workstation and did not experience the problem when certifying MT Predictor, which explains why the issue was not made apparent.

          We are currently working rather diligently with them to transform the data request functionality to a lower level interface and hope to have the work completed by the end of the year, so be sure to stay in tune as it has been fast tracked.

          For any other MT Predictor users that are experiencing any difficulties, please be patient as relief is on the way. On behalf of both organizations, we apologize for your troubles.

          Comment

          Working...
          X