I still maintain, as I wrote earlier this year, that these benchmarks are artificial, and do not really represent real-world performance. That is not to say that Carakan wasn't really needed. In my opinion, it was important for at least two reasons:
- Marketing: Let's face it – marketing matters. Today, SunSpider and Peacekeeper are probably little more than a way to create pretty graphs that people can understand. These benchmarks may be artificial, but most people are not treating them that way.
I'm really impressed and excited by what the Carakan team has been able to do so far. Their work is very important, and the full impact of the new engine will become apparent when a new generation of Web applications enters the stage.
In the meantime, we can at least benefit from pretty graphs showing Opera as #1 🙂
Oh, and don't forget about Vega. Keep in mind that hardware acceleration isn't even enabled yet!