His only primary argument was this:
However, once again, he's provided no proof at all that Speedmark tests applications in unison. Thus, it's still not an accurate test of real-world performance, because in reality, as you mentioned, who is seriously going to run a single application at one time, be it for work, school, etc. I could *maybe* see someone doing that on a system such as a netbook, but even then that's more due to the limitations of hardware (in terms of seeing multitasking performance decrease substantially) vs. any desire to truly run an application one at a time.
I don't know about anyone else, but when I'm using Final Cut Pro or Photoshop, I don't have *solely* those applications open.
What's really amusing though is that his original argument was that it was an unfair comparison because the Lynnfield systems had 4 GB of system RAM to Bloomfield's 8 GB (and that's what we've been trying to argue against). Several of us chimed in pointing out that memory capacity did not matter for the tests Speedmark is running, since it isn't testing applications in unison and thus whether it's 4 GB or 8 GB of system RAM, the available capacity would essentially have little-to-no impact. Maybe if we were talking about 512 MB vs. 8 GB, but yeah, lol. He's yet to actually provide substantive proof that Speedmark does test in unison, instead changing the focus of his argument to "what are real world applications?" basically. Now all he's doing is making insulting comments to others and throwing a tantrum, which is why I'm not bothering to respond to him.