It's definitely one "block". When I was making the labelled version of the photo it was very clear where one block ends and the other begins (you can see the greyish boundaries when you zoom in on the full size die photo). There are some things that look like seams, but are actually artefacts of the fact that the die photo is composed of a large number of small photos stitched together, and this is what's in the middle of J1 (just scroll up and down and you'll see it the whole way up the photo).
Now, that's not to say it doesn't hold two logically separate components, but it's one solid block of logic.
Thanks for the feedback. Yeah, the ratio of the proposed Interpolators to SPUs is a good point to raise. I simply don't know enough to say how these things usually scale. I'd say we shouldn't assume it's a linear relationship with this little to go on, though. Even if they don't usually scale the way we see here, who's to say Nintendo didn't have their own reasons for including all of them? It's something worth looking into, definitely.
In regards to the size of J1, that's another curiosity. I can only guess, but perhaps it has something to do with how the data is passed from the dispatch unit to the interpolators. If we assume that's what they are, in the old RV770 design, all the interpolators border the dispatch unit. Here, only one does. The intricacies of die layout may have prevented it. So perhaps that one has some extra logic on board to make up for this. I'm in over my head already with this stuff, though. So I'll stop there.
I don't think the S blocks have too much logic to be texture caches. Take a look at the texture caches on Llano and RV770. They need logic to keep doing what they do automatically.
You might be right on S. I think the equivalent blocks on RV770 also contain the logic and wiring for the Data Request Bus and Crossbar, but I suppose the same thing could be true here.
Your hypothesis for the size of J1 also seems reasonable.
