In order to obtain information on the repeatability of hydroacoustic estimates of abundance and biomass of the fish stocks in Stechlinsee (Germany) and Irrsee (Austria), we performed synchronised surveys in three expert teams using either two identical 120 kHz SIMRAD EY500 echosounders (Stechlinsee) or one such machine and a 200 kHz BioSonics DT6000 (Irrsee). At Stechlinsee, where the fish stock is dominated by Coregonus albula, the night surveys with vertical beaming were made along 12 transects with two boats, one following the other at a distance of approximately 300 m. Fish-density estimates from the two units were highly correlated, but the slope of the regression differed from the expected value of 1. The regression relating biomass estimates of the two expert teams also revealed a close correlation with the slope not being significantly different from unity. This pattern could be explained by the fact that the differences in abundance were due to small targets differentially encountered by the two teams and contributing only marginally to total fish biomass. Nevertheless, we found that the results of a single hydroacoustic survey can be reproduced reliably by an independent team using similar equipment. The comparison between different machines at Irrsee, where the fish stock is dominated by Coregonus lavaretus, was done at night with the two echosounders mounted on the same boat and running simultaneously without interference. The abundance estimates correlated significantly but more weakly than in the Stechlinsee exercise, although the echosounding systems differed in many technical specifications. Correlation of biomass estimates was found to be robust, with the slope of the regression not significantly different from unity. Hence, we found that simultaneous surveys by two expert teams using split-beam systems from completely independent manufacturers, differing in sound frequency, beam width and shape, pulse length, ping rate, acquisition software, and post-processing software, produce directly comparable biomass estimates.