Text archives Help
- From: "jimkress_58" <jimkress_58@kressworks.org>
- To: <iv3d-users@sci.utah.edu>
- Subject: [IV3D-USERS] RE: Re: RE: Re: RE: Re: Help on data input and Axis Display
- Date: Sat, 16 Jan 2010 15:10:54 -0500
>
Unfortunately the raw reader is generally based around an implicit data
>
layout, i.e. data start at (0,0,0) and change in X slowly and in Z
>
quickly. There's a radio button for "text" input, but looking through
>
the code, all our IO is unformatted so I'm a little wary that the text
>
path is relatively untested.
Yeah. I tried the "text" option and my computer hung.
>
I would doubt it is meaningful, because I don't think the conversion
>
step did something intelligent. Then again, you know your data better
>
than I do :). Do the generated images "look" like your data?
Hard to tell if I cannot determine their x,y,z spatial location. Maybe I'll
try slicing the data and see if the contoured intensity for each slice
corresponds to the volumetric information presented.
>
As a side note, these data are pretty low res. If they're low
>
resolution for visualization reasons, and you can obtain higher res,
>
do so. People don't tend to believe me until they've done it a couple
>
times, but (seriously!) the size of the data is irrelevant to IV3D.
I agree but it's the best we can do with the experimental equipment and the
chemistry involved.
>
C++. Adding a new reader is pretty easy/self-contained, so really if
>
you've got someone that knows any ALGOL-derived language && is willing
>
to google some terms, that will be enough that we could walk them
>
through what's needed.
ARGH! You've hit the chink in my armor. However, I can try to struggle
through it (I have 30 years of FORTRAN experience but am just starting to
learn C++). What's the best existing reader to use as a starting template?
Jim
>
-----Original Message-----
>
From: tom fogal [mailto:tfogal@alumni.unh.edu]
>
Sent: Saturday, January 16, 2010 1:48 PM
>
To: iv3d-users@sci.utah.edu
>
Subject: [IV3D-USERS] Re: RE: Re: RE: Re: Help on data input and Axis
>
Display
>
>
"jimkress_58" <jimkress_58@kressworks.org> writes:
>
> > From a (very) quick read, I'd say this isn't a point cloud. It
>
> > looks as if your data's ranges are [1:16],[2:9],[7:15], and further
>
> > there are no `holes' in the data (i.e. no combination of those
>
> > three indices which is not represented). If this is true, you've
>
> > got regular, gridded data, which is exactly what we need for
>
> > ImageVis3D.
>
>
>
> OK.
>
>
>
> However, when asked, ImageVis3D appears to import the file. It
>
> just asks for the extent (number of points?) in each direction,
>
> i.e. 16,8,9.
>
>
Yeah, you're getting the raw conversion dialog, which (if you say
>
`yes') comes up when we can't figure out any metadata that would help
>
us read the file.
>
>
Unfortunately the raw reader is generally based around an implicit data
>
layout, i.e. data start at (0,0,0) and change in X slowly and in Z
>
quickly. There's a radio button for "text" input, but looking through
>
the code, all our IO is unformatted so I'm a little wary that the text
>
path is relatively untested.
>
>
Worse, detecting errors at that stage is essentially impossible,
>
because we don't know anything about the file. So, ImageVis3D will
>
happily do whatever conversion you've asked for, without regard to
>
whether that conversion makes any sense.
>
>
> Is this imported volumetric image I observe meaningful?
>
>
I would doubt it is meaningful, because I don't think the conversion
>
step did something intelligent. Then again, you know your data better
>
than I do :). Do the generated images "look" like your data?
>
>
>
As a side note, these data are pretty low res. If they're low
>
resolution for visualization reasons, and you can obtain higher res,
>
do so. People don't tend to believe me until they've done it a couple
>
times, but (seriously!) the size of the data is irrelevant to IV3D.
>
>
> Is the grey scale presented in the image controlled by the number of
>
> bytes allowed by the graphics card?
>
>
You're seeing grey because the default renderer is based on 1D transfer
>
functions, and the default 1D transfer function is a ramp for all 4
>
color channels. You can modify the current function with the "1D
>
Transferfunction editor", accessible from the Workspace menu. More
>
details are in the manual.
>
>
> What would happen if I changed the file so that the intensity for
>
> points outside the current data ranges, that is for points [1:16],
>
> [1,10:16] [1:6,16], are zero?
>
>
I'm not sure I follow that syntax correctly, but it sounds like you
>
want to add more points to your domain, outside the original domain,
>
initialized to 0? With the default parameters, the only thing that
>
will happen is that your domain size will increase. This is because
>
the default transfer functions are set to remove / not visualize
>
0-valued data. You could change this in the 1D TF editor by making the
>
leftmost components non-zero.
>
>
> > If there's anybody in your group who can code, have them follow up
>
> > on tuvok-developers [1].
>
>
>
> Code in what language?
>
>
C++. Adding a new reader is pretty easy/self-contained, so really if
>
you've got someone that knows any ALGOL-derived language && is willing
>
to google some terms, that will be enough that we could walk them
>
through what's needed.
>
>
> Also, is the amount of video memory just that which is on the (in
>
> my case) embedded controller or does it include the TOTAL Available
>
> Graphics Memory = Dedicated Video Memory + Share System Memory?
>
>
To be honest, we ignore the graphics memory setting for now, due to
>
limitations of the rendering API. Someday in the distant future I hope
>
that will be fixable, at which point the "GPU Memory" slider would want
>
to know just your "Dedicated Video Memory".
>
>
> Thanks for your help. I hate to be such a nag but the software has
>
> such a huge upside for the work I am doing that could help many
>
> people.
>
>
Not a nag at all. We love to get more people using our software! =)
>
>
Cheers,
>
>
-tom
>
>
> -----Original Message-----
>
> From: tom fogal [mailto:tfogal@alumni.unh.edu]
>
> Sent: Saturday, January 16, 2010 3:41 AM
>
> To: iv3d-users@sci.utah.edu
>
> Subject: [IV3D-USERS] Re: RE: Re: Help on data input and Axis Display
>
>
>
> "Jim Kress" <jimkress_58@kressworks.org> writes:
>
> > The file is attached. It is ASCII and would meet (I believe) your
>
> > definition as point clouds. Please note: cloud is in the plural.
>
>
>
> >From a (very) quick read, I'd say this isn't a point cloud. It looks
>
> as if your data's ranges are [1:16],[2:9],[7:15], and further there
>
> are no `holes' in the data (i.e. no combination of those three indices
>
> which is not represented). If this is true, you've got regular,
>
> gridded data, which is exactly what we need for ImageVis3D.
>
>
>
> If there's anybody in your group who can code, have them follow up on
>
> tuvok-developers [1]. Otherwise I'll add an ASCII reader to my todo
>
> list...
>
>
>
> -tom
>
>
>
> [1] https://lists.sci.utah.edu/sympa/subscribe/tuvok-developers
>
>
>
> > -----Original Message-----
>
> > From: tom fogal [mailto:tfogal@alumni.unh.edu]
>
> > Sent: Friday, January 15, 2010 6:38 PM
>
> > To: iv3d-users@sci.utah.edu
>
> > Subject: [IV3D-USERS] Re: Help on data input and Axis Display
>
> >
>
> > <jimkress_58@kressworks.org> writes:
>
> > > My dataset consists of one file with many single lines of data, each
>
> > > of which contains the X, Y, Z, and Intensity measured at that X,Y,Z
>
> > > point in space. What is the best way to feed this into ImageVis3D?
>
> >
>
> > We'll need a bit more info on this format. Is there any structure to
>
> > these points, or do they essentially form a `point cloud'? Secondly,
>
> > are these ASCII or binary files?
>
> >
>
> > > Also, I need to be able to display X, Y, and Z axes with user
>
> > > configurable scaling and tic mark number, value and location on each
>
> > > axis.
>
> >
>
> > There's currently no support for this in ImageVis3D, though the
>
request
>
> > has been brought up before.
>
> >
>
> > > I can provide a sample input file if you will tell me how to get it
>
> > > to you.
>
> >
>
> > If you've got a very small example (kilobytes), sending it to this
>
> > mailing list is the best mechanism.
>
> >
>
> > If it's anything larger, use ImageVis3D's builtin "Report an Issue"
>
> > dialog, under the help menu. Add a description describe the file
>
> > format, and use the "Add files" button to add one or more example
>
> > files.
>
> >
>
> > -tom
>
>
- [IV3D-USERS] Help on data input and Axis Display, jimkress_58, 01/15/2010
- [IV3D-USERS] Re: Help on data input and Axis Display, tom fogal, 01/15/2010
- [IV3D-USERS] RE: Re: Help on data input and Axis Display, Jim Kress, 01/15/2010
- [IV3D-USERS] Re: RE: Re: Help on data input and Axis Display, tom fogal, 01/16/2010
- [IV3D-USERS] RE: Re: RE: Re: Help on data input and Axis Display, jimkress_58, 01/16/2010
- [IV3D-USERS] Re: RE: Re: RE: Re: Help on data input and Axis Display, tom fogal, 01/16/2010
- [IV3D-USERS] RE: Re: RE: Re: RE: Re: Help on data input and Axis Display, jimkress_58, 01/16/2010
- [IV3D-USERS] Re: Help on data input and Axis Display, tom fogal, 01/16/2010
- [IV3D-USERS] RE: Re: Help on data input and Axis Display, Jim Kress, 01/19/2010
- [IV3D-USERS] Re: Help on data input and Axis Display, tom fogal, 01/19/2010
- [IV3D-USERS] RE: Re: Help on data input and Axis Display, Jim Kress, 01/19/2010
- [IV3D-USERS] Re: RE: Re: Help on data input and Axis Display, tom fogal, 01/19/2010
- [IV3D-USERS] RE: Re: RE: Re: Help on data input and Axis Display, Jim Kress, 01/19/2010
- [IV3D-USERS] Re: Help on data input and Axis Display, tom fogal, 01/19/2010
- [IV3D-USERS] RE: Re: Help on data input and Axis Display, jimkress_58, 01/20/2010
- [IV3D-USERS] Re: RE: Re: Help on data input and Axis Display, tom fogal, 01/20/2010
Archive powered by MHonArc 2.6.16.