Android 4.4 has been announced and brings an interesting new feature: Renderscript Compute is now accessible from the NDK from C++ code. There is very little documentation atm, or at least I haven’t found it but you should download NDK r9b at the time of writing to see a HelloComputeNDK sample in the samples folder. It looks like a fully native solution, i.e. you do not need to write any Java code to access it afaik. The Android 4.4 release notes say that you should be able to access all renderscript functionality including intrinsics and user-defined kernels from the NDK. This is quite a nice development and kudos to the Renderscript team, but they do desperately need to address other concerns such as documentation. I still do think that they need to provide access to a lower-level API, such as OpenCL, to complement the high-level Renderscript. But I guess such is life.
As an aside, release notes also say that Android 4.4 enables GPU acceleration of Renderscript (I am assuming only for Filterscript) for all recent Nexus devices (Nexus 4, Nexus 5, Nexus 10, Nexus 7 2013) with the exception of Tegra 3 powered Nexus 7 2012, which has an ancient GPU architecture.
Some ideas are as follows:
1. Touch responsiveness is an objectively measurable quantity. I think high speed cameras can play a very important role in this field.
Some good initial work has been reported by Tech Report.
It is unfortunately increasingly common to hear “Benchmarks don’t matter” and then some semi-coherent rant about user experience and “smooth” UIs. I think all it means is that the writer had no idea how to measure the touch responsiveness 😛
2. Application launch times: Application launch times can again be measured objectively. For slower apps, you can use a stopwatch and for small fast-launching apps, you can again use a high speed camera.
4. Proper application benchmarks need to be more common instead of synthetics. For example, Photoshop Touch can perhaps be used much as Photoshop benchmarks are now common on the desktop.
5. Synthetic benchmarks are poorly written and poorly understood. For example, Linpack Android version seems to be a poorly coded benchmark. The megaflops reported from the benchmark are far off the capabilities of the chips tested. I am looking at making a better one when I get time. For floating point tests, really what you want are accurate and separately reported measures of fp32 performance, fp64 performance and fp32-with-NEON performance.
6. Further, synthetics should properly report both single threaded and multithreaded numbers. (Linpack does do this but many other benchmarks don’t). I think single-thread performance is underestimated on mobile with most websites reporting benchmarks from multithreaded tests. However, few apps use 4 cores in any modern Android phone. And no, you don’t need 4 cores to multitask.