ah sorry, got some additional questions: if the hardware has a FPU, will it be automatically used, both in java and in native code? for the native code part: will the floating point stuff be transformed into softfloat instructions at compile time, or will the exceptions be caught in the kernel, and the transformation therefor be handled at runtime? thank you :D
On 22 Feb., 21:29, "[email protected]" <[email protected]> wrote: > hey, > thanks for your information, that really helped :D > > The HTC Dream has MSM7201A as CPU. > I found the following spec for > it:http://pdadb.net/index.php?m=cpu&id=a7201a&c=qualcomm_msm7201a > > can you tell me what by "ARM9 coprocessor" is meant? what is that > coprocessor for? > > On 22 Feb., 19:53, David Turner <[email protected]> wrote: > > > The HTC Dream doesn't have an FPU, or we would be using it :-) The Verizon > > Droid and Google Nexus One do have one though. > > > Keep in mind that "VFP" is really an umbrella for different ARM FPU > > technologies: > > > If I remember correctly: > > > VFPv1 is now obsolete > > VFPv2 is an optional instruction set for ARMv5/ARMv6 CPUs. > > VFPv3 is used on ARMv7-based CPU, it is not compatible with VFPv2 as far as > > I know. > > > Also, there is VFPv3-D16 and VFPv3-D32, where the difference is that the > > former provides 16 64-bit FP registers to the CPU, > > while the latter provides 32. Which one is exactly implemented depends on > > the exact CPU model. > > > Whether OpenGL is faster using floats or fixed points depends on the GPU. I > > believe that embedded GPUs only really > > support fixed point at that point (and the driver performs floats->fixed > > point conversions for you). > > > The only way to tell is to benchmark on specific devices and compare the > > results. > > > On Mon, Feb 22, 2010 at 2:26 AM, [email protected] < > > > [email protected]> wrote: > > > I'm currently developing a application on a HTC Dream aka T-mobile G1. > > > afaik the processor has a FPU coprocessor, called VFP (Vector Floating > > > Point) in ARM terms. So far I couldn't find any information about if > > > the FPU is used for floating point calculations or if those are > > > emulated in software. does anyone have any information about that? > > > the opengl implementation supports both fixed point and floating > > > point. will my applications be slower when I'm using floating point, > > > because they get transformed into fixed point internally? or is > > > floating point implemented in hardware? > > > thanks for any information in advance :D > > > > -- > > > You received this message because you are subscribed to the Google > > > Groups "Android Developers" group. > > > To post to this group, send email to [email protected] > > > To unsubscribe from this group, send email to > > > [email protected]<android-developers%[email protected]> > > > For more options, visit this group at > > >http://groups.google.com/group/android-developers?hl=en > > -- You received this message because you are subscribed to the Google Groups "Android Developers" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/android-developers?hl=en
