Re: Large 2D Graph Additionally

From: Chris B (CSBooth_at_noSpam.com)
Date: 04/28/04

  • Next message: Craig: "Managed Direct3D for 2D games" 
    Date: Wed, 28 Apr 2004 21:23:08 +0100

    Thanks for the response Philip

    "Philip Taylor [ATI]" <ptaylor@online.newsgroup-poster.ati.com> wrote in
    message news:uRQtMPVLEHA.1120@TK2MSFTNGP11.phx.gbl...
    > there is "handle the dataset" and there is "handle the dataset optimally"
    >
    > if you control the hw specification and can constrain the hw to recent hw
    > with 32-bit index buffer support and lots of video memory, you may be able
    > to load and render a dataset this large. you will need to test this,
    though.
    OK I will try this. I only need this on a specific machine. This is just a
    test tool for an "in house" application. Any distribution would not involve
    graphics. Depending on what I can achieve I will consider updating the
    graphics card etc. if appropiate.

    >
    > usually loading a large dataset with no abstraction is a bad idea, if you
    > have to run on just any old hw. typically you have to develop an
    abstraction
    > that allows you to work with subsets of the data at any one time. moving
    > them in/out as needed.

    I agree. But this is the bit where I was hoping to get away with "simple
    coding", on the basis that directX would take a subset of the vertex buffer
    as required. (Assuming I have managed to get the vertex buffer loaded) I
    would be happy with a rendering in 5 seconds say. Then merge with anything
    interactive.

    >
    > zoom is a function of the camera and scaling, so you can handle that with
    > matrix maths.

    Yes, I am happy with that. And I only want a 2D view, with simple scaling
    and translation only.

    >
    > have you calculated a vertex buffer memory budget for such a dataset? and
    > the bus bandwidth required to get it over to the card? and the vertex
    > throughput required by the card to render this dataset at interactive
    > framerates?
    No. I do not have any idea how do this. Thanks for the suggestions, but any
    updates on this at this stage will probably lose me. I will for now adhere
    to suggestion 1. Of interest though, is that that the vertex set is largely
    static. I guess that on a highly specialised graphic card any transform can
    be done directly on a vertex set loaded on that card?

    >
    > finally, DirectX isnt about making the graphics coding simpler, its about
    > providing access to hw acceleration through a thin layer. think of it that
    > way and you will be much happier.
    No, I am much happier thinking of DirectX making graphics coding simpler.
    ;-)

    Thanks.

    Chris

    >
    > "Chris" <Chris@nospam.com> wrote in message
    > news:%23n7MqwTLEHA.1644@TK2MSFTNGP09.phx.gbl...
    > > Additionally, each edge can be plotted as a path, consisting of, on
    > average
    > > 8 points per edge.
    > > Using a straight line for each edge is acceptable for my usage,
    selecting
    > a
    > > single mid point from the path would be better, up to optimum of
    plotting
    > > all points on the path fro each edge.
    > >
    > > Obviously I require usual operations of zoom etc.
    > >
    > > Also I am familiar with the idea of sorting the nodes/edges into
    > rectangualr
    > > areas, and by level, then drawing only level 1, 2 say, on a full view
    of
    > > the graph, showing the lower and more numerous lower level node/edges
    only
    > > for appropriate rectangles when zoomed in.
    > >
    > > I am more interested in whether the graphics coding can be made much
    > simpler
    > > because directX can handle a graph of this size. --- ie I can feed the
    > data
    > > in, and just select which layers I wish to display according to the
    zoom.
    > >
    > >
    > > "Chris" <Chris@nospam.com> wrote in message
    > > news:OQ3k3hTLEHA.624@TK2MSFTNGP11.phx.gbl...
    > > > I will be, in next month or so, drawing a graph (map) with around 2
    > > million
    > > > nodes, 3 million edges, using probably 4 layers. Using c# and managed
    > > > directx is it feasible to consider? The graph will be static as
    such,
    > > but
    > > > I would use another layer to hilite routes say.
    > > >
    > > > Thanks for any replies.
    > > >
    > > > Chris
    > > >
    > > >
    > >
    > >
    >
    >

  • Next message: Craig: "Managed Direct3D for 2D games"

    Relevant Pages

    • Re: Which method is better?
      ... The scene I have is one big mesh exported from 3DsMax. ... one vertex buffer, and there's only one index buffer. ... will reference to vertices all throughout the vertex buffer, ...
      (microsoft.public.win32.programmer.directx.graphics)
    • Re: Detecting mouse click on a particular triangle
      ... Ok i posted this before reading your replies... ... > there were functions to do that with just a vertex buffer in the utilities ... >> verts make up the next triangle. ...
      (microsoft.public.win32.programmer.directx.graphics)
    • Re: basic performance questions - vertices/second=>fps
      ... Especially for the vertex buffer handling and locking. ... that a smart split of the vertex buffer will have significant ... which most likely won't support vertex shader 3. ...
      (microsoft.public.win32.programmer.directx.managed)
    • Re: Detecting mouse click on a particular triangle
      ... My confusion is that these codes ... keep a second buffer around with a lockable z buffer ... there were functions to do that with just a vertex buffer in the utilities ... > verts make up the next triangle. ...
      (microsoft.public.win32.programmer.directx.graphics)