The post spends a lot of time talking about flash not utilizing / partially utilizing / fully utilizing graphics cards and makes a lot of conclusions and even takes a dig at apple about it. But then never bothers to ask the same questions about html5. e.g., which browsers are natively using hardware acceleration for H.264, if any? Apparently the author didn't find it relevant. "I didn't contact any other companies because the tests are objective and straightforward."

If flash only beats html5 when using hardware acceleration and html5 browsers haven't introduced hardware acceleration yet, his conclusions end up backwards, or at best completely irrelevant and speculative. shrug