For those who don’t know the tool: The Pixel Farm develops PFTrack, a node-based 3D camera tracking and photogrammetry system used in VFX, virtual production, and survey-driven pipelines. It exports to DCC tools including Autodesk Maya and integrates LiDAR and image based reconstruction.

The Pixel Farm has released PFTrack 26.02, a point update to its camera tracking and photogrammetry software. The release focuses on export reliability, Survey Solver behaviour and data validation rather than new solver technology. According to the published change list, the update introduces improvements to Maya export scripts, OBJ texture handling, LiDAR import logging and node connectivity within the Survey Solver. It also resolves a keyboard shortcut conflict in the Orient Camera node.

Maya export scripts corrected

The updated Maya export scripts now automatically parent background image planes to cameras on export. This addresses a long standing friction point in matchmove handoff. Previously, depending on scene configuration and export path, artists could end up manually re-parenting image planes inside Maya to maintain alignment between the solved camera and its background plate. The revised script automates this relationship at export time.

For tracking teams delivering camera solves into Maya-centric pipelines, this reduces post-export clean-up and the risk of misalignment during layout or scene assembly. The vendor statement does not detail changes to other DCC export scripts.

Survey Solver and OBJ texture loading

PFTrack 26.02 updates the way OBJ mesh textures with numbered filenames are loaded into the Survey Solver node. OBJ sequences or assets that rely on incrementally numbered texture files will now load correctly when imported into Survey Solver. The official notes do not specify naming conventions, but refer to numbered filenames in general terms.

For facilities using photogrammetry meshes or survey derived assets with multiple texture tiles, correct automatic loading reduces manual reassignment inside the node graph. It also limits the possibility of mismatched textures during camera solve refinement. The change appears limited to texture loading behaviour inside Survey Solver and does not reference modifications to the OBJ importer elsewhere in the application.

LiDAR import logging added

The update adds explicit error logging when importing corrupted or incomplete LiDAR survey data files. PFTrack supports LiDAR survey data within its camera-solving and scene-reconstruction workflows. In previous versions, incomplete or damaged files could fail during import without detailed diagnostic output.

With 26.02, the system now logs errors when such data is detected. The vendor documentation does not describe the logging format, verbosity level or where logs are written. It does state that error logging is triggered when corrupted or incomplete LiDAR files are imported.

Photogrammetry node connectivity fixed

A further fix addresses connector logic in Studio and Enterprise editions. Photogrammetry nodes can now be connected to secondary inputs of the Survey Solver where appropriate. The official wording specifies that this applies to Studio and Enterprise only.

This suggests that earlier builds restricted valid node graph connections in certain configurations, potentially limiting more complex photogrammetry driven solves. The updated connector logic restores the ability to route photogrammetry outputs into secondary Survey Solver inputs when supported by the node’s design.

The vendor has not published technical detail about which specific input sockets were affected or how the internal validation rules have changed. Not independently verified at press time.

Facilities using advanced node graphs that combine survey data, image based reconstruction and manual constraints should review existing templates to confirm expected behaviour after updating.

Scope and positioning

This release does not introduce new solver models, GPU-acceleration changes, or major UI revisions. It is positioned as a maintenance and workflow refinement update. The focus remains on export reliability, data validation and node graph correctness. For studios that rely on PFTrack for camera tracking, survey alignment and photogrammetry reconstruction, these adjustments address specific operational irritations rather than expanding feature breadth.

PFTrack is available in three editions. PFTrack Solo is priced at £699 for a permanent licence. PFTrack Studio is offered on a rental basis, with pricing listed at £59 for a five day rental period. PFTrack Enterprise pricing is available on request from The Pixel Farm. As always, facilities should confirm current licensing terms and support options directly with their vendor before procurement.

As always, new tools and updates should be tested in controlled conditions before being introduced into active production pipelines.

The post PFTrack update tidies Maya and LiDAR workflows first appeared on DIGITAL PRODUCTION and was written by Bela Beier.

For those who don’t know the tool: PFTrack is The Pixel Farm’s flagship 3D camera tracking and matchmoving software, used in VFX and virtual production pipelines for precise camera solves, photogrammetry, and scene reconstruction. It works alongside PFClean and integrates with DCCs from Maya to Nuke.

Lens distortion overhaul

The 25.11.13 release introduces new anamorphic and non-linear distortion models, smarter grid detection, and automatic selection of the most suitable lens model for each shot. The Lens Management system also now allows exporting distortion data directly into supported host applications.

Tracking tools refined

Both Auto Track and User Track nodes gain expanded editing controls in the Cinema workspace, including track grouping and motion prediction. The Localised Motion Prediction algorithm enhances feature consistency on fast-moving footage. Camera and Survey Solver nodes receive improved parameter control and faster iteration feedback.

Photogrammetry and modelling upgrades

The Photo Mesh node’s performance and memory management have been re-engineered to allow significantly larger mesh builds, particularly on macOS. Image Modelling introduces user-defined primitives, enabling custom reference geometry during scene reconstruction.

Sensor databases and third-party data

With the 25.12.19 build, PFTrack adds support for external camera sensor databases, beginning with integration of Matchmove Machine’s CamDB. This allows accurate sensor metadata import to improve focal calibration.

Node and UI polish

Across all 2025 releases, many nodes—including Geometry, Solver, and Point Cloud—benefit from small but meaningful updates: faster point cloud generation, improved solver convergence, and clearer error feedback. The general UI receives smoother interaction within Cinema viewports and enhanced undo stability.

Quality-of-life improvements

The 25.10 update improved bundle adjustment speed and tracking robustness on heavy footage. Version 25.09 refined Python scripting APIs, expanding access to internal data for pipeline automation. Collectively, these tweaks make PFTrack less prone to small but annoying interruptions mid-solve.

Licensing reshaped

After user demand, The Pixel Farm has reintroduced perpetual licences alongside subscription and floating models. Solo, Studio, and Enterprise editions remain available, with perpetuals offered for PFTrack only. Pricing is listed on The Pixel Farm’s store.

Tested before trusted

The Pixel Farm recommends reviewing node behaviour in controlled test scenes before deploying 2025 builds into live productions. As always, newer algorithms can subtly alter solver results between versions.

// https://pftrack.thepixelfarm.co.uk/documentation/changelog.html

The post PFTrack 2025: Matchmoving gets its groove back first appeared on DIGITAL PRODUCTION and was written by Bela Beier.

For those who don’t know the tool: 3DEqualizer by Science-D-Visions is a high-end 3D tracking and matchmoving tool used in VFX and compositing workflows. It connects to Maya, Blender, Nuke and others to generate precise camera and object solves for integration of CG elements.

Over the winter break, Science-D-Visions released 3DEqualizer4 Release 8.1. The update focuses on precision and pipeline reliability, with several technical improvements aimed at matchmove and postvis specialists.

Piggyback rigs and postvis tools

The standuot addition is a new control pane for computing main camera motion from a piggyback or auxiliary camera, including 360-degree setups. This workflow allows the calibration of static camera-to-camera transformations directly from calibration or production shots. Science-D-Visions also provides a demonstration video of this workflow. To simplify onset capture, a new import script lets users load footage directly from movie files, reducing dependency on external transcoding tools.

Small but targeted refinements

Release 8.1 introduces a new high-pass filter in image controls, a frequently requested addition for analysing fine detail in tracking plates. It also adds start-frame parameters to USD and Alembic exports, allowing more precise control when aligning sequences in downstream departments. A new Attribute Editor plugin demonstrates how to reference 3D models directly from USD files, intended as an example for studios developing their own custom pipeline tools. The update further includes a rewritten Blender export script (v2.0) that matches the design of the Python-based Maya exporter from Release 8.0.

Technical updates and Python 3.11

Under the hood, 3DEqualizer 8.1 moves to Python 3.11. The runtime update brings better session management, new methods for converting OpenCV-style lens distortion to native 3DE formats, and improved image buffer handling. Several older tools were refined, including decimation of 3D models with correct UV handling, enhanced curve filtering, and better distance constraint behaviour. The development team also addressed multiple stability issues: bug fixes cover 2D track exports to Flame, curve editor display errors, long-lat conversion crashes, and problems with rolling shutter exports.

Export and scripting improvements

Maya export now includes user preferences, the “object pgroup z distance” tool gained motion path transfer functionality, and file requesters now display file sizes in human-readable format. The ScriptDB installer received fixes for SSL warnings, and the “reel in points” and “import 2D tracks” tools were corrected for minor issues.

Pricing and compatibility

3DEqualizer 8.1 runs on Windows 10+, glibc 2.17+ Linux, and macOS 13.5+. Perpetual licences cost €7 699 (around USD 9 065), while subscriptions range from €65 per week to €1 704 per year. As always, new builds should be tested in non-critical environments before being rolled out to production shots, especially given the shift to Python 3.11 and updated exporters, which may affect pipeline compatibility.

// https://www.3dequalizer.com/index.php?id=251111_01&site=devblog

The post 3DEqualizer 8.1 Adds Piggyback Camera Solves first appeared on DIGITAL PRODUCTION and was written by Bela Beier.

In recent years, Foundry has not only tightened up the release schedule considerably, but also made it much more regular. Nevertheless, version 14 is not a maintenance update, quite the opposite. Not only are new features being added, but in some areas the existing ones are being shaken up. The biggest and most important area is probably the 3D system, which has been completely overhauled.

Cutting off old habits

Sometimes the time has come to make a clear cut. Even as a software developer, it becomes increasingly difficult over the years to work on the same old architecture. With version 14, Foundry therefore decided to rewrite the 3D system, which has always been one of the core elements of the software, from scratch. In order to be well equipped for the coming years, USD (Universal Scene Description) was chosen as the basis. USD not only offers massive leaps in performance, but is also being supported by more and more other manufacturers.

To accommodate larger pipelines, it will be possible to select your own version of USD to use, regardless of the version of Nuke you are using. However, those who do not use USD in their pipeline should not have any disadvantages compared to the previous system, as everything that the new 3D system requires is already fully integrated in Nuke. As such a changeover is of course a big step, the old 3D system is not being retired just yet. It is still available in Nuke for now and, perhaps more importantly, the new system is currently in beta.

Why USD?

Better performance alone is a nice effect, but there were more reasons for switching to the USD-based system. A major Achilles heel of the old system was the handling of large imported geometries, where there was no way to isolate individual parts and edit them individually. The new Scene Graph and Path and Mask Knobs form the basis for this.

The Scene Graph

To organise everything from objects and materials to cameras and lights effectively and uniformly, there is a new panel in Nuke 14 called “Scene Graph”. Here you can find all the objects in a scene in a hierarchical structure and can name them (with a new token system if desired), switch them on and off and create dependencies. The Scene Graph and Viewer are closely linked and objects selected in the Viewer are also selected in the Scene Graph and vice versa. A search function is also integrated.

Path and Masks Knobs

All nodes in the new 3D system that generate or import geometry now have a path as an entry, which simply refers to the name of the node by default. However, it is linked directly to the scene graph and can also access the token system. This means that more complex scenes can be named more quickly and cleanly and dependencies can also be created via the path without having to play connection limbo in the node tree.

The name “Mask Knob” is somewhat misleading, as it is actually a way of defining which objects in a scene should be influenced by a modifier node such as a GeoTransfom. The new default behaviour now looks like this: If you insert a node below another node, the result of the last node is affected – no longer the entire stream. However, if you want the new node to influence several objects, you can assign it to the mask knob from the scene graph or via an expression.

USD Lights and Materials

If USD-based geometries are imported into the new system, lights and materials are now displayed correctly and thus provide a better preview of the actual asset. The new Preview Surface Node is a new way to create more complex materials within Nuke and is seen by developers as the first step in a development that is progressing through beta. There are also new USD Lights. Unlike the old system, lights are no longer converted into Nuke’s own format, but are displayed exactly as they looked in the USD-based DCC software they came from.

Old and new

As already mentioned, the old and the new system will both populate the Node Graph for the time being. However, they are not compatible with each other. The old nodes retain both their name and their colour, while the new ones stand out with a new nomenclature (TransformGeo is now called GeoTransform) and a green that complements the old red. They also appear with a small “Beta” notice. Speaking of beta: Foundry wants to use this beta to actively incorporate feedback from the community, which is why there is now also a forum on the Foundry site for the new 3D system.

Unreal

The Unreal Reader Node for Nuke X and Nuke Indie is completely out of beta. It can be used to transfer maps and sequences directly from an Unreal Editor session to Nuke. To do this, a “Nuke Server” plugin must be installed in Unreal, which then provides the session for Nuke. From here, the artist can have exactly the channels and passes that he needs made available, as everything is provided “on-demand” by Unreal. Various types of crypto matte can also be requested. The current OCIO colour transforms are also supported.

Cameras can be synchronised with the Unreal session as well as replaced by Nuke. The whole thing works non-destructively for the Unreal session, so that you now have the freedom to position the camera slightly differently in Nuke than in Unreal. If you are happy with your result and want to make sure that further changes to the Unreal session are not passed on to Nuke, you can generate an EXR sequence of all requested channels and passes from the Unreal Reader Node.

A particularly nice feature is the stencil mode, which allows you to split individual objects in the Unreal scene into separate layers. Even shadows and occluding objects can be removed. You also have the option of rendering the Nuke scene as a CubeMap, which allows you to transform your scene into a Lat:Long image or an Environment Sphere in Nuke.

The cat gets a new home

The machine learning tool “Copycat” is getting a major performance boost. Ten to twenty per cent faster learning should be possible on current Nvidia cards. The automated rotoscoping of human bodies has been accelerated even further. New setting options should dramatically speed up training here. Finally, users now have the option of displaying only the thumbnails relevant to them in the LayerContactSheet view for training.
Outside of Nuke, Foundry has created the “Cattery”, a free repository where copycat files contributed by the community can be downloaded. The whole thing is sorted by category and will hopefully continue to grow over the coming months.

Bokeh

Foundry acquired the rights to PG Bokeh from Perigrine Labs. This popular depth of field tool is natively available in all Nuke versions from version 14.02 onwards. PG Bokeh has already been used in many major productions and is the only depth-of-field tool that can be combined with deep compositing.

Login-Based Licence

Nuke 14 offers Foundry customers a further option for managing their own licence or licences. In addition to the classic node-locked licence or an RLM-based server licence, you can now also opt for a login-based licence. Artists can log in with their email address and a password to activate their licence. This makes it much easier to take a licence to the home office.

The online platform also offers numerous options for managing multiple licences for teams. Foundry emphasises that this licence model is purely optional and anyone who doesn’t want to switch doesn’t have to. This is also important for anyone who still has to work in parallel with older versions of Nuke, as the login is only available with Nuke 14.

Subscription

Subscription is always a double-edged sword. Some see greater flexibility, more predictable costs and the opportunity to get started without a major investment, while others simply see subscriptions as a “money grab”. The Foundry is now also offering the Nuke Family as an annual subscription. The prices are as follows: Nuke costs customers 2,699 Eur per annum, Nuke X 4,099 EUR and Nuke Studio a whopping 4,999 EUR.

As with all Foundry products, there is no monthly subscription model and unfortunately – unlike other products – no monthly payment option. Although there is still the option of quarterly rental, the prices here are not really attractive – they are over fifty per cent of the annual subscription. Existing customers still have the option of purchasing the previous “perpetual licences” until the end of the calendar year and then continue to receive maintenance for them. However, Foundry has also tightened the price screw here: Nuke now costs EUR 5,118, Nuke X EUR 9,758 and Nuke Studio EUR 11,358.

Conclusion

With the new 3D system, Nuke 14 lays the foundation for the coming years and sets a future-proof standard with USD. The system of making a feature available to the user base as a beta for one release and then bringing it out as a final in the next release or the one after that has already worked well in the past with the Unreal Reader or with the sync sessions in Nuke Studio and Hiero(Player). The Unreal Reader may only be of interest to a certain target group at the moment, but the more popular virtual production becomes, the more comp artists will have to deal with Unreal. They have the basis now.

Copycat gets a nice, comprehensive product maintenance – important, because Nuke’s approach to machine learning only works as long as it is regularly updated. The fact that all Nuke users now have PG Bokeh on board is really pleasing – especially because of the support for deep compositing. The login-based licence system will certainly provide more flexibility in the medium term.

One criticism is that, once again, the small Nuke – without X – has hardly any features – of course the new 3D system is also being introduced here. But apart from the bokeh, all the new features are limited to Nuke X and Studio. Copycat is still a feature that is reserved for Windows and Linux users due to its close integration with NVIDIA.

It remains to be seen whether the announced port to Apple Silicone will also let the cats jump over the Mac. And last but not least, the switch to subscription will not be greeted favourably everywhere in the community. For larger VFX studios in the feature sector, an annual subscription may actually offer more flexibility in scaling the pipeline. Customers in the advertising sector and freelancers, on the other hand, will not find an attractive offer here to be able to react appropriately in an ever faster changing market.

The post What’s New in Nuke 14? first appeared on DIGITAL PRODUCTION and was written by Christoph Zapletal.

And above all: how will we look back on our mistakes today? In Adjourned, our contemporary climate summits are already a historical relic, a classic theatre play from the past. Having failed to fulfil their actual task, their former existence now serves only serves to amuse a few survivors. In the dystopia, our present-day climate theatre becomes a perfect source of satire. We realised this in our second film at the HFF Munich under the direction of Prof. Jürgen Schopper as follows.

In the second year of our training, as visual effects students at HFF Munich, we had the opportunity for the first time to develop a short film idea together with other departments at the university, such as production, camera and screenplay, which had to be realised with the help of visual effects. The advantage of this co-operation was that the effort and feasibility of potential visual effects shots could already be planned down to the smallest detail during script development. We tried to work as closely as possible to the principle of virtual production, for which close co-operation with the other departments was a prerequisite. This collaboration was extended by VFX industry professionals, who supported us again and again during the course of the production. Petra Hereth was responsible for team coordination.

by Christian Geßner, Nicolas Schwarz, Chris Kühn,
Malte Pell, Tobias Sodeikat and Jonas Potthoff

From the first script version in October 2021 to the start of shooting in April 2022, we were involved in the preparation of the film in many ways: in parallel with the script development, we created the first concept art, which we used to design the look for our world together with author Larissa Dold and director Matthias Zentner.
At the same time, we constantly updated the VFX calculation with the latest findings from the joint meetings from the start of the script work in order to guarantee the feasibility of the visual effects. In this way, we were able to ensure that the VFX effort was in line with the time frame available to us during book development. It also gave us the opportunity to prioritise during development which parts of the story would benefit from visual effects and where they were not absolutely necessary. In addition to storyboarding and concept art, this process was accompanied by us with the support of Luis Guggenberger. With the help of 3D mentor Berter Orpak, we created a 3D pre-visualisation in a very short time, which not only showed roughly blocked scenes, but already included tracking shots, animated characters and a rough sound layout.

Production: University of Television and Film Munich 
Duration: 07:51 min (25fps)
Resolution: 2K (Aspect Ratio: 2:39)
Render Engine: Redshift, Mantra
Texturing Software: Substance Painter
3D Software: Blender, Houdini
Compositing Software:  Nuke

Once the script was largely finalised in February 2022, we were able to focus on the actual filming preparations. As we already had a very concrete idea of the shots to be filmed based on the 3D previs and the entire shoot was also very VFX-heavy, the shooting schedule was developed directly in the VFX department in close consultation with the producers Luisa Eichler and Michaela Mederer. In various production meetings and shooting schedule discussions, we also planned the concrete realisation of each shot with the other trades – for example, we were able to determine exactly where the real studio set would merge into the digital world with the art department under the direction of Katja Severin and how we could make these transitions as unobtrusive as possible. It also took some coordination with the costume (costume designer: Katharina Ost) and make-up (key make-up: Sabeth Kelwing) departments in order to be able to shoot VFX-compatible costumes and SFX make-up, such as burning hands, as sensibly as possible.

During filming (camera: HFF graduate Teresa Renn), we as VFX students were represented in various positions on set, with the role of visual effects supervisor being swapped on a daily basis so that every student had the opportunity to experience this function for themselves. We were supported by professionals from the industry (Jan Stoltz and Dietrich Hasse). But we were also very busy outside of this position: Both the set and the actors* had to be scanned and data also had to be collected from the set after each take.

After the end of the shooting week, it only took another week until we were able to lure the first VFX shots from a rough cut and start editing (editing: Michael Dervenski & Matthias Zentner). To do this, we updated the latest cut version in DaVinci Resolve with the online material and exported the shot plates and references in 16-bit Linear EXR sequences for editing.

Order in the chaos of assets – the modelling

Hopefully it will still be a few years before it looks like the Vertagt universe here in Munich. As the only thing that already existed during the production period was an empty stage, the world around it had to be created from scratch in the computer. We used a combination of specially created, purchased and scanned assets.
This resulted in a total of over 50 different models, some in up to six different versions. The largest of these was a broken container ship. We reworked the purchased 3D models with our own textures for our film in order to adjust the level of detail for the project and the individual shots. The realistic replica of a whale skeleton, shipping containers and oil drums are some of the many assets that were added to the interior and exterior of the wreck.

Image references of possible shipwrecks, damaged cars, aeroplanes, harbours, pipeline systems and materials affected by sand and weather formed the basis of our research. From the references summarised in mood boards
References summarised in mood boards, we were able to pick out new ideas throughout the project and recreate them in detail. The interior of the shipwreck was continuously filled with first large, then smaller and smaller objects.

The 3D models were modelled and “UV unwrapped” in Blender. In order to cope with the large number of differently coloured and dented containers, we created various textures and models, which were assigned procedurally in Houdini. To ensure that accessories such as cables, ladders and rubbish could still be placed on the containers and that the AO map could be calculated correctly in Substance Painter, the scene was initially blocked with cuboids.

In addition, we modelled a second interior, which is revealed to us in the long return journey as the ruins of an engine room, gradually adding information about the location and the dystopian world. The centrepiece is an old, oversized ship’s engine. The difficulty here was to combine the realistic depiction of the working environment in an engine room with as open an environment as possible to show the size of the wreck. Some elements run through both areas of the ship in order to connect them as seamlessly as possible. Both the rust shader and the cables hanging from the ceiling serve as a link here.
“Among other things, a model looks realistic when the eye and brain cannot grasp all the elements at once, but discover new things bit by bit”. To follow this advice from Dirk Mauche (our active support in the modelling area), a tangle of cables, chains and rope ladders provides the necessary details (and disorder) in the picture.

A crawler learns to fly – rigging & animation

The main CG character, a cockroach called “Despair”, was already delivered with various animation cycles and a rig, but first had to be fed into the pipeline. To do this, we created a motion cycle system in Houdini that could be used to efficiently fade between the various animation cycles. With the help of several controllers, it was possible to run the CG cockroach on a spline curve at different speeds. It only became problematic when more complex movements were required. A rotation around its own axis, when the cockroach has to perform a somersault to escape from a tricky situation, became an endless search for the right point of rotation. The problem was solved using the good old “brute force” method. A second locator served as a new anchor. Despite using ready-made animation presets, we were once again lucky enough to be supported by Prof Melanie Beisswenger as an animation lecturer. We learnt that even the smallest changes in the running speed of a cockroach can provide a lot of information about how it can affect the viewer.

Continuity of light and shadow

We realised early on that we needed low key lighting on set. The only source of light was to be the extremely bright lightbeams that would shine through small holes in the outer wall of our shipwreck, but from whose scorching heat the ship’s inhabitants were to hide. The darkness surrounding them was only to be discreetly brightened up by their bounce lights. However, as is so often the case, it turned out during pre-production that the reality was different. We couldn’t completely do without a few bright accents in the form of artificial light lamps on set in order to capture the mood and create enough detail in the blacks for the camera. This resulted in an interplay of lightbeams, which have a daylight and therefore cool colour temperature, and practicals, which emit artificial light and therefore have a warmer colour temperature.

In order to get as close as possible to the properties of real sunlight in our studio environment, our head lighting technician Torsten Baier organised a large PAR spotlight, which was directed into the right paths using mirrors and whose light cone was suitably shaped. Thanks to the parallel beam path of this spotlight, it hardly loses any brightness at a distance and thus corresponds to our perception of sunlight. A lot of artificial fog on set also ensured that the beam of light was visible in the air and that our surroundings looked appropriately dusty. Now it was time to digitally match the lighting from the set..

The long shots were particularly challenging, in which we had to extend our shot set outwards in order to be able to tell the interior of the shipwreck credibly. The transition between the filmed stage and the digital set was concealed by 3D-scanned scrap parts and models, but the lighting mood had to match exactly for both. The light beams also cross the “boundary” between the filmed and digital plate, which is why they had to be partially supplemented, but also completely retouched and digitally replaced. The aim was to recreate the correct angle, diameter, intensity and colour temperature of the real light beam. This was done in Houdini and then rendered with Redshift. In compositing, we added further details such as animated fog and particles floating in the air to the light beams, further blurring the line between real and computer-generated footage. We received a lot of support from our lecturer Frank Dürschinger, who provided us with the final food for thought to create a realistic image.

Other shots had a smaller CG component, but this had to be all the more seamless. As a completely digital character, our cockroach needed lighting that was as true to the original as possible, including refraction, reflection and shadows, in order to fit exactly onto the plate being shot and integrate itself into the real world. And that in every setting size – whether it appears large in the picture or only takes up a few pixels. To achieve this, we used a 360° camera to take exposure series of the respective location and its lighting during the shoot and then stitched them together to create HDRIs. The HDRIs served as the basis for the lighting of the respective 3D scenes, but were often supported by additional spot and area lights. Finding the right balance between too much and too little additional lighting was a challenging fine-tuning task, but it was also a lot of fun.

FX simulations: Sand and bone dust

The requirements for the FX simulations ranged from simple setups like the main title crumbling into dust and a cockroach kicking up sand, to big challenges like the skull shattering on the ground and a photorealistic desert that had to work in both close-up shots and a super long shot. We were once again lucky enough to have Felix Hörlein on board as a lecturer for our second project at the HFF and were therefore able to draw from the full range of possibilities.

In the final shot of the film, we establish a vast, open desert landscape swept by sand and wind. The biggest challenge in realising this haze was the enormous size of the terrain. With an area of three square kilometres, a simulation approach was needed that could simulate different resolutions for different areas. Thanks to the help of Felix Hörlein, an LOD system was created based on the distance to the camera. In the first step, a point cloud was generated at three different points in time using Mantra’s ray tracer, which subsequently produced a representation of the entire visible geometry.

In combination with the LOD system, this resulted in a significantly reduced footprint that could be used to generate the volume. All parameters that have an influence on the volume must also be controlled globally in order to ensure that the edges of the individual LOD areas blend into each other as much as possible. To save computing power, time and storage space, the simulations on the individual wedges only start shortly before they are visible in the camera. In addition to the volumes, we used a particle simulation in the foremost areas of the field of view to get as close as possible to the “sandstorm” target. This was controlled by the Vel field from the volume simulation.

Despite all the cost-saving measures, one workstation per version took an average of around six days to simulate all the individual parts, generating a cache size of 5 TB. This is largely due – at 1250 frames – to the length of the shot. Mantra became our render engine for all simulations. Rendering was always done with mattes, IDs and cryptomattes to give the compositing specific access to the individual simulations.

The skull crashing to the ground and disintegrating is the dramaturgical highlight of the film and demanded a lot from us due to the size of the shot. The basis of the effect is an RBD simulation with custom constraints for the three areas “skull”, “jaw” and “teeth”. To make the breaking more interesting, the option “switch constraints when broken” is also active for a selected area that contains the main features of the skull. Using a second constraint type, the selected geometry is held together for longer before it finally breaks. The skull has been acquired and prepared for the simulation. Three separate “rbd-fracture-nodes” are used to divide it into fragments in order to get as close as possible to the real properties of the different bone thicknesses.

To save memory space and optimise performance, the RBD simulation was only cached as points. This allowed us to remesh the geometry of the skull only after the simulation and thus determine the final resolution of the mesh. A growth solver, which is initiated by changing the distance between neighbouring points when breaking, starts the “disintegration” process. All infected polygons are passed on to another solver, which freezes them at their position and feeds their centre as a single point into a POP network. After the simulation of the POP solver, the polygons are linked to the animation of the respective points using an ID. This allows the dissolving parts to be simulated and saved in a very computationally and data-efficient manner. The individual points are controlled by a velocity field that was generated with the RBD simulation.

In order to intensify the shattering, we have also added two further particle simulations, which on the one hand provide more volume, but more importantly create the appearance of a dust-covered floor – and thus correspond to the rotated plate. We were also able to use a similar but significantly simplified technique for two shots in which the hand and arm of an actor are burnt by the sun and disintegrate into dust.

Digital film set

The use of the Faro laser scanner provided great technical support for post-production. Purchased especially for the Visual Effects degree programme, we were able to create 3D scans of the entire set, which later enabled us to read off accurate distances and create precise match moves. Our initial expectations that tracking would not be a problem with smaller camera pans were quickly dashed by the bitter realisation of how important a good match move is.
Thanks to Matchmove lecturer Ando Avila, however, we were also able to overcome this challenge. The scans also form the basis for the digital replica of the film set. An initial rough reconstruction provided information on the position and tilt of the camera, and details such as wooden struts, chairs and lamps were subsequently added. The recording of the actors was rotoscoped and projected onto the geometry. The projection of the recorded stage onto 3D geometry was used several times. Especially to create subtle
3D projection was an important tool, especially for credibly narrating subtle camera movements with parallax shifts. Of course, it was also helpful to use rudimentary geometry.

Pipeline

In the previous year and during the course of the project, we learnt a lot about effectiveness and workflow. Many shots with the same environment led to the decision to integrate Houdini digital assets into our workflow. This allowed us to tackle steps such as modelling and animation at the same time. But that alone is not enough to keep a cool head and keep things organised. Our Pipeline TD Jonas Kluger, who supported us in the workflow with Shotgrid Studio and modified the pipeline according to our needs, made sure of this. This enabled us to keep a better overview of the tasks and stick to our production schedule.

Compositing

The final step in post production was compositing. As the production offered everything from simple retouching to elaborate full CG shots, the six students had to finalise the more than 40 visual effects shots in a short space of time. Deadlines are deadlines and they don’t wait any longer than our climate. The comp work took place exclusively in Nuke and for some of us, working on the project was a good introduction to the software. Fortunately, we were able to rely on the support of Christoph Zapletal and Jens Schneider, who were able to teach us better Nuke structures and advanced techniques. This was the only way we could work on many different shots at the same time.
At the end of the film, there were also several shots in which we had to digitally add ash flying through the air. The “Das Element” software helped us enormously here, as plates can be organised and found quickly. These
Plates were inserted after they had been created on the last day of filming and thus supplemented the library of purchased stock footage. From a large ladder we threw dust particles, torn paper or we blew mist through a hose into a rubber glove. As everything was filmed against a black background, we could simply use these elements in Nuke and add them to the shot plates on the stage.

Nice and dirty – the colour grading

Once shooting, editing and visual effects were complete, the film was given its finishing touches. Andreas Lautil from Pharos created an individual look to match the eccentric style of the dystopian theatre. Although the grading is based on fairly classic teal and orange lighting on set, it was pushed in a much dirtier direction here. The dark areas were made dirtier and greener so that the audience could put themselves in the shoes of the run-down world and better empathise with the society in the film. Once again, it became clear how much influence each department has on the final project. Originally, a rather desaturated style was planned – clearly recognisable in our concept drawings. However, this was greatly varied in the grading. Now the contrasting cold tones in the depths harmonise with the warmer highlights and thus also with the absurd events in front of and behind the stage.

The sound design and funny background conversations

The quiet murmuring, the sound echoing from all directions and the creaking of the rusting steel transported us to the gigantic wreck in the middle of the desert for the first time in the making of the film. Thanks to Dr Rodolfo Silveira’s work in sound design, the silent shots that accompanied us for months became an atmospheric experience. In several sets, vocal ranges and positions, we students were even able to lend our voices to the digital audience ourselves and push our vocal chords to their limits. What do old, injured people who are on the verge of dying of thirst sound like when they nevertheless enter a bizarre theatre? The HFF’s in-house recording studio demanded a lot from us in terms of acting, but also brought to light a hidden passion for some of us. As the film was shot entirely in a studio, it was necessary to place all the metallic sounds of a ship in the very first shot of the film.

Thus, the entire opening sequence, which takes place in a corridor on the side of the theatre stage, was completely resonated to convey the echoes, vibrations and thus the dimensions and weight of an abandoned oil tanker. Through this strategy, the sound information creates a new level of understanding that complements and enlarges the film for the audience and prepares them for the climax and the central conflict of the story. The sound design was supported by an atmospheric composition by musician Meredi, who was able to capture a unique mood for the film using specially created sounds.

Outlook

Above all, working with industry professionals has shown us that even a little insight and practical experience in all departments helps us to gain an understanding of other departments. The mentality that visual effects is just a part of post-production is a thing of the past, because film-making begins for visual effects during script development. Virtual production is becoming increasingly relevant for well-planned productions with a visual effects component. The first internal presentation took place as part of the “VFX-Reel” 2022 exhibition in the Audimax of the HFF Munich.

Team

• Script: Larissa Dold
• Director: Matthias Zentner
• Producer: Michaela Mederer and Luisa Eichler
• Director of Photography: Teresa Renn
• Gaffer: Torsten Baier
• Production Designer: Katja Severin
• Costume Designer: Katharina Ost
• Key Make-Up: Sabeth Kelwing
• Composer: Ina Meredi Arakelian
• Editor: Michael Dervenski, Matthias Zentner
• Colorgrading: Andreas Lautil
• VFX Supervisor: Jan Stoltz and Dietrich Hasse
• Project Supervision: Prof. Jürgen Schopper
• Project Consultant: Dr Rodolfo Anes Silveira
• VFX Pipeline TD: Jonas Kluger
• 3D Mentor: Berter Orpak
• Team Assitant: Petra Hereth
• Technical Tutor: Moritz Rautenberg
• VFX Producer: Chris Kühn, Jonas Potthoff
• Modellers: Christian Gessner, Chris Kühn, Malte Pell, Tobias Sodeikat
• Texture Paint Artists: Christian Gessner,
• Chris Kühn, Malte Pell, Tobias Sodeikat
• Simulation Artist: Nicolas Schwarz
• Animators: Chris Kühn, Tobias Sodeikat, Malte Pell
• Compositors: Jonas Potthoff, Chris Kühn, Nicolas Schwarz, Malte Pell, Tobias Sodeikat, Christian Gessner
• Visual Effects Editor: Jonas Potthoff
• PreViz Artists: Christian Gessner, Chris Kühn
• Sound Design/Re-Recording Mixer: Dr Rodolfo Anes Silveira
  • VFX Mentors
• Concept: Luis Guggenberger
• Modelling/Texturing: Dirk “Superdirk” Mauche
• MATCHMOVE Ando Avila
• FX: Felix Hörlein
• Mattepainting: Jens Schneider
• Animation: Prof. Melanie Beisswenger
• Compositing: Christoph Zapletal
• Lighting: Frank Dürschinger
  • Cast
• A Chancellor: Viola von der Burg
• A Conservative Party Member: Claus Peter Seifert
• Green Party Member: Ronja Katharina Brusa
• A Prompter: Patrick Bimazubute
• The Choir Singers: Emil Vorbrugg

The post VERTAGT – VFX for a dystopian satire first appeared on DIGITAL PRODUCTION and was written by HFF Hochschule für Fernsehen und Film München.