1: \begin{abstract}

2: We compare the photometric accuracy of spectra and images

3: in flares observed with the {\sl Ramaty 

4: High Energy Solar Spectroscopic Imager (RHESSI)} spacecraft. 

5: We test the accuracy of the photometry by comparing the 

6: photon fluxes obtained in different energy ranges from the

7: spectral-fitting software SPEX with those fluxes contained 

8: in the images reconstructed with the {\sl Clean, MEM,

9: MEM-Vis, Pixon,} and {\sl Forward-fit} algorithms.

10: We quantify also the background fluxes, the fidelity 

11: of source geometries, and spatial spectra reconstructed

12: with the five image reconstruction algorithms. We investigate

13: the effects of grid selection, pixel size, field-of-view,

14: and time intervals on the quality of image reconstruction.

15: The detailed parameters and statistics are provided in an

16: accompanying CD-ROM and web page.

17: We find that {\sl Forward-fit}, {\sl Pixon}, and {\sl Clean}

18: have a robust convergence behavior and a photometric

19: accuracy in the order of a few percents, while {\sl MEM}

20: does not converge optimally for large degrees

21: of freedom (for large field-of-views and/or small pixel sizes), 

22: and {\sl MEM-Vis} suffers in the case of time-variable

23: sources. This comparative study 

24: documents the current status of the {\sl RHESSI} spectral and 

25: imaging software, one year after launch. 

26: \end{abstract}

27: