<html>
<head>
<style>
.hmmessage P
{
margin:0px;
padding:0px
}
body.hmmessage
{
FONT-SIZE: 10pt;
FONT-FAMILY:Tahoma
}
</style>
</head>
<body class='hmmessage'>
Hi Berk<BR>
<BR>
Thanks for the suggestion. Would I be right in saying that the data in the lookup table part have to be in a specific format i.e. 3 rows of 9 floats in your example? Is there a rule of thumb about the layout of the lookout table?<BR>
<BR>
I think the file format is fine for smallish data volumes but could be a lot of work if it's a large dataset. Any possibility that a future release will allow the user to load a CSV spreadsheet file and generate data volumes directly ?<BR>
<BR>
Thanks<BR>
<BR>
Lester<BR>
<BR>
<BR><BR>> Date: Tue, 14 Oct 2008 12:06:59 -0400<BR>> From: berk.geveci@kitware.com<BR>> To: lester_anderson1963@hotmail.com<BR>> Subject: Re: [Paraview] Data range problem<BR>> CC: dominik@itis.ethz.ch; paraview@paraview.org<BR>> <BR>> Lester,<BR>> <BR>> Your best choice is probably going to be VTK ascii format. It is a<BR>> very simple format. A uniform rectilinear grid would look like:<BR>> <BR>> # vtk DataFile Version 3.0<BR>> vtk output<BR>> ASCII<BR>> DATASET STRUCTURED_POINTS<BR>> DIMENSIONS 3 3 3<BR>> SPACING 1 1 1<BR>> ORIGIN 0 0 0<BR>> POINT_DATA 27<BR>> SCALARS RTData float<BR>> LOOKUP_TABLE default<BR>> 60.7635 107.555 80.5241 85.9694 156.706 105.73 37.3531 84.1445 57.1137<BR>> 100.634 171.37 120.395 149.785 260 169.546 77.2238 147.96 96.9844<BR>> 60.7635 107.555 80.5241 85.9694 156.706 105.73 37.3531 84.1445 57.1137<BR>> <BR>> A curvilinear grid is easy to write as well. The only additional<BR>> information needed is the point coordinates.<BR>> <BR>> More detail can be found here: http://vtk.org/pdf/file-formats.pdf<BR>> <BR>> -berk<BR>> <BR>> <BR>> On Tue, Oct 14, 2008 at 11:28 AM, Lester Anderson<BR>> <lester_anderson1963@hotmail.com> wrote:<BR>> > Hi Domink<BR>> ><BR>> > Sounds like I do't know what the Paraview definition of plain raw format is.<BR>> > I have attached the simple test file - just 3 points with a scalar value. It<BR>> > has a header section but paraview should recognise that anyway.<BR>> ><BR>> > Anything look odd in there?<BR>> ><BR>> > Thanks<BR>> ><BR>> > Lester<BR>> ><BR>> >> From: dominik@itis.ethz.ch<BR>> >> To: lester_anderson1963@hotmail.com<BR>> >> Subject: Re: [Paraview] Data range problem<BR>> >> Date: Tue, 14 Oct 2008 08:25:22 +0200<BR>> >> CC: paraview@paraview.org<BR>> >><BR>> >> Hi, if you have tried all datatypes in Paraview then chances are very high<BR>> >> your data is not plain raw format but includes some header etc. I don't<BR>> >> have<BR>> >> anything at hand in raw format but would be surprized there is none in VTK<BR>> >> example data.<BR>> >> PS. I think some little programming might often help you out. You might<BR>> >> want<BR>> >> to look at python if you want to keep it at minimum (plus it integrates<BR>> >> with<BR>> >> VTK and PV).<BR>> >><BR>> >> Dominik<BR>> >><BR>> >> On Monday 13 October 2008 11:00:22 pm Lester Anderson wrote:<BR>> >> > Hi Dominik<BR>> >> ><BR>> >> > I do not have Matlab to test, but there should be no issue of<BR>> >> > 'endianness'<BR>> >> > as I work on Windows and use PC-based editors. The file type does not<BR>> >> > seem<BR>> >> > to be the issue, but more the data within (just guessing).<BR>> >> ><BR>> >> > I have tried all the data type variants without success: other than<BR>> >> > being<BR>> >> > able to read the points in.<BR>> >> ><BR>> >> > Didn't seem to make much odds using xyzc (0 1 2 3) or (0.0 1.0 2.0 3.0)<BR>> >> ><BR>> >> > Do you have an example raw data file? It would be so much more<BR>> >> > convenient<BR>> >> > to be able to load up data via CSV (but that currently doesn't work).<BR>> >> ><BR>> >> > What would be the best way of loading data into Paraview that does not<BR>> >> > require any programming or major work?<BR>> >> ><BR>> >> > Lester > From: dominik@itis.ethz.ch> To:<BR>> >> > lester_anderson1963@hotmail.com><BR>> >> > Subject: Re: [Paraview] Data range problem> Date: Mon, 13 Oct 2008<BR>> >> > 09:03:58<BR>> >> > +0200> CC: paraview@paraview.org> > A simple test you could do is to try<BR>> >> > to<BR>> >> > read the raw binary file into matlab. > there you can also specify both<BR>> >> > endianness and precision (double, int, etc.). > I bet the reason you do<BR>> >> > not<BR>> >> > get the right numbers is that you import a wrong > datatype.> > Dominik><BR>> >> > ><BR>> >> > On Sunday 12 October 2008 11:54:36 pm Lester Anderson wrote:> > I have<BR>> >> > used<BR>> >> > a plain text editor to write a simple ascii structure, and also> > tried<BR>> >> > using Surfer (Golden Sofware) to prepare .DAT file do the same (eg> ><BR>> >> > space<BR>> >> > delimited or comma delimited). Basically I am testing a simple ascii> ><BR>> >> > text file structure and it ckearly loads the data (points plot where<BR>> >> > they><BR>> >> > > should be), but it is unclear how to set the data type. If we are<BR>> >> > dealing> > with positive an negative numbers then I assume that means we<BR>> >> > need a signed> > integer or real number type.> >> > Perhaps if you could<BR>> >> > show an example data file that works in the raw import> > as an<BR>> >> > xyz,value<BR>> >> > it might be clearer as to how to proceed. Thanks for the> > help, I'm<BR>> >> > sure<BR>> >> > I'm missing something very obvious here just can't put my> > finger on<BR>> >> > it<BR>> >> > yet!> >> > Lester> From: dominik@itis.ethz.ch> To:<BR>> >> > lester_anderson1963@hotmail.com>> > Subject: Re: [Paraview] Data range<BR>> >> > problem> Date: Sun, 12 Oct 2008 20:58:52> > +0200> CC:<BR>> >> > paraview@paraview.org> > How do you write the data to a file?> >> ><BR>> >> > Dominik> > On Sunday 12 October 2008 02:36:27 pm Lester Anderson wrote:><BR>> >> > >><BR>> >> > > Hi Dominik> >> > So how can I define the correct datatype? For<BR>> >> > > example,<BR>> >> > if> > it is just signed> > integers like the test I made (-5 to +5), or<BR>> >> > real> > numbers (eg -2.54 to> > 5.55). I set the file to LittleEndian as<BR>> >> > I'm on a> > PC> >> > Importing the data via the raw format is working,<BR>> >> > it's<BR>> >> > just> > working out the> > correct type to make it work!> >> > Thanks<BR>> >> > for<BR>> >> > the help>> > >> > Lester > From: dominik@itis.ethz.ch> To:<BR>> >> > paraview@paraview.org>> > Subject:> > Re: [Paraview] Data range problem><BR>> >> > Date: Sun, 12 Oct 2008> > 12:17:03 +0200>> > CC:<BR>> >> > lester_anderson1963@hotmail.com> > You most likely> > imposed wrong> ><BR>> >> > datatype when importing the raw stuff.> > Dominik> > On> > Sunday 12<BR>> >> > October> > 2008 10:55:18 am Lester Anderson wrote:> > Hello> >> >> > I<BR>> >> > have<BR>> >> > just created> > a test as a set of random values:> >> >> >> ><BR>> >> ><BR>> >> > 'x','y','z','grav'0,0,0,20,0,1,30,0,2,50,1,0,-10.1.1,-30,1,2,-50.2.0,10,2,1<BR>> >> >> >> >>> >> ><BR>> >> ><BR>> >> > > >,20,2,2,51,0,0,-31,0,1,-41,0,2,-11,1,0,01.1.1,-21,1,2,21.2.0,11,2,1-11,2,2<BR>> >> >> >> >,0><BR>> >> > >2,0,0,02,0,1-52,0,2,-42,1,0,-12.1.1,12,1,2,22.2.0,52,2,1,42,2,2,4>> ><BR>> >> > > >><BR>> >> > >> > It is a cube of point values (x,y,z) with a test scalar (grav).<BR>> >> > >> > The> ><BR>> >> > data> >> > range of 'grav' is -5 to +5> >> > I imported the data into> ><BR>> >> > Paraview as a> > raw(binary) format file which seemed> > ok, with the> ><BR>> >> > extents: 0 to 2 on> > (x,y,z). The result was a Image (uniform> >> ><BR>> >> > Rectilinear Grid)> >> > When I> > look at the data information it shows<BR>> >> > the> > data range as 2573, 13612> >> > (min,max) - what is going on<BR>> >> > here?><BR>> >> > >> >> > Thanks> >> > Lester> >> >> ><BR>> >> > _________________________________________________________________> ><BR>> >> > Make><BR>> >> > > a> > mini you and download it into Windows Live Messenger> >> >> ><BR>> >> > http://clk.atdmt.com/UKM/go/111354029/direct/01/> > > > -- > Dominik> >><BR>> >> > ><BR>> >> > Szczerba, Ph.D.> Computational Physics Group> Foundation for Research<BR>> >> > on><BR>> >> > >> > Information Technologies in Society> http://www.itis.ethz.ch> >> ><BR>> >> > _________________________________________________________________> ><BR>> >> > Make<BR>> >> > a> > mini you and download it into Windows Live Messenger> >> ><BR>> >> > http://clk.atdmt.com/UKM/go/111354029/direct/01/> > > > -- > Dominik> ><BR>> >> > Szczerba, Ph.D.> Computational Physics Group> Foundation for Research<BR>> >> > on> ><BR>> >> > Information Technologies in Society> http://www.itis.ethz.ch> ><BR>> >> > _________________________________________________________________> ><BR>> >> > Make a<BR>> >> > mini you and download it into Windows Live Messenger> ><BR>> >> > http://clk.atdmt.com/UKM/go/111354029/direct/01/> > > > -- > Dominik<BR>> >> > Szczerba, Ph.D.> Computational Physics Group> Foundation for Research on<BR>> >> > Information Technologies in Society> http://www.itis.ethz.ch<BR>> >> > _________________________________________________________________<BR>> >> > Get all your favourite content with the slick new MSN Toolbar - FREE<BR>> >> > http://clk.atdmt.com/UKM/go/111354027/direct/01/<BR>> >><BR>> >><BR>> >><BR>> >> --<BR>> >> Dominik Szczerba, Ph.D.<BR>> >> Computational Physics Group<BR>> >> Foundation for Research on Information Technologies in Society<BR>> >> http://www.itis.ethz.ch<BR>> ><BR>> ><BR>> > ________________________________<BR>> > Try Facebook in Windows Live Messenger! Try it Now!<BR>> > _______________________________________________<BR>> > ParaView mailing list<BR>> > ParaView@paraview.org<BR>> > http://www.paraview.org/mailman/listinfo/paraview<BR>> ><BR>> ><BR><BR><br /><hr />Try Facebook in Windows Live Messenger! <a href='http://clk.atdmt.com/UKM/go/111354030/direct/01/' target='_new'>Try it Now!</a></body>
</html>