Hi Carsten, sorry for the thread confusion.
Yes I am concerned about precision. Of course rather with transformations and other background calculations than with the geometry data. But I am just wondering where the journey is going in terms of graphics hardware. I would expect the graphics hardware to be operating on 64Bits with all data in the long run, too - please correct me if I am wrong. I would also expect the use of floats on hardware that is (or will be) designed for doubles to become a bottle neck in the future. E.g. for an application, which's core has 64 Bit precision and OpenSG is used to do the rendering, takes mainly a lot of transformations and sometimes new mesh data. Every time a 64Bit precision transformation is copied to the 32Bit OpenSG system and later to the graphics hardware, everything needs to be converted to float. Will this lead to performance losses at the point where simulation data crosses from 64 to 32 bit? Can OpenGL handle mesh data in floats transform them with transformations given in doubles? Thank you in advance, Yours Georg. -----Ursprüngliche Nachricht----- Von: Carsten Neumann [mailto:carsten_neum...@gmx.net] Gesendet: Montag, 7. Juni 2010 02:24 An: opensg-users@lists.sourceforge.net Betreff: Re: [Opensg-users] Depth of Field in OpenSG Hello Georg, On 06.06.2010 11:06, Georg Wünsch wrote: > I have nothing to contribute to that particular question but... sorry an > important question that came back to me when I read these mailings. ok, to make life easier for others when searching the list archive in the future it is helpful if you start a new thread and adjust the subject line. Thanks! > What is the best way of porting a 32 bit opensg 1.8 app to 64bit& OpenSG > 2.0 lateron? > > First port to 2.0 then to 64bit or the other way round or all together in > one step? I wouldn't expect much 32/64 bit porting to be involved. The OpenSG data types all have an explicit number of bits (OSG::UInt32, OSG::Real32) so the data does normally not change size. Even if you are not using OpenSG types but float/double directly, those types do not change size when switching to 64 bit. > Some more stupid questions I am considering: > How does the graphics hardware behave when it is corresponding with a 64bit > application. the hardware mostly sees the data you upload to it and the OpenGL interfaces specify what data type you upload. > Is that happening behind the scenes or do I have to take care > and think about every type of data - if they need to be converted or not? hm, I'm confused, what data do you expect needs conversion? > E.g. do I have to convert trimesh data from float to double? that is independent of having a 32/64 bit application, so no you just keep on using floats if there is no other reason (like higher precision requirements) to switch. > Do I need to > keep the data as 64 bit values in Opensg already? Does the graphics card > work in 64bit or 32 bit at all? More recent hardware has gotten better dealing with double data without performance loss (mostly driven by the compute APIs CUDA/OpenCL etc.), I'm not sure if it can handle it without penalty in all cases yet. Cheers, Carsten ---------------------------------------------------------------------------- -- ThinkGeek and WIRED's GeekDad team up for the Ultimate GeekDad Father's Day Giveaway. ONE MASSIVE PRIZE to the lucky parental unit. See the prize list and enter to win: http://p.sf.net/sfu/thinkgeek-promo _______________________________________________ Opensg-users mailing list Opensg-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/opensg-users ------------------------------------------------------------------------------ ThinkGeek and WIRED's GeekDad team up for the Ultimate GeekDad Father's Day Giveaway. ONE MASSIVE PRIZE to the lucky parental unit. See the prize list and enter to win: http://p.sf.net/sfu/thinkgeek-promo _______________________________________________ Opensg-users mailing list Opensg-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/opensg-users
