In Gonzalo’s class we were given the task of doing some roto on around 100 or so frames of animation regarding a man running down a bridge where both the man and bridge were meant to be cut out.
I started out with getting a roto for the ground and pillars on the bridge as the camera isn’t quite a still and would need to be tracked. I made a track and then had it tracked fairly well as while there is movement on the camera, it’s very subtle and doesn’t move a whole lot.
I managed to get a roto of all parts of the bridge and afterwards went about slightly adjusting and matching the tracker’s movements onto the roto coordinates.
Afterwards i went about attempting to track portions of the guy running’s body in an attempt to make rotoing slightly easier as opposed to moving the roto every 5 or so frames but this ended up not working in the end and ended up scrapping the tracks.
I went about breaking the roto into different sections of the body to get a bit more control, starting with separating the legs into four different roto portions.
Afterwards i went on to doing the chest and torso portions to try and separate out the roto to get more overall control.
Lastly i went about separating the arms and head from the roto, as I’ve learned prior that using as few points as possible and separating it into as many shapes as possible gives both more control over a roto and makes it easier to do as you’re not tweaking thousands of points.
This is what the alpha looked like on just the guy running, as well as what the mask looked like on him while moving.
Next i combined the alphas using a merge and this was the result of that. I plan later on to do some more experiments with just refining the roto, feathering in places and getting both the feet and hands, but as this was mostly just to experiment with how Nuke’s roto works, for now i’m content with it.
We started this week by attempting to use Mudbox to create some proper bricks for my fireplace. To start i went about making a proper displacement map out of an existing brick texture in Photoshop. I did this by adding a black and white filter, and then adjusting colour levels to try and make this texture as black and white as possible, meaning making the divets in the bricks as black as they can be and the bricks being as white as possible with little grey inbetween to get a more defined bump map.
After i’d used things like photo filters and literally painting out some of the bricks into white, i went into Mudbox and made a simple plane to test out the brick texture on. I first increased the subdivision level on the plane to get more detail on the bricks, and then used the bumpmap i had created as a stencil overtop of my plane. I then used the sculpt tool and started painting the stencil onto the plane, tweaking the strength on the sculpt brush to a higher sensitivity occasionally.
The bricks weren’t as defined as i had originally wanted, so i went about going back into photoshop and a combination of levels, curves and a gaussian blur to remove a lot of the noise and grey little pieces inbetween.
I then made another plane and repeated the same process, making a new stencil with the new bump map.
This was looking a lot better compared to what i had previously and was something i was a lot happier with. I then scaled it a bit more on the Y Axis to make the bricks more defined with the divets and liked the result a lot more.
I Then went about trying to extract a displacement map through mudbox’s extract texture map, though unfortunately, Mudbox’s 2022 version absolutely refused to work in terms of exporting this, meaning this now stands more of a proof of concept with mudbox, as i’ll have to use Maya to do my bump mapping.
After this i went about using a simple cube and using the Duplicate special, and spaced them apart and created a load of identical copies of the cube in a row.
After this i sculpted a couple of them to look not so linear and made one in particular higher then the others in the middle as i wanted this to be the keystone of the fireplace, then i used a nonlinear deformer and bended the cubes into an arc. I also built a very rudimentary fireplace interior simply using the extrude tools.
After this i was having some problems with the UV’s on the interior of the fireplace, this was due to some of the way it had been modeled as i needed to use a combination of multi cut tools and adding divisions as the mesh wasn’t symmetrical, and then smoothing the UV’s using the smooth UV tools.
After this is textured the keystones, using the sculpt tool to also make them look more non uniform and adding a high bump value. I also textured the actual floor and interior of the fireplace, I also used a high bump value on this to give these a lot more geometry. I also extruded out some parts to make the fireplace look a bit more blocky on the hearth.
This fireplace was looking a lot better now.
However a big empty fireplace was looking a little bit empty, i remembered how i’d tried to build out some logs for a log cabin earlier on, so i repurposed that and used it to put some logs into the scene, making them have a high bumpmap for geometry and using the sculpt tools to make them look a bit different from eachother.
I then wanted to populate my scene with a couple other things, like a couple wineglasses, so i used the same skills i had used for the bottle but did the same thing in making the spline of a wine glass.
I also took the time to build a book, using the extrusion tools and adding edge loops along the way.
The glass was textured fine as it was simply a default glass texture, however the book still needed to be textured properly. I sorted out the UV’s to a decent amount and went about separating out the textures, i originally had some issues with this due to some small bugs but managed to work it out and put some placeholder textures on just to make sure it would work.
After this i added in some better textures. I then went through and made some variations in photoshops, and created two better bumpmaps for the pages and leather book casing.
I then started adding all of these things into my scene, attempting to make the scene more lively and full of items.
An issue that i had for a long time was attempting to make my lantern cast light properly, as the glass object encasing it was preventing light from coming through even with transmission on, i originally tried to fix this with multiple different ways, such as tweaking transmission, specularity, intensity and exposure of the light, and finally i found some solutions. I set the texture to turn shadow casting off and it was finally going through the glass.
The problem with this was that it didn’t cast shadows on anything at all, meaning the light was shining through objects that it wasn’t supposed to. I fixed this by turning cast shadows back on and then going into the attribute editor spread sheet and turning off cast shadows on the actual object instead of the light. this then made the light look a lot better.
I then went about positioning the objects around and tweaking the lantern a bit more and then rendered out some still that i’m much more happy with.
I wanted to make a mounted deer head for my rustic cabin scene as to experiment around with ZBrush. I’d seen that the smoothing and sculpting tools were easy to approach in ZBrush and could make some highly detailed things but when making things before i’d started off with a basic human head shape. I Decided to start off with a basic reference photo from google images and make a geometric low poly deer head.
After this it was just a simple task of adding in edge loops with symmetry and pulling apart polygons until i had a very rough shape of what it was going to be. I did this in an orthographic view port facing head on just to pull apart polygons as i found it easier to do then using a proper spline setup.
After this it was looking fairly good in an orthographic viewport but it had no depth to it on the Z Axis. Using just the perspective view and the soft brush i went about pulling out vertices on the Z Axis and giving it more depth.
It started coming along with adding more edge loops into parts like the horns and pulling out the edges very quickly.
After this the low poly head was done, i got both versions of what it would look like subdivided and polygonal and took it into ZBrush afterwards.
In ZBrush i went about both using the DemStandard cut tools and the move tool to pull apart and smooth the mesh down to get some results that i liked. I focused on deforming both the horns to look a little more organic then before and tweaking the eyes a lot. A lot of the difficulty i found with doing this was finding out the organic shape of what a deer’s head looks like with muscular structure, as all deer have fur covering the skin, and had to deduce through a lot of refrence images what it would look like.
Overall i’m happy with the way it looks now aside from the hole in the mesh that i’ll need to sort out, and may just add some more polys to flesh it out a bit more. Afterwards i’m not concerned on adding texture to the skin as this will be covered with fur and i may do so using Maya’s Bifrost engine.
After working with my professor Nick later on, i managed to fix this small hole in the bottom left of the mesh by closing up the hole using the move tools as much as i could, and then remeshing the subtool. I made sure to go inbetween undoing and redoing this to get the hole closed while still trying to keep the mesh as detailed as possible, as remeshing at a high level destroyed all detail. I let the project lay dormant after a while as i had some other things i needed doing, and then came back into the project after revisiting some of the ZBrush work from Week 5. I used the polypaint to make a very simple colour outline, and then decided to record the rest, using a combination of different colours in painting, smoothing the colours together, adding in textures via alphas, going back and subdiving the mesh more to get better detail on the alphas, and overall making the deer look much better.
Below is the process of doing this at over 2500x speed.
After this the deer was finshed, after being subdivided at around 5 million polys. i took this into Maya after using the UV Master tools to create a UV and making a displacement map / texture, after adjusting the bump on the displacement map to not explode all over the place, i realized something was wrong with the actual mesh.
I had some theories as to what was happening, but most of them stemmed from when i began subdividing in the process of creating this in ZBrush, as after remeshing my mesh had lost a lot of polys and when adding in alpha detail i had to divide again. I believe this broke the mesh in a weird way in ZBrush, as after consulting my prof Nick and getting his gracious help, The problem was in fact that ZBrush had broken up the mesh when exporting, which would cause major issues when applying textures and displacement. This was solved by turning off textures and displacements, grouping together all subtools, and then checking using the grid feature inZBrush to make sure the mesh was all in one object. I eventually used this exact process again when creating my Nordic Axe to make certain it would transfer properly into Maya. I imported the Deer into Maya and fixed the bump map once again blowing up even though it looked fairly neat.
After this it was finally done. I imported it into my scene and rendered it out. Overall i learned a fair amount i feel, and got a bit of a better feel about what i should and shouldn’t do when using ZBrush in terms of the workflow and also helped me learn to troubleshoot some issues like things not grouping or holes in a mesh.
I also built a very small plaque for the head to rest on in the scene out of wood in the shape of a pentagon simply using some edge loops and moving them around, though as this was in a dimly lit area in the scene for my room its not as visible
The Blair Witch Project was a movie in 1999 that challenged a lot of the conceptions of what a movie of the time should be. The movie is based off the premise of a camera crew going into the woods to film a documentary about the legends of a witch in the nearby woods. This adds an element of the film being very Meta and feeling more real as the cameras, audio equipment and crew are all referenced in the story of the movie. The movie seemed so real in fact, that in a quote from The guardian: https://www.theguardian.com/culture/2018/may/21/how-we-made-the-blair-witch-project
The co-director of the film, Daniel Myrick stated that: “We were listed as deceased on IMDb. Our parents started getting condolence calls”
The film consisted of only 3 cast members and cost around 26,000 pounds to shoot.
At first, the film doesn’t appear to have a very concise beginning, middle and end. The end is very abrupt and the climax at first appears to happen within 5 minutes of the ending. This works to the movie’s effect of realism, as in real life, things can proceed or go down hill quickly, but as a movie it’s somewhat different from the traditional Three act structure seen in most movies.
Exposition: The Exposition for The Blair Witch Project is longer then most films, with the conflict not starting until much later on in the film. The exposition starts with introducing us to the main characters and giving us a bit about each of their personalities through their interactions with each other, and then giving us small hints about what or who the Blair witch might be through interviews with local residents.
Conflict: The conflict in the film occurs when after filming shots over a day in the woods, the group finds themselves lost on the second day and can’t find their way back, attempting to stick to a map and compass to find their way home.
Rising Action: The rising action in this film takes up the longest portion of the film, with the climax, falling action and resolution only taking up around 15 minutes at the end of the movie. The rising action consists of the characters fights between each other, stressing over being lost, and having an unidentified entity consistently antagonize the characters night after night. This leads to a lot of smaller elements like losing the map to get out, interpersonal fights between characters, and revisiting the same areas or finding weird symbols on trees, all being separate conflicts that come together as one to quickly deteriorate the character’s mental states.
Climax: The climax of this film happens very quickly over a short period of time and its slightly difficult where to determine where the climax actually is. One of the characters is taken in the night, and throughout the next day the remaining two characters attempt to find him with no success. The small action portions in this movie occur over night, leaving the days to feel like a reprise, with this day being the shortest out of all others and not letting the viewer feel much relief at all from the previous night.
Falling Action: The falling action for this movie comes with both the viewer and character’s slow realization that the group most likely isn’t going to make it out of this situation. Over the course of around 5 minutes, the characters find an abandoned house while hearing their friend call out to them and scout around to try and look for their friend. It could be perceived that this is also a continuation of the climax, with the falling action not being a main emphasis and the story ending abruptly.
Resolution: The resolution of this film ends with the implied death of the characters on screen. With this being made to be a handheld or found footage style film, the audience only sees what the character sees as long as the character is filming. This leads to intentional gaps in the viewers knowledge when the characters are not filming, leaving the audience to piece together or interpret what has happened or other story elements throughout the film.
Overall: This film is difficult to analyse in a typical 3 act structure as by design, it is made to have portions that are very quick and others that are very long. The beginning of the film with both the exposition and conflict follows the typical three act structure, but the rising action and climax go far past the 2/4rd portions of the runtime, with both the climax, falling action and resolution all happening so quickly that it’s difficult to place exactly where the climax starts and where the rising action does. This does make the story feel rushed towards the end or even chaotic, but the intention was to make this story feel real with chaotic quick segments making it feel all the more real.
Firstly, i noticed in my room build that there were some issues with the walls, as this was made to look like a log style cabin, the logs are meant to be lined up and roughly the same size, though one of the walls was not the same. I used the viewer tools to disable the grid in the viewer, then went into the UV Editor and lined it up by moving it and scaling around more.
I went about making some tweaking to the lighting of my scene, as i needed to start making a couple changes as the scene seemed much to bright. I used the lofting techniques similar to the revolve technique used for the glass. i used the CV Curve tool to build a short wavy spline to look like a curtain, then used duplicated this twice and pulled them down to use for the loft.
After this i used the Loft function on the splines and set it to use control points and output polygons.
After this i applied a brown fabric style texture and started and added a low amount of transmission allowing light to pass through similar to glass. I went about tweaking the curtain’s control points as it was looking really blocky originally, this was due to some of the control points overlapping so i went about going through these points and separating them. I also used the soft selection tool to move some of the control points around and make the curtain look a bit more organic and like a fabric.
After this i put a side by side of the spline before and after i had fixed the blockiness and made it look a bit more organic,
Unfortunately i noticed that the top of the curtain was clipping into the top of the windowsill fairly hard, so i went about adjusting those control points to make it just overlay on top of the windowsill. I also saw that there were some holes in the above board mesh and that there was a lot of messy and really sharp geometry, so i used the smooth sculpting tool to fix it up a little bit.
At this point i really was hating the colour and design of this fabric, so i went about changing it to a more grey fabric texture, and adjusting both the transmission and the specularity, as a semi see through cloth shouldn’t have almost any reflectivity.
I liked these curtains a lot more, though i felt like a snowy log cabin might be better for the outside HDR Box, and may change the overall feel of the lighting.
I Experimented for a while with the lighting, however i realized using an HDR Of a snowy background wouldn’t have a whole lot of lighting difference and would be a lot more blue in terms of lighting and decided to stick with the original, though i tweaked the Lamp’s colour tempurature and the exposure levels of the HDR lighting.
I Then decided to add in some other things i’d built, such as a plate and the chairs i had built in week 2, i tweaked the bump mapping and textures on both and also added in the cheese i had built, then scaled everything properly and edited them in scene.
After this i tweaked the lighting for a couple hours as i was really finicky with it, experimenting with exposure and intensity, and eventually settled on this. Overall i’m happy with the end result.
Next i’d like to add more of the models i’ve previously worked on into the scene.
I had originally wanted to build a Chinese style lantern for my Room build in Maya, though there were a lot of changes in the design process that ended in me essentially throwing out a model, starting from scratch again and just doing a new lantern. I started with simply building out the base of a lantern, utilizing both the extrude tool and the symmetry tool. I had issues with this and had to restart a couple times as the symmetry wouldn’t co-operate due to the amount of faces not being equal, but eventually managed to get it working. I used these tools making sure to add divisions to the extrusions as i went along, and using the soft selection tool to get curves on the sides of the lantern as i went along. I also build a simple bulb in the middle by scaling and moving the verticies in a cylinder.
Eventually through using these techniques, i ended up with this Chinese style lantern. Unfortunately i hadn’t taken UV’s into account, and there were a lot of issues with the model such as they symmetry on certain things being wrong, the model being over complex, and not fitting in with the overall scene i was going for. I originally also thought i had lost the model, so i ended up just starting over from scratch, but used the techniques i was using prior and the mistakes i had made with symmetry especially to focus on making the new one much better.
I went with the same way of building this out, and came out with something that looked more akin to a regular lantern. I used a torus at the top for the ring, and the ring can be positioned upwards or downwards as I’ve set the pivot point at it’s bottom. I also built a glass encasement for the actual light so that the light could go through it. I did this by just adding a glass texture and making sure the transmission would work well with the light going through it.
Previously i had tried multiple ways of putting the light inside of the fixture, such as using a single point light which kind of worked but didn’t work too well as i wanted the light to come from a single point. I originally had wanted the light to come from a single ball or sphere but putting the light in it would immediately diffuse it as it had nowhere to go out from since it was in a sphere. This also didn’t make sense as for an old lantern it shouldn’t really have a ball lighting it, in terms of real life it didn’t make sense. I wanted a coil light and asked my professor, Nick for some help. I learned about using Mesh Lights; objects that acted as a light source, and made a mesh light out of a helix to make the lantern appear to have a lit coil inside, tweaking the helix to be long and thin along the way.
Along the way i made a texture to apply to the lantern as well, a cast iron texture i had simply found online, and gave the lantern a very light bump map to make it feel a bit more organic. I also tweaked the specularity to give it a bit more roughness and a cast iron feel.
After this as my previous blog post stated, i cleaned up some of the UV’s and did some tweaking to the atmosphere rendering and how it would interact with the mesh light, and this was the final result, which while i’m happy with it overall, it took a lot of different tweaking with the mesh light to get what i felt like to be the correct colour tempurature, intensity and diffusion.
Unfortunately once again my bread’s displacement bumpmap had just disappeared due to Maya’s weird ticks. I fixed this and then upped the bumpmap on the cast iron and was happy with the result.
I started building out the face with a Zbrush primitive, which is very polygonal when starting off and is going to need a lot of work to subdivide and make look like an actual human head.
I did some smoothing and started giving him some actual face shapes but unfortunately he looks a little bit like a goblin.
Afterwards i started tweaking the facial structure, attempting to add face lines and do some things with the eyes as opposed to just having some dented holes in his head and adjusting the blockiness on his neck.
After this i adjusted the eyes giving him eyelids and starting to add in some more facial structure feelings, as at this point it needed some more lines to feel more organic.
After this i got it to a point i liked more, with it having a lot more wrinkles and face creases and somewhat resembling the old man from squid game.
I managed to bring it into Maya and export it, but i’d like to go back at some point and add in some more textures with painting on the face and potentially tweak the brow line / add some facial pores.
I went about using the polypaint tool to then start adding some colour, switching the options on the brush to RGB instead of Zadd, i also made sure to turn on the fast shading in the textures to make sure i was getting the actual colour as opposed to the polyskin texture.
I Started adding colour to the model at this point, trying to use the smoothing tools and using the focal shift to make the paint on the face look somewhat blended together. I also then went about importing some alphas to use as textures on the skin, for the face pores, lips and skin etc.
I Then went about using the UV Master and making some UV’s for my face to be able to put it into Maya, also making a texture map from polypaint, and a displacement map to get the detail of the model when importing it into the scene.
The model imported fine and intact, meaning there were no issues with the mesh, so i went about adding in the displacement map first, turning off quadratic filtering and making sure the colour space was set to RAW for most detail.
However the color balance wasn’t set properly on the displacement shader, as it had no deference between colours and needed to be set with the offset being half the gain. I tweaked it again and accidentally put in an incorrect value and got the complete opposite of the first image that almost kinda looked like an alien.
After fixing this, the model was looking much better.
At this point all that was left was to simply add in the actual colour, and upon doing so, this was the result.
I narrowed down the issue to being that when creating a proper displacement map, i both had saved as a tif file which maya hated, and i also hadn’t set the map to adaptive. However upon trying to fix this, i had ZBrush crash constantly when trying to export. I believe this was due to the number of overall polys, as it was sitting at over 21 million, and even on the lowest subdivision level would still not export. I went about researching ways to get this down using decimation, however potentially due to hardware constraints, even this couldn’t fix the issue, as decimation master would repeatedly crash and even cause hardware issues where half of ZBrush would stop outright with errors of decimation master stopping. However, as i learned the techniques to get there and knew the issues, i decided to use these same exporting techniques for other projects like the deer head i was working on.
The room that i had built had some small things in it but needed to start to be populated with some items to look more lively, i started by making a simple green glass bottle for the scene. i Did this by instead of building it out meticulously through using a cylinder, hollowing it out etc, i used the CV Curves tool in the sideways window viewport to draw out half of a bottle in splines.
After this i centered the pivot point to where the bottles halves would meet by hitting D on the keyboard and snapping too the first control point.
Then i went into the revolve tool and made the revole axis preset set to Y and had it sweep at 360 degrees and set it to only 8 segments as this was a bottle and wouldn’t benefit from having a high number of polys. After that i simply applied it and came out with a bottle.
Unfortunately when i was making this Bottle it ended up having an issue with putting some control vertexes where they shouldn’t have been, as i was new to using the tool. This was creating some weird edges in the bottle that shouldn’t be there.
I started deleting the control points that were a problem and also moving around some control points to tweak the shape of the bottle, as the spline would directly corollate to what the bottle’s object would look like.
After this I Imported the bottle into my scene, scaled it down to size on my table build from earlier, then gave it a glass texture and set the transmission to green to give it that wine bottle look.
After this, the wine bottle was finished and rendered into the scene, and was working properly as a reflective green bottle.
While working on this, i started wanting to fix some things with my room build. For one there was no proper ceiling, and the floor and walls were extremely drab with the way they were presented. I originally had wanted to build my room as a log style cabin, though i went about this in potentially the worst and most convoluted way possible, specifically in building out cylinders and stacking them on eachother, then deforming each of them and adding separate wood textures to each of them.
This was a bad way of working as it would both up the poly count and bloat render times. Instead i simply went with making a log style by using a texture from polyhaven that was a wood walls texture that looked like logs, and making the bump strength on the displacement set very high. Previously i had extruded some of the alls and repurposed some of the wood assets to build out windowsills as well, though since this was a log cabin i wanted the wood from the log cabin to build into the wood from the windowsills so this overlapping wasnt as much of an issue. I also cleared some of the unused textures and also fixed up some of the UV’s on it.
While fixing the UV’s i had a fatal Maya bug in which i couldn’t scale textures at all. Previously i’d found ways to get around this error but i needed this solved to progress as the floor in the image needed to be scaled to look larger and add more segments to the tiling. I ended up using a way of troubleshooting where i simply reset Maya by renaming the Prefs in it’s userfiles to OldPrefs, forcing it to create a new preferences folder in the process and getting rid of the bug.
I also used this time to go through and change some things from what i’d previously done. For one the Lantern i had built prior had some pretty awful UV’s, so i went about fixing these by doing a combination of segmenting the UV’s into better shells and scaling them or just using Camera Based to fix some. UV’s are not my strong suit, so a lot of the time Camera Based UV’s can be a life saver for me or just fix UV’s that are having Bugs or Problems.
I also messed around a bit with making the atmosphere on my renders, specifically around the coil light have a bit more atmosphere to bounce light around on, so i experimented with making it a bit stronger.
I tweaked it a bit afterwards as now there was too much atmosphere and it simply looked cloudy, but while rendering this i noticed an issue, the bread i built being very smooth for no particular reason. I noticed that somewhere along the way, the bread’s displacement map had just failed to load in properly at all. I took this opportunity to fix this and also to add a slight bump map to the bread’s plate, and to add some roughness to the plate’s specularity as it was a bit too shiny.
After this, i re-rendered these things out and they looked much better.
Discussed file formats and their relevance in Nuke, with EXR being the most prevalent file formats to use in Nuke, that being mp4’s or EXR’s and the software used in compositing, with After Effects and Nuke being the main programs while also having davinci resolve fusion as an option.
We also talked about the digital production pipeline and the main lines of pre production, production and post production and a short overview of the roles done inbetween.
I’ve previously used Nuke before so relearning most of the controls were a bit of a refresher course for me, but after around a year of last use it was definitely a much needed refresher.
Starting off with the refresher was simply just hitting Tab and importing a file through the Read node, and toggling the viewer to see the footage using the 1 Key.
Afterwards was using both the gain and Gamma sliders to effect the video in the top right. I’d actually not properly used these before as i’d never really had a reason too or would usually add a different node to do so, but having these around is nice and may save time from crating and plugging in new nodes.
Next was testing out and applying some blur nodes and keyframing them to create a Gaussian blur into frame. I’ve previously done this in after effects a lot of times, but a lot of my experience in Nuke came from Tracking using CaraVR, 3D Integration using merge nodes and 360 Stitching, so something as simple as doing a blur node felt fresh to me again.
Next i was taught about just rearranging the panels in Nuke to fit your layout. I’ve had many times in Nuke prior of accidentally messing up the layout and removing a pane that i shouldn’t have, so being able to finally edit this is one of the greatest strengths i could probably learn. Layout and knowing your interface can truly make a difference between working both quickly and efficiently.
Next was creating a simple composition using the merge tool and some grading / transform nodes. Back when i previously had worked in flame, i’d had it really drilled into my head to keep my node graphs very meticulously organized, which is something i want to keep doing going forward using Nuke.
After this it was just a simple thing of using the write node again and exporting this composition to a jpeg after fixing up the composition a little more.
And our little test composition was now finshed!
With this proof of concept i decided i wanted to test out the color grading skills of Nuke. I Personally use Davinci Resolve for all of my Compositions and have some footage that i previously color graded in Davinci, though it was graded incorrectly on a monitor that had very low contrast. I no longer have the raw footage, but to get a little refresher on both Nuke and Davinci, i decided to import different shots and see how it fared on color correcting compared to eachother.
First of all, these videos are both way too large and also as MOV, which are probably not going to play well in Nuke, With some of these videos clocking in at nearly half a gig, these definitely need to be compressed down.
These were converted to an H.264 Codec and the resolution was halved, with the bitrate being dropped to just around half of what it originally was. Now that the file size and codec has changed these should play well in Nuke and Davinci.
While attempting to find the Histogram in Nuke to start compositing i wanted to use the color correct node but accidentally put in a deep color correct node. I hadn’t seen this before and looked on the foundry website here: DeepColorCorrect (foundry.com) to attempt to find the difference. It seems as if deep color correct is mainly used for getting a matte, so for this time i’ll stick with using a regular color correct node.
I found a way on the foundry site to use histogram both as a panel to view in real time, and a node to use. Though the node doesn’t seem to have too much use when editing the way i’m doing as it doesn’t update in real time and only updates after switching to the node after every edit with the color correct node.
Originally the video had a lot of issues with the highlights being very oversaturated and the reds being blasted to the point where the video looked very yellow and green tinted in some places. I fixed these by using 3 different nodes, being the hue correct to bring out more of the reds, a color correct node to tone down the highlights and up the shadows, and a grade node as a master correction node to adjust gain and lift if needed.
This was the difference between the videos, the first one being the color corrected: and the second being the original version.
Next was to use the same clip in Davinci and see if i could get the same result or better. I used initial primaries adjusting the Lift, Gamma Gain and Offset, then a secondary to try and bring back some of the red tones while toning down the yellow tones. This originally didn’t work as well and i had to almost brute force remove some of the color using the qualifier tool, with some more refining i feel like i could have gotten it a bit more red, but it started to look close enough to what i was going for.
The end result was a little bit red blasted in some of the highlights but overall achieved the comparison i was going for.
At the end of the day these are two very separate programs. However i learned a lot more about Nuke and it’s ways of color correcting from this small process. In terms of which is better for color correction, i think Davinci still ends up coming out on top when it comes to specifically making very detailed corrections with its easy qualifier tools, access to many LUTS and power windows. While i feel Nuke being mainly used for compositing can still provide some very quick and easy color correction.
I started this tap build by simply using a cylinder and decreasing the subdivision count then using the edge selection tool and deleting some faces to make the shape hollow, then using the extrusion tool to build out the shape of the faucet attachment to the wall.
After this i went about using the extrude tool, building out the faucet and adding subdivisions as i went along, trying to round out the faucet in the middle and then create a dipped neck, using the soft selection to dip it down and rotate it properly.
After this I realized i had made a mistake, as i originally had built out the faucet all as one object, and this was going to create major issues in terms of UV’s and texturing the object, as well as this, there was some weird topology issues going on with the faucet’s end that needed to be fixed. I went about deleting some faces and building out the faucet’s top using other pieces of geometry to have more control over the textures.
Next i went up the actual faucet head, building as i went along with different shapes, building segments and using both the extrude tool and symmetry to build out things like pipes and nuts for the head of the faucet, then using both the extrude, scale and symmetry tools to build the head. With the UV’s on the head, originally the UV’s looked fairly ugly, but this was fixed by using a UV Based camera and adding seams in some places where it wouldn’t be seen easily.
After this i went about texturing the object. Originally in viewport it would simply look like this, as reflective textures do this in the maya viewport without lighting. As this isn’t really beneficial for use in actually working out the texture, i used a combination of viewing the textures in the live render viewport and using the hypershade to tweak the textures.
Originally the texture that was supposed to be applying wasn’t linked in properly so after linking this in it looked a lot better. I also tweaked the bump map a bit to make the cuts in the metal look a bit more realistic, tweaked the specularity roughness and added a colour correct node to change the colour of the metalic texture overlayed to slightly tweak it a bit and give it a bit of a darker colour.
After this the tap was looking a lot better, and i went about implementing it into my room scene.
While implementing this into my scene, i really didn’t like the way my sink was looking, as originally it just had a plain white texture. I decided to check the UV’s to see how easy it would be to go about doing some things with the texutre as i wanted to make it have a black marble ish basin. The UV’s were a mess as this was originally extruded a lot without thought for the UVs, so i went about cleaning them up a bit, either by the grace of Maya being finicky or a bug in Maya 2019, Maya refused to let me move my UV’s around properly, however i went about cutting the UV’s and sectioning it off into shells, then using Camera based UV’s on those shells to make the Textures look a little bit better. As this basin’s textures were both marble patterns and also very dark, a lot of the cuts in the UV’s wouldn’t look as noticeable, and i’d be able to add in a new texture.
I then added a small bump map to the texture, reduced the specularity a little and added a bit of roughness to the specularity and finished up the basin and taps. I may come back at some point and just slightly drop the bump map just a little more as i went back and forth between adding and removing it, but i may just tweak it a bit later in some renders to make it a little bit less bumpy on the edges.
