I notice that people are doing some initial benchmarking of BBB and other boards on the RPF forum. Results roughly as expected I guess:
Using just a simple
time echo "scale=2000;4*a(1)" | bc -l
as a lightweight benchmark, I see these numbers reported (smaller Time is better):
[table now updated with extra datapoints reported in current thread below]
| Submitter | Time (s) | Board | SoC | Clock (MHz) | O/S |
|---|---|---|---|---|---|
| shuckle | 26.488 | Raspberry Pi B | BCM2835 | 700 | Raspbian 3.1.9 |
| morgaine | 25.719 | Raspberry Pi B | BCM2835 | 700 | Raspbian 3.1.9+ #272 |
| shuckle | 25.009 | Raspberry Pi B | BCM2835 | 700 | Raspbian 3.2.27 |
| trn | 24.280 | Raspberry Pi B | BCM2835 | 700 | Raspbian ? |
| morgaine | 22.456 | Raspberry Pi B | BCM2835 | 800 | Raspbian 3.1.9+ #272 |
| morgaine | 21.256 | Raspberry Pi B | BCM2835 | 800 | Raspbian 3.6.11+ #545, new firmware only |
| selsinork | 21.0 | Minnowboard | Atom E640T | 1000 | Angstrom minnow-2013.07.10.img |
| shuckle | 17.0 | Raspberry Pi B | BCM2835 | 1000 | Raspbian ? |
| morgaine | 16.153 | BB (white) | AM3359 | 720 | Angstrom v2012.01-core 3.2.5+, user-gov |
| selsinork | 15.850 | A20-OLinuXino-MICRO | A20 | 912 | Debian 7.0, 3.4.67+ |
| selsinork | 15.328 | Cubieboard | A20 | 912 | Ubuntu/Debian 7.1 |
| pluggy | 14.510 | BBB | AM3359 | 1000 | Debian |
| morgaine | 14.153 | BBB | AM3359 | 1000 | Debian 7.0, 3.8.13-bone20, perf-gov |
| selsinork | 13.927 | A10-OLinuXino-LIME | A10 | 1000 | Debian 7.0, 3.4.67+ |
| Heydt | 13.159 | Cubieboard | A10 | 1000 | ? |
| selsinork | 12.8 | Sabre-lite | i.MX6 | 1000 | Debian armhf |
| selsinork | 12.752 | Cubieboard | A20 | 912 | Ubuntu/Debian 7.1 + Angstrom bc |
| selsinork | 12.090 | BBB | AM3359 | 1000 | Angstrom dmnd-gov |
| pluggy | 11.923 | BBB | AM3359 | 1000 | Angstrom |
| selsinork | 11.86 | BBB | AM3359 | 1000 | Angstrom perf-gov |
| selsinork | 9.7 | Sabre-lite | i.MX6 | 1000 | Debian armhf + Angstrom bc |
| selsinork | 9.606 | Sabre-lite | i.MX6 | 1000 | LFS 3.12, gcc-4.8.2, glibc-2.18 |
As usual, take benchmarks with a truckload of salt, and evaluate with a suitable mixture of suspicion, snoring, and mirth. Use the numbers wisely, and don't draw inappropriate conclusions. 
Morgaine.
