Viridien has entered a collaboration with NVIDIA to improve the high performance computing foundation behind seismic imaging workflows. The focus is on optimizing the algorithms used to process and image the subsurface on NVIDIA accelerated computing platforms, aiming to reduce the time and cost required to turn large seismic datasets into images that are usable for technical and commercial decisions in energy and geoscience projects.

 

How the Workflow Optimization Will Work

 

The effort centres on tuning Viridien’s seismic imaging algorithms to take deeper advantage of NVIDIA’s HPC capabilities, including tensor cores and mixed precision computing. In practical terms, this type of optimisation is about increasing throughput without sacrificing the reliability of the output, because seismic imaging is constrained by both physics based computation and the scale of the data. Mixed precision approaches can materially improve performance when implemented carefully, but they require validation to ensure numerical behaviour remains stable enough for decision grade interpretation.

 

Why Tensor Cores and Mixed Precision Matter for Imaging

 

Seismic imaging workloads are computationally heavy because they involve repeated modelling of wave propagation through complex geology, often across large 3D volumes, and frequently require iterative refinement. Hardware features like tensor cores can accelerate specific matrix operations that are common in modern computing workloads, and mixed precision can increase speed and reduce power use by using lower precision where it is safe and higher precision where it is needed. The value proposition is clear if the output remains robust, because sharper images produced faster can shorten interpretation cycles and improve the economics of exploration and development decisions.

 

Read more: European Marine Energy Centre Secures €17.5M to Expand Tidal Test Facilities in Orkney

 

What Each Partner Brings to the Table

 

Viridien is positioning its role around expertise in seismic imaging and fully managed HPC solutions that support these workflows at scale. NVIDIA contributes accelerated computing platforms and a software ecosystem designed to extract performance from modern GPU based infrastructure. The collaboration is framed as a way to align algorithm design, platform optimisation, and operational deployment so that subsurface teams can run complex imaging jobs more efficiently rather than treating hardware and software as separate procurement decisions.

 

Decision Impact for Energy and Geoscience Clients

 

The companies are linking performance improvements to business outcomes that matter in subsurface work. Faster, more reliable imaging can reduce uncertainty in prospect screening and improve well placement decisions by giving interpreters clearer structural and stratigraphic signals. The expected downstream effect is lower exploration cost through reduced dry hole risk and less compute time to reach decision grade results, which is particularly relevant as energy companies balance technical ambition with tighter capital discipline and rising scrutiny on operational efficiency.

 

What to Watch as This Moves from Promise to Practice

 

The credibility of this collaboration will depend on how improvements are measured and validated in real production environments. Performance gains are only valuable if image quality and repeatability remain consistent across different basins, acquisition types, and processing objectives. The most meaningful signal will be whether users can reduce turnaround time, compute spend, and operational friction without introducing new verification burdens, because seismic imaging teams will not trade speed for uncertainty when the stakes involve drilling decisions.