The use of photogrammetry in the field of underwater archaeology has developed considerably in the last twenty years to the point of becoming an indispensable tool today. This technique is widely recognised as a formidable improvement for our discipline, and presents many advantages, among which the possibility of recording quickly a large quantity of very precise data. During the last fifteen years, the Centre Camille Jullian team has had the opportunity to work together with Croatian colleagues on nine ancient submerged shipwrecks in the north-eastern Adriatic region. From the very first campaigns, the use of photogrammetry seemed to be an obvious solution which was then quickly systematized. It is during this period, marked by a strong evolution of the documentation practices, that we noticed a significant reduction of the time necessary and thus devoted to the excavation of a wreck. If this represents an appreciable advantage in the context of preventive archaeology, it seems to us that it is also accompanied by more negative aspects. Indeed, given that the time for recording the remains of a ship has been considerably reduced, consequently our time of direct observation and familiarization with the wreck has also been greatly reduced. We can therefore question the possible consequences on our knowledge of the remains and on our final interpretations. A project has recently shed new light on this issue. In 2019, thanks to a grant from White Levy Program, we have undertaken a project in order to complete the study and publish the iconic excavation of the Madrague de Giens wreck (1st BC), almost fifty years after the first dives were done on the site. One of the objectives of the project was to digitize all the underwater pictures taken during the excavation (more than 11 000 in total) in order to try to generate a 3D photogrammetric model of the site. The new 3D model of the Madrague de Giens allows us to consider the important contribution of this tool in term of precision and quantity of data, while underlining the qualitative importance and the absolute necessity of the direct observations made by the archaeologist on the site.