800 bits - 27 days later...

I've made it 27 days so far - and I have to say I impress even myself in just how many small CG projects I've managed to pull out of my hat so far.  I thought its worth a short catch up on some of the things that I've done/used/etc to get me this far.

Before I do, I felt its worth also mentioning a fairly amusing comment I got from someone who spotted my work online.  They thought I'd been posting up photographs - and said that I should really make them look 'less real and more 3D' otherwise nobody would realise I was creating 3D images...

Its not often I will hear someone tell me that they think it looks too real and ask me to make it look more computer generated.  Too funny, but I know where they were coming from on an artistic direction.

"This is too real... Make it less real so its obvious that its fake"

Its all in the software

For this challenge I decided to use LightWave rather then Maya.  Originally I thought it could be a great challenge to pull this off with Maya (having taught Maya exclusively for the last 4 or so years) but knowing the amount of time that some areas of Maya's workflow take was enough to warrant going back to my roots (as it were).

There are also some tools and features that are just not found in Maya, and these would allow me to get what I wanted at the quality I wanted with the tight time frames of a daily project.

What took the most time?

In the last few images, a series of isometric-styled computer hardware renders, production time and rendering can be done in 1 to 1.5 hours.  That's because they are comprised of simplified iconic models with basic colour textures.

However for the first 20 or so, an average of 3 hours was normal with some of the more complex ones taking up to 5.  These I attempted to go for a more complex realistic visual, which is why the time was so vastly different to the more recent projects.

In general, my gut feeling is usually 10% modeling and 90% everything else.  In this case, modeling objects was more around 15%.  Where the most effort seemed to go was on the composition, lighting and rendering.  Surfacing didn't take as long (with fairly low amount of Photoshop effort required - but more about that later) for most things, however the amount of additional tweaking I would do while rendering was where this would increase.

Hmmm, deciding on an angle...  Tricky...

It takes a surprisingly long time to get that shot just right, and often the shot I had originally envisaged didn't look as great on screen once it was rendered - which meant a lot of adjusting and testing.  However this is the area where LightWave really shined...

Don't over-model...

The tip of the day - just model what you see in shot.  For instance in the first image I produced of a ZX81 I never modeled any of the sockets or insets of the machine because they weren't seen - and I simply positioned the cables to sit in the correct locations.

Another example here is a BBC computer where the image rendered was of the BBC logo in the top-right corner.  Did I need to model every key?  Nup...

Model what's seen, not what you think you might see...

Bumps, bumps, bumps...

A lot of 80's computers were built with plastics, and as we know, not all plastic surfaces are smooth and shiny.  This is where adding bumps to the surface helps break up the surface and simulates the appearance of these types of materials - and I can honestly say that I did this with literally every project (given just how much plastic was involved obviously)

It may be tiny, but its essential to getting the look

As I said, this wasn't as time consuming as the cinematographic aspect of each project.  One reason being that I made use of a lot of procedurals for adding extra texture to plastics and other types of surfaces through their bump channels.

Bumps, bumps and more bumps...

I also took advantage of LightWave's gradients to assist in controlling some of these to behave the way I needed them.

Using a slope-driven gradient let me produce smooth edged keys on the BBC

Cameras

To pull off some shots, many of the scenes were fudged.  That is, elements are moved, scaled and rotated to suit the shot and assist in making effects like fogs and depth of field behave the way that you need them to.  Its pretty common in CG to 'work to the camera' with anything - its often the only way to pull off many shots.

Tiny tapes, HUGE TV set, fish eye lens and fog to haze it out in the distance

I also took advantage of a Photoreal camera type that allowed me to apply a lens profile to the render to generate a true fish-eye effect.  While that effect can be faked with Photoshop, you won't get quite the same result as a camera that sees the scene around it by processing the render through a lens.

Depth of field can also be viewed in real time (and very quickly) in the viewport, and that assisted a lot in fine tuning the final effect.

DOF preview in real time helped a LOT in getting the shot

Themes

I knew I wanted interesting interesting ways to display items, and while camera angles and various rendering styles are one thing, I figured I'd also try playing with how to interpret the ideas.  One in particular was when I wanted to produce something to represent the arcade game Dig Dug.  Originally I'd thought all the video game related artwork would be sprites built from 3D cubes (which can look cool) but in this instance I decided to try something different.

We'd been talking at work about post-it artwork just the day before this piece was done. In the last building we were, I'd drawn up some Defender sprites on the windows with post-it notes...  And I figured why not replicate that same idea in 3D.  Then I thought a glass window or office may be a little ambitious - so I went with what we do at work often - producing display boards and hanging the work from thin nylon...  Well, virtually that is.

Virtual art exhibition, anybody?

This was built from about 5 variations of post-it note (ie. a divided plane and a variety of bends) model and by placing them a little imperfectly with some random rotation I managed to get something that looks a little more human.

Lighting and Rendering

Along with surfacing, one of the key features that LightWave has is its Viewport Rendering - or VPR for short.  This feature alone is worth its weight in gold, providing real-time rendered feedback as you adjust and move elements about in shot.  When I was surfacing I used this a lot to see the result of tweaks to bump maps, reflections and node-based materials.  As its a render, it also allowed me to preview my fish-eye camera, as well as depth of field and adjust as I needed.

New Style

All that photo-real style work does take a lot of effort in modeling, surfacing and rendering.  So, I decided after 20 days had passed I would swap my style and try something new and different to keep the work interesting...  I also wanted to see if I could make things faster to create.

A whole new style

For the next set of images, I went for an isometric-angled simplified design - creating machines I recall from my youth (and mostly the ones I wanted, not the ones I owned) in their iconic form and giving them a simple shading style.

Advantages here - I didn't have to worry about fine detail modeling, as I was after the shape and form of the machine and not an accurate model.  Keyboards could be simple blocks, icons and branding didn't have to be added to the model...  Though I decided to model the branding and place it next to the machine.

Taking a break from detail modeling and rendering for a while

The shading was all done by driving the surface luminosity (or incandescence in other applications) through a mix of occlusion shader and a gradient ramp driven by the surface slope to allow shading of curved surfaces.  The floor was a flat plane and simply used a radial gradient to create a 'spot light' style effect from the center.

That meant no need for lighting, calculating shadows - which then made render times a little faster to boot.

So yeh...

There's still 70+ days to go - its a fun project, though at times it feels like it overwhelms everything else around me when I know I have to complete something that night.  With work, time is tight so I squeeze in things when the day is done and I'm sitting in the office...

I'll do another update when I make any new changes in direction...  And obviously a post when I'm complete!  Feel free to check out the project so far HERE.

Continue

Day 1 done - 800 bits - the challenge is officially on!

Well, day one has ended.  However I broke my own number one rule - not to be too ambitious and try to keep things simple and achievable.  It started out great, but then it managed to slide down into my usual 'perfectionism' trap, followed by a mind that keeps coming up with more ideas and going off track very quickly.

Thankfully I put the first image up - however with so many ideas I want to create, I let the first one become a huge daunting task that went haywire at the start.  So as day one finishes, its time to self-reflect and enforce some rules for myself so I don't get caught out again.  Essentially - its time to start breaking some habits when it comes to my own personal projects...

Reference and planning

I did actually do a lot of initial brainstorming and research...  I felt I was fully prepared, had everything I needed at least churning away in my cranium...

I did a handful of thumbnails - I even have an overall plan for some structured shot layouts when it comes to a few.  I finished the 3D model - great, so it should have just spun off from there and been done in just a couple of hours...

BUT...

I went and started to refine details, after which I modelled a few other items in for the next few images.  I had an idea in my head for the composition, but of course next thing I know, I'm deciding that I want to make something else because my brain is seeing possibilities - a dangerous way to work.

Generally, here's what happened...

What I did with the first image was to  

(a) decide to start with my first computer (that was the hardest part - truthfully!).

Then...

(b) I did a quick grab of reference images I've been collating and built a 3D model.

This was actually the easy (and quick) part of the project

That was actually real easy (heres a timelapse video showing the modeling process).  I then...

(c) collated a lot of items from other projects (my ZX Spectrum projects - Tapes, cables, cassette cases)

and built new textures for them based on classic ZX81 games I had back then.  Where time started to get wasted was when I...

(d) started to fumble about with my composition.

I moved cassette models, I tweaked the camera, I placed lights, I tested a few things...  It just became a case of fiddling way too much.  However I opted on two camera angles...  I rendered them out.

Finally...

(e) I did some post grading on them

... with a little lens distortion, abboration, vignette and grain - plus a dash of subtle sepiatone and compiled the images into my bubblegum card template.  Upload - and I was done...

until I...

(f) discovered I'd not paid attention to a few cable details

...which I wouldn't have had to worry about if I'd just done the machine on its own like I'd originally planned, and quickly adjusted my 3D scene, re-render and update the one online.

KISS

One of the challenges with CG is of course to meet deadlines, often under a lot of pressure.  Time pressure will be my daily job during the week, which in itself can take up almost 12 hours of my day when I include travel to and from home.  That's going to put a lot of stress on meeting that daily deadline.

Part of the plan is to definitely "keep it simple, stupid".  Composition will play a big part in this, and I've decided that I can definitely focus on key details that don't require me to model a vastly complex mesh.  As much as I'd love to build a fully detailed 3D object that I can render from any and all angles (such as this ZX Spectrum) I know just how long a project like that takes.

Keeping my compositions focused on the key subject matter using effects like Depth of field and building a themed layout will let me lower the complexity and hopefully retain the visual appearance I'm after.  That's just going to have to be how it is from now on - come up with some pre-prod at the start of the day just to clarify an idea and then stick to it.  There's no more first-day headache of changing direction as I go...

It doesn't actually take THAT long

I built my first new asset for day 1 - my very first home computer, the Sinclair ZX81. It was less then 40 minutes from start to end.  I limited the details down on the model - based on the composition not seeing them.  That shaved a lot of time down.

Textures are also mostly colours and bumps.  That in itself shaved plenty of time not having to manually paint up image maps.  (Here's that timelapse link again, as well as this lovely 22 second timelapse of the little work it took to surface the model (its so short, but it indicates just how easy and quick the process was))

Round, eh, I mean day 2 and onwards...

Day 2 I'm going to continue with this initial history.  I'll make use of models I've already built in the past rather then reinvent the wheel as it were.  Its going to be a test and a half, but there's 99 days to get good at it.

My process from here on...

I figured its worth noting a little about the process I've decided to go with here. 
Obviously I can't muck around like this every time I make an image.  So the general process from this day on will be

(a) Make sure once I'd done my pre-production thumbnails... I STICK with the one I plan to do and do not deviate.
(b) Reuse what I can from my own personal library (ie. my work, I'm not one to use free 'stock' off the net)
(c) Model anything new based on composition
(d) Render, grade and slap it together.
(e) Upload, and then prep up for the next one...

Its part (a) that's the clincher really.  We'll see how we go for tomorrows challenge...

Continue

More CG retro goodness...

In our 3rd year 3D diploma, our students first assessment is to create a realistic rendered scene from photographic reference.  They've just been learning Mari, ZBrush, Renderman and Nuke - all tools that they utilise to produce the final image itself.

As always, I tend to find that there's not a lot of things I can do when these roll around.  The 3rd year is very self-directed, and by this level they should be very competent with their tools and require just a little guidance creatively now and again.  So - with my mid-life crisis still burning away at my soul, I decided to just do my own mini project with the inspiration again coming from my obsession for the 80's computer era.

Recent Acquisitions

If you saw my other post, I recently refurbished a C64 breadbin case.  I'd taken a few photos for that blog post, and I figured I'd use one as reference to create a small scene.  I was also inspired by other things around me, but more about that as we go...

Starting with something simple as reference

Getting started

Obviously the first rule is to not do more then you need to when working on a project.  For the C64, I created the visible corner rather then the whole machine in this case (no pun intended).  There were some background elements - a screw driver, a tablecloth and flat white paper.  I figured that I may toy with these a little and take some artistic license...

The screwdriver luckily was pretty simple.  I modeled the attachment for the end, but only the main shape rather then indent the head detail.  I expect this detail to be out of focus anyway, so there was no need to produce a high detail model.

Materials and textures

Textures in Maya aren't hard, but things such as the procedural textures Maya offers I've never been a big fan of - and I wanted to make use of a fine fractal to generate the molded plastic.  Maya just doesn't make it easy to achieve this quickly, creating odd skewing and weird mapping issues...

So I resorted to the approach of 'best tool for the job'.  One of the students also needed a fractal texture to apply to the bump of a plastic case as well...  Using Lightwave's 'texture image filter' plugin, I created large 8k images of small, clean fractal noise that we could use to add textured detail to the plastic.

I was inspired by work going on around my desk in the office, where air-con repair created a lot of plaster dust as ceiling tiles were moved...  That meant I also needed some dust along edges of the case grooves, and some more fractal variations for breaking up the surfacing subtly...  Using Lightwave's Surface baking camera and the UV map I'd generated, I baked out a few more textures I could use in Photoshop to paint up other maps.

Put a model in the oven, bake on high for 30 seconds...

Crud...

Another detail I wanted to add to my dusty concept was of course larger particulate matter.  Again, I made use of Lightwave's vertex based particle emitter to spray and randomly spread particles into grooves, edges of the case for placement of 'solids'.  Baking the final frame out to an object layer, I then used them to randomly clone a 'plaster particle' with variations onto each vertex.

When bought back into Maya, these aligned perfectly.

But what of that background?

I had a piece of black art-board sitting on my desk that was collecting dust from all the dust being generated...  That gave me the idea to use the board as part of my background...  Plus a bonus was that with a little wiping and shaking, I could slap the real art-board on a scanner to pull off realistic textures that I also used to add plaster streaks to the computers case as well.

A few small screws that I modeled and placed on this board added to the story behind the image.  The screws are perhaps a little higher detailed then they needed to be, but they were relatively quick to create (for this I used LightWave and its very useful lathe tool to create the thread)

Completely screwed...

The tablecloth - also using Lightwave (its nice to be able to work around multiple applications) - I created a cloth simulation for a subdivided mesh, then pushed the board geometry on it to wrinkle the cloth close to the edges of the board.  That I also baked out as a mesh for modification

I took that into ZBrush for touching up, and then textured it with 2 layered renderman shaders to simulate the tablecloth which had a shinier threaded pattern through it. I drove the mix of the two materials through a tiling damask pattern.

Setting up some of the background details

Finishing off

As I had done previously with the C64 dust particles, I repeated the same process with the cloth and art-board to add more details...

Background dusted up and ready for render

The same dusty smeared textures from the art-board I also threw onto the table cloth and C64 case to detail it more, then a quick adjustment of camera angles, some very simple lighting (primarily a large area light and a simple HDR environment) and DOF produced the finished image

Done and dusted as they say... Eh, I mean dusty

To be honest, I'm not 100% happy with it - it could do with work with the lighting - but its a project I enjoyed working through.  The benefits of doing something like this presents the occasional challenge, and its those challenges that encourage more learning of the tools and technology.

Continue

"Great scott! Its time to travel back to 2014"

There has to be a point where a project finishes - in the case of the ongoing 8-bit project I've worked on for a while now (not constantly, mind you - just a little here and there when I had a moment to spare), I've decided to toy with the camera, lighting and do a final rendered still.


(You can view all my posts on this project here)

Just adding a little dutch bank to the camera, a few texture tweaks...   I ended up with this image.  A classic moment - playing Manic Miner in a dark room, with the sunny day outside being blocked by the bedroom curtains.  Just like it was back in the 1980's...

Done...

Some subtle comping of reflections/spec passes back on top of the image, and then grading the colour and exposure a little (all passes were rendered out as half float in .exr format) after slapping in a bit of lens distortion, glare from the curtains and noise grain...  I'm fairly happy with this.

Mix-n-match

Getting the angle right...

As for the camera - I wasn't really sure just where I was going to go with it.  In my original "finished" render, the camera was pretty much straight on.  It looked ok, but it felt a little too dull and uninteresting.  I decided to just throw on a Golden Spiral guide and used the inward spiral to take the viewer towards the focus of the game itself.

Composition guides can be useful tools

Compared with the last camera angle, I find this much more appealing.  Of course, beauty is always in the eye of the beholder as they say.

Tweaked a few elements...

I did tweak a few small things in here.  Some positioning of elements for one - the data recorder I actually moved forward (and slightly scaled up) to subtly influence the perspective of the shot.  The screen, while its not obvious, has a very fine dark band that dulls the pixels (across that creature/green platform) to create the effect often common with the scan line on TV's when filmed or photographed.

In this image below, I've cranked the effect up to make it more obvious - It was simply a soft black block blended with an Overlay to deepen the image...

The effect - though more exaggerated in this example then the final version

I looked at a few photo's and old videos of CRT sets from this era and found they varied - some were very dense and dark.  Some, as in this case, were very slight - but enough to 'just' notice when it slowly moved down the screen during filming.

I'm done and dusted... Honest... Well...

So, for now I'm signing this project off.  In reality, I'm more then likely to return to this now and again and do more "fiddling" but til then its time to consider new projects.  Such as looking at the real thing - given my brand new stash I recently picked up (as seen below).

Keep with the virtual?  Perhaps just go for the Physical...

For those curious, the large keyboard you see at the bottom is from an old Commodore PET - the person I picked these up from has rewired it to work with the ZX Spectrum...  How - I don't know (I'm no electronics boffin) but its pretty cool...

Until next post...

Continue

Swedish Space Odyssey... Kinda...

Back in June 2012, I was asked if I would be interested in working on some VFX for a possible short film a friend-of-a-friend was pitching.  It sounded like fun - with a little space action as well thrown in, I was more then interested.

As part of the pitch, to get funding (which was relatively low) it was important to show that producing Space effects were going to be achievable.  In this case, part of the story revolved around astronauts in a damaged old space capsule - so a concept image showing the ability to produce reasonably good CG was needed to prove that point.

This is the image that I completed for the pitch.  Its based on a Russian Soyuz capsule, featuring a Swedish flag (since the director was Swedish) and orbiting an image of the earth.  An aspherical lens was simulated to create that slight optical distortion, and a lot of additional glare and detailing were finished in post (which you'll see as you read this blog article).

Ready for launch... Ahem, I mean pitch...

Given that the time frame was short to get the pitch submitted (as in a day or two - no pressure, honestly!)  I built and rendered a 3D Soyuz-styled capsule with basic textures.

Basic capsule and background plane

Basic textures were painted up

As you can see, I did start to model in quite a lot more detail then I needed for this single image (I had planned to build a reasonably production-ready high detail model).  There are also additional plates and details on the back of the capsule (based on some of the reference imagery I researched) that we don't see in the render...

All those bolts, etc - when seen on the render, perhaps a little too ambitious.

The problem is that to fully build a detailed capsule with all the textures and detailing in place would take a few days longer, so I finished up the final details manually in Photoshop. Tight deadlines mean that being overly ambitious is not a good idea at this early stage (and if the film went ahead, there would be plenty of time to build on this model anyway)

From basic render, to finished image (mostly done in PS)

Final tweaks were to touch up some of the lighting on edges of panels, add that slight diffuse glare to the bright spots of the image and throw on a few additional details, etc.  After a year or so, I can now see a load of things I should have done, but for what was needed at the time it looked the part to get the ball rolling...

Unfortunately the pitch didn't get accepted as a whole, but this small project - like most small projects I tend to get - was a lot of fun and a chance to create something I would have never thought about doing...

Continue

And even more free scripts... Ho! Ho! Ho!

If you've already seen the previous post, you'll have likely downloaded and looked at the collection of scripts I uploaded.  Here is a second zipped archive of more of them, along with the source for a handful of C plugins I also developed some years back (199x era)

Note that I'm releasing these as a free learning resource.  I'd like to think that anybody who downloads them will share updates or modifications back to the community.  All I ask is that you don't abuse this gesture by releasing 'similiar' tools based on things you picked up from the scripts here without credit or community sharing...

Continue

Free LW scripts - Just in time for Christmas!

While cleaning up one of my hard drives, I found a collection of some of my LScripts I developed over many years.  By all means, these are not everything I've developed and I will upload more as I find them, but I figured that there is bound to be pieces of code or ideas that others can use or learn from...

Note that I'm releasing these as a free learning resource.  I'd like to think that anybody who downloads them will share updates or modifications back to the community.  All I ask is that you don't abuse this gesture by releasing 'similiar' tools based on things you picked up from the scripts here without credit or community sharing...

Continue

When is a cow NOT a cow?

When its a flying machine for a Swedish childrens show called Ljudhjältarna...

 Late end of 2012, I was approached by film maker, director and editor/compositor Arvid Eriksson (whom I'd met through a colleague when creating a conceptual 3D space capsule image for an indie film pitch) to produce a 3D steam-punk styled airship for a Swedish childrens show.  The plan - to create a 3D model that he could then import into Adobe Aftereffects and animate in post.

Arvid's a very talented director, producer, editor and compositor - and he's also a lot of fun and infectiously enthusiastic.  Obviously, I just had to say yes - and hey, nothing says fun like a flying air ship that travels through outer-space... On top of that, the conceptual art we had been given had a stylised "cow" look to it - 4 hooves, rope tail and wings that formed the ears.  This should be a blast...

The finished model as used in the opening titles

Getting it started...

Based on the concept art, I produced a very low quality proxy model using LightWave3D's modeler.  Exported out as an obj, this meant the model could be imported into AE by Arvid to animate and get his project blocked out quickly.

Rough out proportions, and something for the director to get started quickly.

Taking it from there...

There were a few things to consider during this project when it came to building this model...

Obviously one was that it really couldn't be too poly heavy or texture heavy since it had to be importable into a compositing package.  While most CG applications will quite happily render high poly detailed models, After Effects isn't exactly a full featured 3D rendering package - and that also meant a lot of internal custom shaders and texture mapping techniques I could use with my rendering tools would need to be instead based heavily on colour image maps and the general capabilities After Effects offered.

Most the detailing came from image maps, with small detailing modeled where needed.

One other thing was that the ship wasn't going to need to hold up to extreme close-ups.  That meant that detail models and precision texture mapping could be relaxed a little more then I'd normally need to with a more CG detailed project.  There did still need to be at least some detail to make the machine look real.  Details in the leg mechanics and wing/engine/rudder areas needed a little planning...

Details such as legs were low-poly - however the 'mechanics' had to still hold up.

However, there was a close-up pulling back from the bridge to show the pilot flying the ship (video composited into the shot) - but this luckily was achievable with higher-res texture maps and minimal additional modeling (there was some, but only on details where the lower poly tessellation stood out in shot)

Luckily for me, the only real textures that were noticable at full HD was the overall wood on the main ships hull.  To get this to look clean and sharp in a close-up, I simply created the texture map at 4096 rather then the original 2048 resolutions I'd originally been using.

Replacing the 2048 res with a 4096 higher quality image fixed close-up issues.

Then it all changed.

After Effects is not a 3D animation tool - we knew that - and as Arvid's project went on, the limitations of what it could do started to become quite apparent.  In particular, objects that would move a fair distance would render with artifacts (caused, I can only assume, by depth calculations messing up the back face culling).  It also doesn't offer any real animation capabilities to add secondary motion - and this meant the model did have a rather "stiff" appearance.

Worse still - Adobe's decision to remove 3D model support suddenly meant that this project would need older versions of the software to be edited (and make any future changes quite difficult).  With the newer releases of the software having no backward compatible replacement - Well...  You can guess the answer on that one.

It was decided that we render it in a 3D application (in this case, LightWave3D) and instead plug the image sequences back into After Effects.  This meant I could throw in more polys, and while I did add some more "detail", I still let the texture maps control a majority of the look and detailing.

Most the look of this section are image maps. The geometry is fairly simple

Additional small secondary motions such as the hook hanging below the belly of the machine could be added.  It also meant I could throw in some occlusion to add to the final look of the machine, as well as softer lighting and shadows that AE could not.  More believable motion blur for the spinning of the prop's later on were also very easy to achieve...

It all worked out in the end

I had a real blast working on this project.  In total, around 25-26 hours were spent with modeling, texturing and rendering on my end.  Arvid's work can be seen below.  Credits also go to David Giese, who also produced this project along with Arvid for Cinelandia and Scandinavian TV.

Make sure to check out both of these guys' websites to see some of the amazing commercial work that they do.


Ljudhjältarna, Title Sequence from Arvid Eriksson on Vimeo.

Continue