Wednesday, 21 September 2011

Different Visions

Back to Laxton tonight, wearing a by now slightly unfamilar hat, to talk to the Laxton History Group about the past excavation and survey work in the village and castle.

I can't resist including one of the CryEngine visualizations of the castle survey in the presentation, and have settled for that produced in CE2:

It strikes me how different the same basic data looks in CryEngine 1:

Nothing deep here, but it strikes me as how CE1 has something of a cartoonish, fantastical feel about it.  Now we may deride that as lack of photo-realism, something that undermines the authenticity of the visualization, and if we did we might have a point.  But there's also something attractive about this, rather like naive art, the studied lack of realism invites a different kind of engagement and reject criticism for not looking real enough.  This might just be my liking for the speed and simplicity of CE1 speaking, but if game based rendering of landscape is about more than simply realism (and I believe it is) then maybe naivety and a simple pallet have something to offer.

Friday, 16 September 2011


Two exciting pieces of research affecting Stonehenge have been pointed out to me and are too good not to pass on.  The English Heritage funded 3D Stonehenge model has used high resolution terrestrial laser scanning to produced an unparalleled, accurate three-dimensional model of the stones.

Meanwhile, at Bournemouth University the team responsible for one of the recent campaigns of fieldwork at Stonehenge have completed the Google Under the Earth: Seeing-beneath-Stonehenge project. Funded by the Google Research Program. Birmingham Postgraduate Lawrence Shaw has worked on this project and blogs about it here.

Great work, and nice visualization in both cases.

Thursday, 8 September 2011

Exploring Visibility and Atmospheric Occlusion in CryEngine

The role of visual appreciation of landscape has taken a leading role in the archaeology of landscape, encouraged both by theoretical approaches and the utility of GIS softwares for automatic analysis of the visual characteristics of large landscapes.  Various writers have critiqued GIS studies of visibility and proposed enhancements to and refinements of visibility studies (eg. Llobera 1996 and 2000; Ogburn 2006).  In essence arguments for the refinement of visibilities studies focus on the need to appreciate the impact of range, view direction and target size on visibility. Additionally, a number of authors have pointed out the shortcomings of both theoretical and practical approaches to visibility studies, in particular the fact that many studies ignore or misrepresent the potential impact of vegetation and past vegetation patterns on landscape and visibility. 

I have been experimenting with using CryEngine to model different degrees of visual occlusion of landscape, based on atmospheric fog and rain.  Sandbox conveniently allows varying vegetation scenarios for the same landscape to be stored as layers that may be turned on and off to explore the visual impact of changes in vegetation.  Sandbox's environment controls allow alteration of the character, density and occlusion distance of atmospheric fog and addition the impact of rain on visibility may be simulated through particle effects.  The graphics here illustrate CryEngine simulations of the effect of increasingly dense atmospheric fog (with view distance decreasing from 200m to 500m) and the addition of rain on visibility of Stonehenge as viewed from The Cursus, a distance of just over 2km, and can be compared with a traditional GIS-derived two-dimensional viewshed diagram from the same location.

I think this approach has some potential for exploring changing visibility in landscapes where views and indivisibility are considered significant.  I'll be exploring different vegetation patterns in a similar light in the near future.

Llobera, M. 2001. Building past landscape perception with GIS: understanding topographic prominence. Journal of Archaeological Science 28, 1005-14.

Ogburn, D.E. 2006. Assessing the level of visibility of cultural objects in past landscapes. Journal of Archaeological Science 33, 405-13.

Wednesday, 7 September 2011


One of the nice things to come out of Technological Advances in Landscape and Heritage Management Recording has been the chance to collaborate with colleagues in Ireland on some experimental visualization.  

I'm very grateful to the Discovery Programme for supplying some of their very high resolution Fli-Map airborne lidar (c.2m ground resolution) to play about with, including attempting a visualization of the iconic Hill of Tara in CryENGINE.

Tara is an interesting landscape to work with.  The earthworks are well defined and the lidar data quite startling in its clarity, there are also numerous crowd-sourced ground level photographs to work from.  Working in CE1 was a little disappointing, although layering textures from derivatives of the lidar terrain data (slope mapping, solar insolation modelling, etc) shows promise.  Far better at the moment is CE2, with its much improved lighting and vegetation.  Below is a rough render of the initial model, with more to follow in due course. 

Getting the vegetation right here is key and requires considerably more work before even beginning to think about adding arcitectural detail. Great to have such fantastic data to work with.