Solar models use a frequency-averaged opacity (the Rosseland mean opacity) to calculate radiation diffusion. However, while the use of a diffusion model of radiation transport is very accurate, the actual opacity values are the least well-known parameters in solar and stellar modelling. This has lead to at least one paper where the authors literally pleaded for more accurate opacity values. Rosseland mean opacity calculations require detailed evaluation of the atomic physics contributing to opacity, but at the solar densities and temperatures, there have been almost no measurements that could be made to 'benchmark' these calculations in order to evaluate their accuracy. I will describe a programme of work where the plasma densities and temperatures present in the Sun are re-created in the laboratory in plasmas produced by high power laser irradiation. We have developed a probing technique using extreme ultra-violet lasers to measure the transmission through and hence opacity of such plasmas.