I don't think this is entirely accurate. From what I can recall, the case against Landis was from two tests. The first showed that his T/E ratio was well outside the normal ratio for humans. Note that this is different from having elevated testosterone, in fact I believe his testosterone was actually lower than normal. It was just that his epiesterone was WAY lower than normal (these are normally about equal, his ratio was 12/1). As far as I remember this test was somewhat discredited due to shoddy procedures at the lab. The one that stuck was an isotope test which showed that the testosterone in his sample had a different isotope ratio than is found in humans. From this they concluded that it was synthetic and thus upheld the ban. I don't claim to entirely agree with all of their methods or even the results of the test, but I really don't think it's fair to say that he crashed, had a surge of adrenaline and subsequently tested positive.

This is slashdot, so I suppose it should not come as a shock that the summary makes claims that don't stand up to even a casual examination. About 15 seconds on google scholar produces the following paper:

    Correa, A.A. and Bonev, S.A. and Galli, G, Carbon under extreme conditions: Phase boundaries and electronic properties from first-principles theory. Proceedings of the National Academy of Sciences of the United States of America.103, 1204 (2006)
link to article

The second paragraph of the article in Nature Physics (subscription required) that this story is about mentions at least 11 other papers on theoretical calculations and experiments on the melting of diamond. So no, this is not in fact the first time that the melting of diamond has been studied. Indeed, the linked article itself refers to previous experiments at Sandia National Laboratory that melted diamond, but were unable to accurately determine the temperature and pressure.

This is truly impressive work by some very skilled scientists, but let's discuss it for what it is and not what it isn't.

While this sort of machine is useful (I just built one for quantum Monte Carlo calculations 6 months ago) it is hardly news. NVIDIA has been pushing this sort of machine since the launch of the Tesla. In fact, they have had a parts list on their website for some time telling exactly what is needed to put together a computer with 4 C1060's. This is not even the first commercial offering of this nature, with companies like appro and microway having similar products for at least a year (see nvidia) for a more complete list.