The trial of capillary electrophoresis (CE) instruments from 1993 has been updated. The test procedure which was used there could be applied with few modifications. The data of 10 instruments are presented in a detailed table. Additional instruments are included this time. Aspects of method and instrument validation and robustness are discussed and listed in tables. Relevant instrumental features are deduced. Two-dimensional detection and a protocol of the current are very helpful for method development and validation. Some very useful methods that improve the limit of detection or resolution require pressure-driven counter-current and the possibility to control polarity and voltage during runs. Intermediate precisions corresponding to R.S.D. values of below 1% have become state of the art during the last few years. An overall day-to-day precision of 0.1% R.S.D. seems already possible if the multiple injection mode can be used. The detector technology has been improved, therefore injection became the main error source. This error source can be decreased by using internal standards or relative peak areas. In the future CE will be superior to other separation techniques not only in terms of performance, but also in terms of precision.