I don't want to comment too much about the results, because you will see that they speak for themselves. However, before we get into the application benchmark results, I'd like to discuss the scores we saw in Sandra's memory performance tests, because they are exemplary for KT266A.
Those results speak a clear language. You can see that VIA's first shot, the Apollo KT266 chipset was providing a particularly bad memory performance, inevitably leading to bad scores in real world applications. The new Apollo KT266A chipset however, is leaving the whole competition far behind, giving a pretty good idea what the results of the application benchmarks will actually look like.
Now let's skim through the other results without any delaying comments: