Applications of Computing in Industry : Lecture

Imagination Technologies designs GPUs for high performance mobile devices, deep within the chip lies the hardware responsible for performing fundamental GPU-compute tasks such as sin(x). The hardware implementation therefore needs to be fast, small (take up little chip area), low power and (importantly) correct. This talk will take a wander through the plethora of design options when computing these transcendental functions, touching on diverse fields such as computer arithmetic, numerical analysis, logic synthesis, number theory, automated theorem proving, floating point accuracy, decision procedures... Attendees should gain insight into these not so 'simple' functions and an appreciation of the range of challenges their hardware implementation presents to the various research fields involved.

Theo Drane started working for the datapath consultancy Arithmatica in 2002 after completely a Mathematics degree from the University of Cambridge, UK. He moved to Imagination Technologies in 2005 where his interests are datapath optimisation, verification and validation. After a two year sabbatical to work for an independent financial provider, Markit, he returned to Imagination to head up their datapath team while studying for a PhD at Imperial College London's Department of Electrical and Electronic Engineering.