For those brave souls who have ventured into the arena of overclocking their SL-5600 (and those souls who are thinking about it), I've thrown together a table comparing the standard kernel performance with the overclocked (and preemptif) patched kernel.
The benchmark that I used has been optimized (using GCC compile time paremeters) for the different ARM processors. The goal was to see which GCC optimizations worked best with which processors for different target operations. I did not compile the software (and I can't seem to find where I got them), but you can run your own tests with them: nbench-arm.tar.gz or nbench-arm.zip.
The data was collected from the better of two runs on each kernel. The original data collected for each kernel can be found here: Standard kernel data and Overclocked kernel data
nBench optimized for ARM version 3 (nbench-armv3).
| TEST | Iterations/sec. (no overclock) |
Iterations/sec. (overclocked) |
Percentage Increase |
| NUMERIC SORT | 75.01 | 141.57 | 88.7 |
| STRING SORT | 5.9102 | 8.181 | 38.4 |
| BITFIELD | 1.9658e+07 | 2.3315e+07 | 18.6 |
| FP EMULATION | 5.9453 | 7.1799 | 20.8 |
| ASSIGNMENT | 0.8597 | 0.98309 | 14.4 |
| IDEA | 212.44 | 250.87 | 18.1 |
| HUFFMAN | 25.349 | 41.096 | 62.1 |
nBench optimized for ARM version 4 (nbench-armv4).
| TEST | Iterations/sec. (no overclock) |
Iterations/sec. (overclocked) |
Percentage Increase |
| NUMERIC SORT | 73.421 | 140.88 | 91.9 |
| STRING SORT | 5.9195 | 8.1609 | 37.9 |
| BITFIELD | 1.9673e+07 | 2.3343e+07 | 18.7 |
| FP EMULATION | 8.8374 | 10.857 | 22.9 |
| ASSIGNMENT | 0.92285 | 1.0567 | 14.5 |
| IDEA | 232.56 | 274.51 | 18 |
| HUFFMAN | 25.595 | 41.118 | 60.7 |
nBench optimized for ARM version 5te (nbench-armv5te).
| TEST | Iterations/sec. (no overclock) |
Iterations/sec. (overclocked) |
Percentage Increase |
| NUMERIC SORT | 73.741 | 165.32 | 124.2 |
| STRING SORT | 5.9148 | 8.1739 | 38.2 |
| BITFIELD | 2.8382e+07 | 3.3614e+07 | 18.4 |
| FP EMULATION | 9.4899 | 11.702 | 23.3 |
| ASSIGNMENT | 0.92868 | 1.0604 | 14.2 |
| IDEA | 266.46 | 314.71 | 18.1 |
| HUFFMAN | 25.53 | 41.334 | 61.9 |
nBench optimized for Xscale (nbench-cpu-xscale).
| TEST | Iterations/sec. (no overclock) |
Iterations/sec. (overclocked) |
Percentage Increase |
| NUMERIC SORT | 76.278 | 165.24 | 116.6 |
| STRING SORT | 5.9079 | 8.1544 | 38 |
| BITFIELD | 2.8382e+07 | 3.377e+07 | 19 |
| FP EMULATION | 9.6 | 11.982 | 24.8 |
| ASSIGNMENT | 0.94056 | 1.078 | 14.6 |
| IDEA | 269.96 | 318.73 | 18.1 |
| HUFFMAN | 27.686 | 43.617 | 57.5 |
nBench optimized for Xscale (nbench-tune-xscale).
| TEST | Iterations/sec. (no overclock) |
Iterations/sec. (overclocked) |
Percentage Increase |
| NUMERIC SORT | 75.29 | 165.24 | 119.5 |
| STRING SORT | 5.9172 | 8.1546 | 37.8 |
| BITFIELD | 2.8404e+07 | 3.3604e+07 | 18.3 |
| FP EMULATION | 9.4205 | 11.706 | 24.3 |
| ASSIGNMENT | 0.92765 | 1.0605 | 14.3 |
| IDEA | 266.36 | 314.71 | 18.2 |
| HUFFMAN | 25.707 | 41.345 | 60.8 |