These lasers blur the line between 'laser' and 'really bright light source'. On the one hand, they are one of the most efficient light emitters (with nearly 50% electrical to optical efficiency, they were king of efficiency until the development of modern LED technology), and are capable of outputting huge amounts of light-in excess of 50W (yes, watts)-without requiring any particularly exotic external equipment. In fact, I have seen people power them with just a rechargeable battery and carefully chosen length of wire to limit the current! On the other hand, their output is not monochromatic by any laser standards (with line widths on the order of several nanometers or thousands of GHz wide). Furthermore, it is not possible to get anything resembling a collimated beam out of one, partially because there are many independent emitters sitting next to each other (so at best you could get a grouping of beams), but even if you singled out one of the emitters the active region is about 200um by 1um wide and very highly multimode, so at best you could get a beam which is collimated in one axis but diverges quite quickly in the other axis. That said, these bars were designed for 1 specific purpose--to act as an optical pump source for a Nd:YAG laser such as can be found elsewhere on this site--and they do this extremely well.
As fate would have it, I came across a large lot of raw unmounted laser diode bars on eBay, which I purchased and mounted on copper carriers, resulting in the mountain of bars at the top of the page. I actually never ended up using them for my planned laser, because the coherent avia turned up and presented a significantly easier route to a high power Nd:YAG laser. In any case, I decided to do some stress testing on them to see what they were really capable of, so I mounted one on the water block I had designed for my DPSSFD Nd:YAG laser (which also got scrapped when the mystery 532nm medical laser came) and brought out the 100A power supply:
The little guy took a surprising amount of abuse, at 84A it pegged my power meter at 70w, although I suspect its lifetime was significantly shortened after that run.
This run was done at 25ºC, and the slope efficiency was right about 1W/A, dropping down to about 0.9W/A at higher currents.
There is a bit of information on Sam's laser FAQ. If you still have questions, I can be contacted at email@example.com