Rasvar - I definitely agree. I remember the days when the GFDL was the greatest thing around; then, a couple of years back, they tinkered with it and it's never been the same. It's still the best purely tropical model, however, and never too early to start preparing. But, that said, track errors (yet alone intensity errors) are so large at 5 days that the model is almost not worth looking at.

A note on TD #3 - over the past few hours, there has been a flare up in convection near where the LLC should be, even accounting for a slight reformation a bit to the north. After the convection died away earlier today, a band began to develop on the west side (just before 8p ET), worked it's way to the south (~9p), and now has consolidated around the center. If it only slowed down a bit, we'd have something; even still, it's slowly getting it's act together. The outflow appears to be very good in almost all quadrants and waters are certainly warm enough.

As a couple of quick, related asides - it's interesting how every person you ask has a different "feel" on a storm. I've heard a wide range on TD #3, both here and in person with people and around here, ranging from Allen to Gilbert to Lily and a few others. Heard the same with Bonnie, too. It's likely these two storms will be the first Bonnie/Charley and not the next anything, but it's interesting where some of the comparisons are drawn from.

And, I'm not sure if anyone else caught it earlier, but seeing Steve Lyons try to explain the four-quadrant jet model (from quasigeostrophic theory, favored regions for synoptic-scale ascent and descent) in layman's terms was rather interesting. I'm not sure that it's terribly relevant with TD #3, however - the "jet" he showed only had a magnitude of 20mph at it's strongest and it's more dry air ahead of the storm (albeit retrograding) that will impact the storm as opposed to rising/sinking motion.

Definitely a couple of things to watch for now, though.