VIA EPIA PX10000 Pico-ITX Review

Posted on June 2, 2007

Go to: Introduction Unboxing Specifications External Overview The PX-O I/O board Benchmarks Audio and Video Playback Heat, Power and Noise Final Thoughts

Benchmarks

Sisoft's Sandra is a comprehensive diagnostic, analysis and testing package. Benchmarks do not always represent a true real-life performance, but they are useful to compare the speed of various CPUs, and elements of the system. We've used this program in previous Mini-ITX reviews, so had good figures to compare against.

We chose four boards in comparison. The EPIA CN boards have 1.0GHz and 1.3GHz C7 processors. The EPIA N 10000 is the fastest currently available Nano-ITX board at 1.0GHz, with a C3-based Luke processor. The EPIA SP 13000 was the fastest of the previous C3 generation EPIA boards at 1.3GHz. We expected the Pico-ITX to perform somewhere between the Nano-ITX and 1.0GHz CN board in our benchmarks.

CPU Benchmarks

The Dhrystone benchmark is a long standing industry benchmark used to measure CPU performance using a standardised sample of mainly numerical operations. The result is given in MIPS (Million Instructions Per Second).

The PX 10000 produced approximately 1.4 MIPS per 1 MHz of CPU speed - level with the EPIA CN 10000.

The Whetstone benchmark measures FPU (Floating Point) performance, although many modern processors have a number of newer features such as out-of-order execution, pipelining and SSE2 which are not tested using this benchmark.

Again we are level with the EPIA CN 10000. The weaker FPU performance of the C3 processors reveals itself here with the SP and Nano-ITX figures.

CPU Multi-Media Benchmark

The CPU Multi-Media Benchmark uses all the Multimedia Extensions available to the CPU to draw a Mandelbrot fractal. Multimedia Extensions are additions to the x86 instruction set designed to make repeated or parallel operations run faster. Digital imaging or streaming video applications can make good use of these extensions, which use Single Instruction Multiple Data (SIMD) techniques. Changing the contrast of an image or MPEG decoding and encoding all require a large amount of data to be manipulated by the same instructions. This benchmark is therefore a reasonably good test of raw and MPEG2 recording and playback functionality. The PX has some tricks up its sleeve to deal with MPEG videos though. We'll look at that a little later.

The PX 10000 is level with the CN 10000 again, and not far behind the SP 13000.

The Floating-Point Multi-Media results paint a similar picture to the Integer results. Real world results will depend on the particular SIMD instructions used by a particular task.

Rip WAV to MP3

Next we used dBpower AMP Music Converter to convert a 70MB WAV file to an MP3, using the default settings. This test is a good raw CPU power test - a similar (but more lengthy) task would be compiling a Linux kernel. We've still got the same file from all our previous tests. The PX 10000 placed itself where we expected - between the N and CN 1.0GHz boards. The slower CPU speed and limitations of the C3/C7 architecture show themselves here - a modern Intel or AMD CPU will do the same test in under a minute.

Padlock functionality

The VIA PadLock Security Engine integrated into the VIA C7 and VIA Eden processor family supports secure computing through the inclusion of up to five security features: SHA-1 and SHA-256 encryption; AES Encryption; a Montgomery Multiplier; NX Execute Protection; Random Number Generator.

More information about VIA's Padlock

We have had great successes in the past testing the Padlock features of VIA boards, but were thwarted by software with the EPIA PX. Under Windows, VIA's AES Benchmark tool (circa 2004) and Zip Utility unsurprisingly both didn't recognise the PX CPU. The latest version of OpenSSL (0.9.83e) under Linux had the same problems. The nearest we got to Padlock was an enable/disable NX Execute Protection option in the BIOS. If VIA update their software we're sure this will all function correctly.

Audio and Video Playback -->