The 2009 Mac Mini, Cont.

But hey, HDMI is supported by the 9400M, so where is it on the new Mac Mini? Apple is probably playing games with us, holding it back for some future HDMI “upgrade”.

So where's the HDMI port?

However, Apple did go with the higher performance DDR3 1066 memory capabilities offered by the 9400M chipset, as opposed to slower speed DDR2 800. (The older Mini used 667Mhz DDR2 parts.) Because DDR3 system memory is tightly integrated with the GPU, if you double RAM from 1GB to 2GB, you also double graphics memory, which goes from 128MB to 256MB.

The 9400M also supports Nvidia’s CUDA processor that accelerates applications, games, and can transcode videos in real-time. But the big CUDA news is that it is a general purpose parallel computing architecture that leverages the parallel compute engine in Nvidia’s GPUs to solve many complex computational problems in a fraction of the time required on a regular CPU. It includes the CUDA Instruction Set Architecture (ISA) and the parallel compute engine in the GPU.

To program the CUDA architecture, developers can currently use C, one of the most widely used high-level programming languages, which can then be run at great performance on a CUDA enabled processor. Other languages will be supported in the future according to Nvidia, including FORTRAN and C++.

Bottom line, many types of compute hungry, parallel processing applications can see as much as a 35x speed up by using CUDA.  Rumbles in the jungle also suggest that Apple may be incorporating the CUDA architecture directly into one of its upcoming MacOS X iterations.