Slow-motion videography and recording experimental archaeology (part II)

Michał Gilewski

Experiment, documentation and processing

Witold Migal, an experienced flint knapper and archaeologist from the State Archaeological Museum in Warsaw, agreed to perform flint knapping experimentation that would be recorded. He also advised the author on issues of flint knapping.

I filmed the knapping of a biface, using a flint stone core, a smaller stone for preparatory grinding and soft antler hammers for percussion. Unfortunately, lighting was limited to natural sources. During the experiment, various steps of the knapping process were shot at differing magnification levels. The process of filming took two hours during which 83 slow motion videos were produced. The slow motion shots were shot at two speeds: 400 frames and 1200 frames per second. The camera saves them as .mov files encoded using the popular MPEG-4 AVC codec set to replay at the speed of 30 frames per second. This results in the time being perceived as 13.33 and 40 times slower, respectively. Due to technical limitations, the camera could produce only five seconds of slow motion video at a time, so we were unable to produce a continuous video documentation of the process. The 83 slow motion videos of both types already have a total slowed down playback time of more than 92 minutes. Four videos were shot at regular speed, including a “faster standard” of 60 frames per second. After the whole process, a short direct testimony recording of Witold Migal describing the process was produced. An additional objective was to determine how usable the regular speed videos are in comparison to the slow motion videos, and if the software-generated slow motion-like effect (Twixtor plug-in) is of any use in such an application (see details below). If so, the vast amount of archival video footage could be slowed down and provide a new quality for adequate analysis. All imagery produced for this paper was created using a the Nikon 1 V1 camera with a standard 10-30 mm 3.5-5.6 lens, which can be bought for around 250 Euros. All videos were produced in progressive mode, meaning each frame was shot in full resolution, and no videos were made using interlacing. The regular speed videos were shot at 1280×720 resolution, while the slow motion videos were shoot at 640×240 (400 fps) and 320×120 (1200 fps).

Some popular software was used to process the video. For the regular-speed digital video, a visual effect that uses existing frames to interpolate “intermediate” frames was used by taking advantage of “plugins” and software editing programs. Here, a trial version of a plugin called Twixtor [1] and the Adobe After Effects software (a part of the popular Creative Cloud software suite) were used to process the video. I selected one video that featured a rapid blow to the core. The video was intentionally shot at a regular but more frequent rate of 60 fps in the progressive mode. Adobe After Effects was also used to produce a “match moving”-based visualization.

The results: regular video processing, motion recording and visualization

After processing, a regular speed video converted to slow motion video was analysed. While some slower motions (preparatory grinding for the strike) are presented adequately in the playback and can be efficiently followed by the human eye, the same cannot be said about the main strike. The preparatory motions were slow enough to not have significant changes occur between the registered frames, allowing the software to produce adequate artificial frames. These can be observed during slower speed playback. However, the main percussion motion itself is too quick to produce such imagery which would provide the software a basis to calculate non-existing frames that would properly display the motion. Instead, what results is the same blurred image being displayed for a much longer period [2].

Such movements can be efficiently portrayed only in proper slow motion recording. In the 400 fps video, it can be observed that some percussion movements last approximately 30 frames or one second. At the regular video speed, image sampling is too slow to capture more than 2-4 frames in the movement.

Because the slow motion video records around 30 frames during the motion of a single strike [3], more aspects become clear. This also allows for comparisons between strikes, for example between good strikes and those that were poorly executed. Selected key frames can also be used to represent the motion (see Figure 1).

Figure 1. A percussion movement presented on selected key-frames.

Also, a simple illustration and interpretation of the movement can be executed using a simple match moving feature (“motion track effect”) in Adobe After Effects. In the selected video, this was used to track the bone tool used to strike the biface. The program produced a path that was converted into a vector drawing (see Figure 2 ) [4]. Such a path, although it remains a two-dimensional projection, provides a good, approximate overview of the motion and thus enhances our ability to understand it [5] . Such snapshots of different movements can also be used for comparison. This interpretation represents the whole motion so they can be efficiently integrated with movie files in computer databases.

Figure 2. Match-moving-based visualisations of the percussive motion.

It must be noted that the files are very easy to store and share. What is of even greater importance is that the recordings can be integrated into databases, bringing some interesting formalized frameworks for researching ancient craft-production related gestures.

The video materials were reviewed by Witold Migal and myself. Of great importance was Migal’s critique of such techniques from experience. An experienced flintknapper, his initial skepticism changed to emphatic approval after seeing multiple videos documenting the difference in and noting how such few videos could scrupulously document details of (i.e. ineffective moves). Of some concern to him was that the static character of flintknapping archaeological material makes advanced interpretations of what techniques and gestures were used limited. However, he also noted that the modern experimental re-creation can be viewed from a different perspective, for example for permitting the recording of a level of detail not visible to the naked eye in the individual variability of practice of modern flintknapping. In this sense, the results can be described as enabling new, more reflective and insightful perception of archaeological experiments and may lead to interesting new developments of the discipline (see Gilewski, 2015, p. 140) [6].

Any discussion of technique would be incomplete without mentioning some technological alternatives to recording data. Motion capture is currently only being utilized for recording human movement and related cultural heritage (see Dunn et al., 2012; Dunn and Woolford, 2012; Stavrakis et al., 2012). The technique is, however, based on specialized equipment – it requires the use of multiple high-speed cameras. Using this technique, much sophisticated three-dimensional information about motion (for example positions of objects per frame) can be recorded. However, the use of motion capture means that significant effort, preparation and budget must be secured to create such documentation. In the case of slow-motion recording, the minimum equipment only includes a camera (and tripod), which making it a cheap and easy alternative to motion capture.

The emergence of consumer digital cameras able to record in slow motion is of great benefit to the field of archaeology. The technique, which has a long history in archaeological film making, now has a much greater breadth for potential implementation. This not only means that archaeological experiments, which are often conducted on a low-cost basis, can also use this kind of documentation, but that a large number of these recordings can be made.
This can be crucial for examples such as flint-knapping, where slow motion films were limited to educational and popular science purposes due to costs, while regular videography (that is less applicable in this application) has very frequently been used for analysis, self-evaluation purposes by both scientists and hobbyists. Now, slow motion can be more freely applied to these and other and similar purposes while benefiting from the digital nature of the media. Not only can the videos be easily shared (see Whittaker, 2012), but research footage can be produced much more often, and this greater amount of data can be handled by means of integration into databases for computer based processing and analysis.

Finally, it must be added that such videos facilitate very “reflective” analyses. The ability to self-reflecta virtue of great value to archaeologists. Archaeologists who are trained to imagine a past which cannot be seen are now permitted to observe events that occur in the blink of an eye. This also produces a very aesthetically pleasing by-product that may be used not only in science, but also in art. As I already mentioned, slow-motion flint-knapping was used as a portion of an artistic experimental film. Perhaps, wider possibilities can contribute to other film projects mixing archaeology and art. The application of this technique brings archaeology closer to other disciplines that use videography to observe various phenomena.


[1] Because a free trial version was used, the images are marked with thin red lines indicating the full version was not used.  For the intended quality test purpose, such diminished aesthetics are of minor concern.

[2] The video can be accessed here:

[3] See example:

[4] See also video here:

[5] As Tim Ingold (2007, p. 72-75) observed such ideas of tracing the line, and fragmenting it to “freeze-frame points” lead to analyses of the gesture and the ability to capture it as tangible “finished object”.

[6] See Wickstead and Barber (2012) for example of proposition of how greatly visualizing methods of archaeology and its “notions of vision” were shaped by new technical developments.

Figure Acknowledgements
Special thanks to Witold Migal for organising and performing the flint knapping experiment and to Prof. Mariusz Ziółkowski and the University of Warsaw for financial support. I also thank Szymon Ozimek for language help and Michał Przeździecki, Witold Grużdź and Kasper Hanus for their comments.


Barbash, I., Taylor, L. 1997. Cross-cultural Filmmaking: A Handbook for Making Documentary and Ethnographic Films and Videos, Berkeley.

Beale, T.W., Healy, P.F. 1975. Archaeological Films: The Past as Present, American Anthropologist 77 (12), pp. 889-897.

Blain, L. 2014. FPS1000: The low cost, high speed slow-mo camera,, (

Dunn, S., Woolford, K. 2012. Reconfiguring experimental archaeology using 3D reconstruction, [in:] S. Dunn, J. Bowen, K. Ng (eds.), EVA London 2012: Electronic Visualisation & the Arts. Proceedings of a conference held in London 10-12 July,London, pp. 172-178.

Dunn, S., Woolford, K., Norman, S.-J., White, M., Barker, L. 2012. Motion in place: A case study of archaeological reconstruction using motion capture, [in:] M. Zhou, I. Romanowska, Z. Wu, P. Xu, P. Verhagen (eds.), Revive the Past. Computer Applications and Quantitative Methods in Archaeology (CAA). Proceedings of the 39th International Conference, Beijing, April 12-16,Amsterdam, pp. 98-106.

Van Dyke, R.M. 2006. Seeing the past: Visual Media in Archaeology,American Anthropologist 108 (6), pp. 370-375.

Heider, K.G. 2006. Ethnographic Film: Revised Edition, Austin.

Gilewski, M. 2015. New imaging techniques and Polish Archaeology. Impressions on RTI and Slow-motion techniques in Archaeology, Contributions in New World Archaeology, vol. 9, pp.133-143.

Johnson, M. 2010. Archaeological Theory: An Introduction, 2 edition, Chichester.

Knoblauch, H., Tuma, R. 2011. Videography. An interpretative approach to video-recorded micro-social interaction, [in:]E. Margolis, L.  Pauwels(eds.), SAGE Handbook of Visual Research Methods, Los Angeles, pp. 414-430.

Malkiewicz, K., Mullen, M.D. 2009. Cinematography: Third Edition, New York.

Morgan, C. 2014. Archaeology and the Moving Image, Public Archaeology 13, pp. 323-344.

Stavrakis, E., Aristidou, A., Savva, M., Himona, S.L., Chrysanthou, Y., 2012. Digitization of Cypriot Folk Dances, [in:]M. Ioannides,D. Fritsch , J. Leissner, R. Davies, F. Remondino, R. Caffo (eds.), Progress in Cultural Heritage Preservation: 4th International Conference, EuroMed 2012, Lemessos, Cyprus, October 29-November 3, 2012, . New York, pp. 404-413.

Verrall, G.M., Slavotinek, J.P., Barnes, P.G. 2005. The effect of sports specific training on reducing the incidence of hamstring injuries in professional Australian Rules football players, British Journal of Sports Medicine 39 (1), pp. 363-368.

Vollmer, M., Möllmann, K.-P. 2011. Exploding balloons, deformed balls, strange reflections and breaking rods: slow motion analysis of selected hands-on experiments,Physical Education 46 (7), pp. 472.

Weinberger, E. 1992. The Camera People,Transition, No. 55, pp. 24-54.

Whittaker, J.C. 1995. SILICA ON CELLULOID: Some Current Flintknapping Videos. Lithic Technology 20, pp. 149-151.

Whittaker, J.C. 2004. American Flintknappers: Stone Age Art in the Age of Computers, Austin.

Whittaker, J.C. 2010. Weapon trials: The atlatl and experiments in hunting technology, [in:]Ferguson, J.R. (ed.) Designing experimental research in archaeology examining technology through production and use. Boulder, pp. 195-224.

Whittaker, J.C. 2012. Report on Knapping Activities in America. Lithic Technology 37, pp. 51-56.

Whittaker, J.C. 2013. Dart Speed Measurements, The Atlatl 26(1), pp. 9-12.



Be the first to comment

Leave a Reply

Your email address will not be published.