TV.3 Asset Development Analysis (or, the pre-post production blog post)

The last page of TV.3, air -44-, ATC page 635 (story page 612) rolled off the assembly line around 0600 EST this morning. That makes for a total of seven pages in the buffer. On the current Sunday schedule that covers the rest of October and all of November, with the last page finished this morning having a projected post date of November 30. Air -42- will be posted on the 16th and is closest to ATC’s 20th birthday. I can never remember the exact date; I know it’s November 1994, somewhere in the middle. One of the “teen” days.

The comic is currently two pages into the last major production block of the chapter. I took advantage of a scheduling hole in late September and got the rest of the Daedalus shots rendered in a marathon work session. After that, pages 3.36 through 3.38 were done in one production block and 3.39 through 3.41 were created the week after. After finishing line art for the chapter I spent a fortnight on asset development, which is the point of this blog post. As I’ll be discussing material that will be technically spoilery for the next few weeks I’ll tuck the rest of this post below the fold.


The major asset developed during the last 3d production block of the chapter was the Astra, a Heirotus warship first mentioned in 2008. The hull was developed during downtime in March of 2009 and was shelved when I decided to finish Dead City Radio before progressing Transitional Voices. After finishing line art for the chapter I blew the dust off, assessed the geometry, and proceeded to finish the ship. The shape is essentially unchanged – it was a big step forward at the time and i found it acceptable and easy to work with. Unlike the original version of the Castores, which I threw out.

For good reason.

For good reason.

The Astra represents the current direction in 3d asset development – reliance on geometry over textures. Like the Castores and all non-Sabrosa vehicles the windows and hatches are geometry. The hull plating is a simple texture and the hull numbering has been placed using a separate texture channel. The end result is that while I spent about ten days noodling on the Astra, only one of those days was spent in 3d Studio, and the Astra has three maps to the Sabrosa’s ninety six.

The final shot of the chapter originally called for a Loki with Banshee escort. With eight weeks between finishing the Astra and the last page of the chapter I figured I had time to create a new modern asset, something that would be as easy to work with as the Astra, something that would be less of a pain in the ass to specialize for future re-use. The “Bodkin” – a pointy little space fighter with clear Amarr, Caldari, Kilrathi and Earthforce ancestry – popped into existence in a mere two days (one and a half work sessions!). The “Peregrine” – an armored VIP transport – came into existence almost as quickly. Three new ships in a little over two weeks, which gives me seven weeks to work on the website, work on the script, and do things other than page production.

The fact that the Astra appears in three panels and the Bodkin and Peregrine appear in one got me thinking about chapter asset development. Since the epic struggle to finish the Banshee and associated launch tube I’ve had it up to here (some point roughly 12″-18″ above my head) with asset development. I’m not done with 3d development, mind you – I just don’t want it to take forever anymore. I don’t want it to suck. It’s going to take awhile to get interior work to something nice and simple but I’m there or nearly so with vehicles – the stuff I’ve developed this year is less reliant on texturing than previous vehicles and environments – the cargo bay is an exception but the cargo bay has to match the rest of the Daedalus (and in fact was designed to complement the interior and exterior) and is actually a huge step down from the craziness of the rec area or the bridge. The fact that I was able to produce useable vehicles – the Bodkin and the Peregrine – in a very short period of time speaks well for the future developments. The list of assets required for the next chapter is, it turns out, shorter than the list of assets developed for this chapter.

More on those at a later date. Possibly a much later date; possibly a post like this one after I finish work on the next chapter. What I’ve spent seven paragraphs getting to is this, a BULLETED LIST! of assets developed for TV.3 and the number of panels they appear in.

  • The Castores appears in eighteen panels, though that’s including every instance of just the cargo bay or just a piece of deck plating. Its sister ship appears in the next chapter and will be a palette swap with additional mission modules, so most of the development on that is complete. If I’ve interpreted file creation dates correctly the Castores hull only took about five or six days to model, with creation taking place between the cargo bay and the shuttlepod during the spring modeling sessions. The Foundation class ICG cruiser was originally a much blockier hull developed in August of 2008 – a year after I’d started picking away at the Hemera and a couple of weeks before I finished that ship for use in Incursion. While I was comfortable blowing the dust off of the Banshee and the Astra, the first version of the Foundation class had aged badly (see image above). The old low-poly hull looked horrible and would have required extensive work to bring up to spec – hull edits followed by kibbling and extensive texturing. I figured it would be faster to start fresh and the existing Castores hull was the result.
  • The cargo bay (see last panel) is a critical part of TV.4 and technically appears in ten panels of TV.3. This includes all exterior shots in which the nose doors are any amount of open. The cargo bay required an extensive amount of modeling and texturing work to develop, in large part due to the fact that it has to match Daedalus production values and design decisions. Modern knowledge of lighting and modern modeling skills reduced texturing efforts, though the environment still took about two weeks to develop.
  • The shuttlepod is a critical part of TV.4 and appears in seven panels of TV.3. The vehicle has minimal-but-detailed geometry, ten texture maps, and a simple design that impressed Robot Jason. Development time will pay off in the next chapter. The asset took around ten days to develop.
  • ELG-2623.4 appears in thirteen panels counting shots of the rings.
  • ELG-2623.4.21 appears in seven panels. The surface was developed with LunarCell, which I will likely use for Hrazny MrDat.
  • The Astra appears in three panels. Begun in spring of 2009 and finished in October of 2014 this asset has the longest development periodĀ of all ATC vehicles to date, though it was shelved for the supervast majority of that time.
  • The Bodkin appears in one panel and was accurately implemented from sketches in two days.
  • The Peregrine appears in one panel and was developed from a series of sketches exploring possible replacements for the Loki. The ship had a development period of roughly three days.
  • The Bedlam sets in air -3- were developed from existing assets and used for five panels each.
  • A spartan sickbay environment was used for four panels. As all of these are close cropped and moodily lit or heavily distorted this turned into an experiment with as-needed modeling. Not much was needed, so I can start fresh in TV.4 if I want to.
  • The Pittsburgh interior took a couple of days to work out and was used for ten panels. Built from memory and reference, this is the last planned appearance of Thad’s apartment. As the panel count was small and the focus was on characters and not the environment I kept things minimal, with an emphasis on lighting.
  • I did some brief spot modeling to the rec area to accommodate the the script – a drawer was added to the dangly bit over the table, and the fire extinguisher bracket had to be made useable instead of decorative. That didn’t take long. The real gain there was new hardware, which has enabled me to merge the the central corridor, rec area and cargo transfer lock environments into a single seamless environment. No new material (aside from the medkit drawer) but that one optimization saved quite a bit of hassle. The central corridor and the rec area were composited from separate renders in the previous chapter, and very carefully didn’t appear in the same shot in the chapter before that. Yay hardware!
  • Oh and the general accuracy of the Space suits is due to the creation of props of both suit types. I got a lot of mileage out of Inga’s. Less out of Xand’s, though we’ll be seeing more of it in the next chapter.

As you can see that was a lot of effort for a relatively small number of panels. The shuttlepod, bodkin, peregrine, cargo bay and Astra will all be reused in the next chapter and beyond. The cruiser hull will be tweaked into its sister ship for the next chapter, and may appear again after that. I was able to reuse the launch tube from water -1- and I never thought I would until last night so anything’s possible for the Pittsburgh environment. I’ll probably develop the sick bay set out of the sick bay rough, though at this point I can take it in any direction and don’t intend to commit to any kind of interior design until there’s a script and layouts – two things that are starting to inch up there on the to-do list.

So, in summary… two new planets, one new high detail interior, one new low detail interior, one super-sparse suggestion of an interior, some minor set redressing, and five new vehicles. The next chapter will have at least two new interiors, one new planet, and possibly some other stuff of indeterminate complexity. Hopefully I’m over the hump for the book – this thing’s been in production for a long time and I can’t wait to be finished with modeling. Of course, as much as I’d love to just dive right into the next project on the list I need to take a breather, work on the script, prioritize, work on the website, manifest November rent, etc, etc.

I suppose the tl;dr is this: It took awhile, but I’ve finished work on this chapter. The remaining 14.28% of this chapter will post every sunday between now and the end of November regardless of what’s going on in my life, even if I have to mass upload and schedule the posts.