I'm still waiting for an in depth exploration of an FX 8 core with one core from each module disabled, how well it would perform and overclock compared to regular FX 8 core CPUs and the rest of the CPUs worth buying. It has been shown that disabling 1 core from each module improves single-threaded performance significantly (about 10-15% as shown in a previous test). It should also use less power with only half the cores so should overclock further, maybe making the FX-8120 viable in gaming. Would this be enough for it to become considerable? I'd like to know although I don't think it would beat Intel's stuff but it might meet or beat AMD's Phenom IIs that way.
I don't remember the site where this was checked but it was only glanced at, not looked into seriously. They just disabled a core from each module of an 8120 or 8150 and compared it to itself without disabling any cores, both at stock clocks.
Besides that, good article. Really tells us about how far behind AMD has gotten. We all saw what happened when Intel tried to get huge clock rates to make up for huge pipelines and other IPC killers with Netburst so we all should know how that failed. Honestly, AMD could have pulled it off if what Cliff A. Maier said about the automated design used in FX chips is true, 20% more power usage for 20% less performance and 20% larger die size. If not for this then AMD would have an okay product but it still wouldn't be a good replacement for the Phenom II line.
http://www.xbitlabs.com/news/cpu/display/20111013232215_Ex_AMD_Engineer_Explains_Bulldozer_Fiasco.html
There's plenty of sites that covered this problem so look around if you want to, I saw it first at news.softpedia.com.
It would also have been nice to know when each of those CPUs become CPU limited in each game. Basically see what video cards are the best each one can make good use of in gaming overall. That would be more relevant then using a 7970 with each CPU since it probably has about the same or a higher price than what people looking at these CPUs want to pay. It makes a great show of telling us what each CPU can handle and is definitely nice to see but it doesn't give us a good picture of what these CPUs can do with a budget-oriented card. Obviously, any CPU that didn't do good enough with the 7970 won't do good enough with anything else but it's still seems like using a synthetic benchmark to show performance instead of a real life benchmark.
Also, I know games don't use many threads very well but I'm curious, does using multiple GPUs instead of a single, fast GPU make a difference on CPUs with multiple, slower cores? For example, the GTX 295 and Radeon 4870X2 have similar performance to a single GTX 480, GTX 570, and Radeon 6970 (according to Tom's). Would the dual GPU card's outperform the single GPU cards with similar performance on a CPU that uses many threads to make up for low single threaded performance like the Phenom II x6s or the FX CPUs? I assume it wouldn't do much better, if at all, but it seems to be a valid question, especially if you have two GTX 295s or Radeon 4870X2s against two Radeon 6970s or GTX 480/570s. Ignoring micro-stuttering and similar multi-GPU problems of course, that would skew the answer(s) of my question too much anyway.