Why I remain unconvinced:
There are two main issues I have here with the proof that there is a problem. So, I'm going to take a stab at explaining why I am not convinced by the evidence presented. Note, I'm not saying the problem does not exist. It very well might and the evidence might actually be accurate. I'm simply stating why it's not convincing to me.
1. DPR analysis. This analysis is very useful and it is a good benchmark, but, there are some very glaring omissions in the analysis. For one, it does not measure opportunity cost. It only measures damage potential. For example, in order to Bless, the cleric has to cast the spell. Say the cleric rolls low in initiative order, after the fighter(s) attack. That means for the first round of combat, the cleric has contributed nothing to the combat. Compare that to, say, a cleric that leads off with Flame Strike, even at the end of the round. Which has a bigger effect on the combat? Also, let's not forget, there is the opportunity cost of Concentration. Are there any other spells with concentration that the cleric could be casting instead? For example, a Light Domain cleric might be casting Flaming Sphere instead of Blessing 2 PC's. Which is more effective? I don't know. I'm picking rather random examples, but, the point is, DPR does not take this into account.
There is also the opportunity cost of taking the GWF/SS feats. A few posts up, we were comparing a human (variant) Lvl6 fighter to a hill dwarf Lvl 6 fighter. Thing is, our hill dwarf, with the Tough feat and an ASI has considerably more HP than the human. How much DPR is that worth when the hill dwarf can eat a dragon breath without saving while the GWF/SS fighter drops like a stone?
My point is this: DPR is a metric that can be used for comparison and it's an important one. Certainly the DPR analysis that has been done points in the direction that there might be an issue. It's not conclusive, but, it does highlight the issue and I feel that it indicates that we should be taking a closer look.
2. Anecdotal evidence. When you think about it, play testing is pretty much purely anecdotal. But, there is a significant difference here. As it stands, with the anecdotes we have, I know that if I play the say [MENTION=5834]Celtavian[/MENTION], [MENTION=6786202]DaveDash[/MENTION] or [MENTION=2011]KarinsDad[/MENTION] play and my players did the same, I'd likely have similar issues. But, are these issues the result of the mechanics or a result of how they play the game? We don't know. There just isn't enough evidence here. To properly explore this, you need about 10 groups playing the same half a dozen scenarios multiple times and recording the results each time. All I really know at this point in time is that some people are apparently having issues with these feats, typically at quite high level. What I don't know is why they are having these issues. And without doing a LOT more testing, the anecdotes don't really prove anything.
So, after this thread am I convinced there is a problem? No, not in the least. I am convinced that for some people there appears to be a problem. But is it systemic or an issue with the idiosyncrasies of their specific tables? Again, I don't know. I do know that my group is not having these issues, so, until we do have these issues or I see a LOT more hard evidence presented, I'm not going to worry about it too much.