From: Satheesh Ganapathi Subramanian (gsat++at++evl.uic.edu)
Date: 08/31/2000 10:49:40
Hi Angus,
Thanks for your detailed explanation. But I've been having some problems
trying to get the
I tried doing a subload with this understanding. My destination texture
has a dummy data filled with all blue (1024x512). My subload is dummy data
with all white (760x486). When I do a subload, what I expected was that
the portion of the texture that was subloaded would be white and the rest
of it would be blue. What I got instead was that the portion of the
texture that was subloaded was white but the rest of the texture was
having corrupt data. I couldn't figure out why that was happening.
Also, I've another question. What is the difference between using
pfTexture::subload() and using pfTexture::setLoadImage() inconjunction
with setLoadOrigin() and setLoadSize() with the texture format enabled for
PFTEX_SUBLOAD_FORMAT and PFTEX_BASE_AUTO_SUBLOAD. I tried both the
approaches and the result was exactly the same as described above.
Again, if there is some sample code that can illustrate the use of
pfTexture::subload() it would be of great help.
Thanks
Satheesh
>The key thing to note here is that the source image may be a different
>size form the destination image. In addition the image being subloaded
>may not occupy the full width of the source image and may not be
>positioned at the origin of the source image.
>
>If you consider this problem it becomes clear that you need to specify
>the location of the subload in the destination image, the overall width
>of the source image and the width and height of the subload from the
>source image, and the position of the origin of the subload in the
>source image.
>
>The height of the source image isn't needed because the image is stored
>column minor, row major in OpenGL, but in order to properly address
>height you need to know the image width for the stored source image
>unless you are subloading a full width image in which case xsize is all
>you need.
>
>So, srcwid is needed to support source images of widths which do not
>match the width of the image in texture memory, xsrc and ysrc is the
>origin of the download in the source image (which doesn't have to be the
>same as the origin in the destination image).
>
>There is a degree of redundancy because an application could simply
>adjust the pointer to address the appropriate pixel of the source image
>and zero the xsrc and ysrc values, but that's more tricky for the
>application to code for I suppose.
>
>Strictly speaking the manual description is wrong w.r.t. srcwidth.
>
>Cheers,Angus.
>
>Satheesh Ganapathi Subramanian wrote:
>>
>> Hi Performers,
>>
>> I'm trying to use pfTexture::subload or pfSubLoadTex and I'm not able to
>> understand the meaning of - srcwidth, xsrc, ysrc - in the list of
>> paramenters to that function. I'm creating an image buffer (unsigned
>> int*), the data in which keeps changing every frame. I would like to load
>> this into texture memory using pfTexture::subload. It would be great if
>> someone can tell me what each parameter of the function call means. I'm
>> not able to get much out of the man pages.
>>
>> If anyone has some sample code that they can share, explaining how to use
>> this function, it would be of immense use.
>>
>> Thanks
>> Satheesh
>>
>> -----------------------------------------------------------------------
>> List Archives, FAQ, FTP: http://www.sgi.com/software/performer/
>> Submissions: info-performer++at++sgi.com
>> Admin. requests: info-performer-request++at++sgi.com
>
>--
>For Performer+OpenGL tutorials http://www.dorbie.com/
>
>"In the middle of difficulty lies opportunity."
> --Albert Einstein
>
===================== members.xoom.com/malargal ========================
Satheesh GanapathiSubramanian
gsat++at++evl.uic.edu
Work: Home:
Research Assistant 901, S.Ashland Av, Apt# 107-A
The Electronic Visualization Laboratory Chicago, IL-60607
851, S.Morgan St., Room# 1120 SEO
University of Illinois at Chicago Ph: 312-492-8164
Chicago,IL-60607
Ph: 312-996-3002
Fax: 312-413-7585
=========================================================================
This archive was generated by hypermail 2b29 : Thu Aug 31 2000 - 10:49:50 PDT