• Advertisement


  • Content count

  • Joined

  • Last visited

Community Reputation

100 Neutral

About serg3d

  • Rank
  1. Quote:Original post by frob Your experience is with C++, not java. That should make your choice obvious. I wouldn't completly agree. While I myslef am doing development on Symbian C++ it's only because applications I'm working on are extremely CPU-hungry(AR), so java is out. Symbian C++ framework have notoriously high learning curve, documentation is often lacking, UI framework classes and APIs are huge and not well documented. However while threshold is quite high as soon as you managed to build your framework and go to "meat" of the game - graphics/gameplay, it's pretty much normal C++ and OpenGL ES. Of cause if you target phones without floating point (VFP) you have to do all the math in fixed point, but as I understand you have to use fixed for java either. So I'd say for small, short-term project go with java. For huge, long, performance-critical projects go with Symbian C++.
  2. I’ve finished porting my AR Game prototype to Symbian 9.* (Nokia N95 and like). OK, what have I learned ? 1. 300+ Mhz CPU is quite enough for multimarker-tracking based AR. 2. Floating point and fixed point calculations. First version was for Symbian 7.0s, without hardware floating point support, so I tried to make exposure to floating point minimal. As result, on the Nokia N95, which have floating point (VFP) I don’t see any difference in performance between software floating point version and hardware floating point. 3. Symbian platform fragmentation. It bad. 4. Use optimization ! performance difference between O3 and no optimization is huge.(I use GCCE compiler) or you can read it here. blog posts [Edited by - serg3d on October 29, 2008 12:29:02 AM]
  3. iPhone 3D Engine

    What is the situation with floating point on iPhone ? For symbian 9.* hardware FPU is not supported in SDK but could be accessed with some cfg modifications. Though there are reports that at least for some devices fixed point still faster. Is hardware floating point supported in SDK for iPhone? If not physics could be quite difficult.
  4. Robust contour detection?

    You can also use morphological operations to get rid of artifacts like tails, branches etc.
  5. Yep, I have come to similar conclusion. Camera is not capturing enough sunlight because of limited FOV. I think the other solution, instead of letting artist edit images manually, I can increase the weight of top hemisphere spherical harmonic. That way sun will go into mix better. edit: May be pass image through some non-linear filer, like x**2. Make bright part brighter, and dim part dimmer, and after that average.
  6. I'm using image based lighting with spherical harmonics, produced from cubemap hdr images. However there is a big problem with my implementation: Outdoor images with blue sky cause top of the lighted object become blueish. The same problem with indoor cubemaps - color of the wall transfer to the object. We don't see any top-blue humans outdoor, this blue color is neither physical nor realistic. Any idea how get rid of this problem in physically correct way ? I can decrease image saturation, but that would cause *all* light become white, I still want colored light too. Any advices, pointers to some articles ?
  7. Quote:Original post by bestseclub2 Quote:Original post by GuyWithBeard Quote:Original post by bestseclub2 OK,so this means i can use gmail or yahoo accounts to register right?Some applications dont allow gmail or any other free emails... Yes Yeah,cool!Same for the SDK,right? Yes, SDK is free. However there are some trikcs with it's installations - don't install anything into /Program Files/ - that cause name being too long and could cause problem with some SDK versions. Another - you need Active Perl (also free) to be installed to use SDK. But not the latest version of Active Perl(5.10.*). SDK works only with 5.6.* Active Perl. Also some of the examples can not be moved to other directory unmodified - they have binary paths hardcoded in their pkg files. And biggest of all - the problem of Symbian Signing. That is too long to discuss here, there is plenty info on the net.
  8. I have used GSL for SVD some times ago. It has C interface. IIRC compilation was not exactly trivial but not too difficult either. http://www.gnu.org/software/gsl/
  9. Computer Vision

    Quote:Original post by VprMatrix89 I'm interested in learning the techniques of 'computer vision', specifically taking a 2d image and defining geometry from it. It doesen't have to start out complex.. Don't be put off by intimidating responses. In it's simplest form it's extremely easy, if you fluent with elementary vector/linear algebras. If you don't want code himself just download ARToolkit or ARToolkit+ - open sourced marker registration pakages. But I'd recommend to start with basics - it would also help understand ARToolkit better. Start with simple black square marker on the white background. Try to build coordinating system from it. 1.Binarize your image into black/white (simple threshold would do for start. use adaptive threshold for real work). 2. Segment your image into connected components (google for algorithm if you have problems with it - there is a plenty). 3. Approximate each component with polygon - make "line segmentation" of the boundary. Again there are several method google for it. Also I did in depth explanation at usenet sci.image.processing couple of years ago. Actually that is hardest part of "simple" approach. 4. Chose components consisted of four lines. - suspected markers 5. build coordinate system from the lines. Suppose V0, V1, V2, V3 - vectors representing lines (counterclockwise) Use V0 as X, V1 as Y Z axis could be calculated as (V0 X V2) X (V1 X V3) "X" -is vector crossproduct You have your coordinate system. You can use additional structures (inside the marker for example) to choose which corner of the marker to use as zero. 6. Download Artoolkit manual and read it. 7. Subscribe to ARToolkit mailing list. I reiterate - if you are fluent with vectors, matrices and experienced coder it's not a big deal. The problem starts then you try get rid of markers, or have bad lighting conditions, or use several markers, or want better stability etc. However if you have problem understanding even this simplified outline better don't start right away. Read some books on matrix/vector algebra, projective geometry and image segmentations instead. Or just use ARToolkit blindly, as end-user.
  10. Processor development for game AI

    Quote:Original post by harsh1987 ... 1)C macro style instruction set, that is same code for all data widths; code and data treated differently in my processor. I think that code be a disadvantage for functional programming.
  11. Quote:Original post by 4everlive Can I create (hardware accelerated) 32 bit RGBA textures even if eglChooseConfig supports no alpha on the surface? Yes. You don't need alpha on the rendering surface, only in texture. Quote: glEnable(GL_BLEND); glBlendFunc(GL_SRC_ALPHA, GL_ONE_MINUS_SRC_ALPHA); no joy on the mobile Should work. May be you havn't created the texture properly - make sure your specify RGBA texture, with proper value in A, with all May be it's a driver bug. Or may be device doesn't support alpha blending. From your code snipped you are using multitexture (I didn't check your code, your should make sure your texture combination is correct). Try single texture for test, with simple GL_MODULATE.
  12. But SDK doesn't officially support native code, last I heard. That is severely limiting it's appeal.
  13. It seems to me you are mistaking two completely different objects - alpha channel and alpha component of the texture. RGBA is an alpha texture - texture with alpha component - (semi)transparent texture. Alpha channel is the alpha component of rendering buffer - your rendering window. If you want to use alpha channel you should set it during OpenGL window initialization -eglChooseConfig But you don't need alpha channel if you want use (transparent) RGBA textures. Just use them with glEnable(GL_BLEND) - no alpha channel needed.
  14. Hmm, what do you need alpha channel for ? Why don't you use alpha textures ? Alpha channel - destination alpha - used not often, for postprocessing or for render target.
  15. Math help - Integral Equations and more

    You have to get introductory course in the functional analysis first, probably. http://en.wikipedia.org/wiki/Functional_analysis It's essentially a calculus with variable functions. You also have to have some knowledge of linear and abstract algebra too. Before the start you should know Linear algebra - vector space, matrices, eigenvalues and eigenvectors, decompositions. Introduction to abstract algebra - abstract vector spaces at least. After that you can advance to introduction to functional analysis - functional spaces, linear operators, norms, basic theorems. At this point you will understand which methods you problem require and can find your way yourself.
  • Advertisement