Bob Friesenhahn wrote: > On Mon, 16 Jun 2008, James Cornell wrote: > >> On Solaris GCC is used to compile Opera. We do not know what options >> were >> used to build Opera since it's proprietary. Not that I'm against >> closed-source software, this is hardly the case, but I'd suspect >> Intel would >> have better tuning since it constitutes a large installation base, and >> through deduction more engineers on all fronts working toward Intel >> support. I'll compare AMD and Intel regardless with all the tests I >> can run > > I see that you keep copying my private emails to the list. > > It is wrong to make such sweeping judgements regarding "Intel" or > "AMD" CPUs since they all perform differently and offer different > features. > > The tests you are performing are only valid for your systems. > > Bob > ====================================== > Bob Friesenhahn > bfriesen at simple.dallas.tx.us, > http://www.simplesystems.org/users/bfriesen/ > GraphicsMagick Maintainer, http://www.GraphicsMagick.org/ > Heh yeah sorry about that. The judgement they are only good for my systems is short sighed. I do not necessarily recommend that people compare statistics between the two unless their job is heavily dependent by proxy on what is being benchmarked. For those who happen to have either, simply reading only the data that is for their particular setup is the best thing to do. These machines are quite popular, so some people will find it to be useful information, mostly in terms of what browser is the fastest, a way to stop the arrogant marketers from claiming they are the fastest.
Now, these tests between browsers is useful, it's tedious, but easy as the tools are there. Currently I'm only doing the tests on one machine, the Intel, which will cover a wider subset of people. Since I'm not a Mozilla contributer I have to wait so I can rerun over the old RC3 bench results. I'm doing about 60 configurations, later plotting them if time permits. James
