following the same processes. For instance, the range of data captured was different with some teams electing not to capture addresses of voters, and with other teams having more than two operators working on one machine. This was corrected on the afternoon of the first day Comparison of the Two Vendors.  The hardware and software of both bidders met the requirements outlined in the tender advert, with none of the equipment sets breaking down during the site validation exercise, with both sets working after the shock and waterproof tests. However:  Laxton (Chinese Firm) had a better operator interface. It was easier to use and understand. The operators were guided on what steps they would be working on during the process. However, the least reliable component on the Laxton BVR kits appeared to be the Machine Readable Zone Scanner (MRZ), which often struggled to read data on the plastic identity cards, but could quickly read the details on the passport's data page.  Dermalog (German Firm) appeared to have better equipment and technology. Their MRZ reader did not struggle to read the details of ID cards. Their camera and light source was more convenient to use; both devices could be moved beyond the kit to scan documents. Their fingerprint reader provided much clearer fingerprints. However, there appeared to be a problem regarding the issues of charging a back battery for one of their kits. Since Observers could not interact with the Assessment teams or the bidders, it was unclear whether it was an issue of equipment failure or human error on account of an operator not correctly connecting cables. Process Progress and Conclusions  ZEC awarded the tender to Laxton Group on 30 May 2017, and the equipment and software procured were administered during the BVR process. 7.3 AFIS Tender Processes ZEC conducted a ‘limited special tender’ to upgrade the existing IT infrastructure for the Biometric Voter Registration (BVR) system, through inviting companies to submit bids to supply Automated Fingerprint Identification System (AFIS), a principal software that checks for multiple registrations that may occur during the registration exercise. ZEC informed stakeholders that seven companies expressed interest, but only 3 submitted bids within the required time, with only 2 of the 3firms (Laxton and IPSIDY) meeting the set bid requirements. The 2 firms were invited to demonstrate how their proposed solutions for the de-duplication process would work over a one and half hour period through PowerPoint presentations and live demonstrations of the software. Below are brief remarks about critical elements of the process. The presentations were attended by a few civil society organisations, some representatives of various government departments, and ZEC personnel, the bulk of whom served as technical evaluators of the presentations. The low turnout could have been due to the rushed nature of the process and late dispatch of invitations, which ZEC only dispatched a day before the presentations. 33