I haven't really experimented much with it myself, but reports from others seemed to indicate that the results were more aesthetically pleasing from merging of multiple exposures vs. using a single EXR. I don't know why this is the case, perhaps there is just so much information in the full EXR that the top and bottom of the scale are putting off the balance of the mid-range and the compression isn't as effective. Or perhaps there is some bias due to extremely bright pixels (e.g. sun) skewing the tone mapping algorithms. I'm just speculating. I'd really hope that the EXR would actually give the superior results as it has the most information in it. I think it's worth someone doing some more formal tests, particularly in comparison to other apps (like the recently released Photoshop CS5 with enhanced HDR and tone mapping functionality).
- Oshyan