There are 10 sessions for the WCRP Workshop on CMIP5 Climate Model Analyses. After Session 1 (Historical simulations and projections), I feel we (meteorologists and climate modelers) are still more or less like BLIND MEN who try to figure out what an elephant looks like.
Each poster presenter has one slide and three minutes to advertize her/his poster. They all managed to finish within the time limit. Some concluded that CMIP5 is better than CMIP3, and some said there is "[n]o progress." It really depends on HOW one looks at these models and which aspects one focuses on... A good study, in my view, is to link model bias (or improvement) with physics or dynamics.
The "line up" was informative enough for me. I left without looking at any posters this morning because they are all very far from what I am working on. The same for the afternoon session.