Rebuilding Titanic: Part 3 - Woes of a Millennial Titaniac (Patreon)

By Kyle Hudak

Ancient Times

The landscape of Titanic research and gaming today is one of relative abundance. There are countless websites and blogs full of interesting and often high-resolution photos. Numerous Facebook groups and forums full to the brim with imagery. Entire Discord servers for the sharing and discussion of research and reference material. Museums, libraries, and archives all over the world are increasingly putting their collections online, with high-resolution images either free to access, or free to preview, or purchasable for a price. And, of course, so many books that are available both digitally and with the click of a button from Amazon. Then there's the power of computer hardware, with GPUs now seemingly competing over how many atoms they can simulate and software like Unreal Engine 5 doing away with the limitations that used to shackle game developers.

My meager Titanic collection as of March 2012. Just 3 years prior, it was even smaller.

If you’re a Gen-Zer making ship models in [current year], you may not have much appreciation for just how much more difficult it was for us old folks (Millennials) to make our little Titanics back in the day. I see so many young people now building incredible renditions of Titanic in many mediums, with the benefit of so much information at their fingertips. The average Roblox builder probably has far more reference material at their disposal than Matt did when he started modeling Titanic in 2006 and when I started in 2009.

In the 90s and early 2000s, there were a host of websites dedicated to Titanic largely run by the "old guard" of the Titanic world. A lot of the popular Titanic historian names you hear of today were at least somewhat online back then, either active in the early Titanic forums or with their own websites. A particular favorite site of mine was the late Roy Mengot's Titanic Wreck Model site. Roy's site was a feast of cool information sprinkled with low-resolution photos of an absolutely fascinating wreck model and a touch of 90s Dot Com garishness. Back then, if you wanted to see the wreck in any clarity, this was probably one of your best bets.

Roy Mengot's Titanic wreck model page, which still somewhat exists in archived form. Mengot was a wreck researcher whose major contribution was the bottom-up breakup theory, which is probably one of the most popular theories today. Mengot passed away in 2015.

Another major institution for online Titaniacs was the Titanic Research and Modeling Association, or TRMA. It was a great general resource for those looking to learn about Titanic, but its best feature was its forums. Over many years, countless Titanic enthusiasts, modelers, researchers and historians posted and engaged in threads dedicated to practically all aspects of these ships. If you spent enough time there, you probably would have happened across many awesome things that would have otherwise remained hidden. Even I posted there back in 2008 to show off my 1/350 Titanic build. Sadly, TRMA went the way of many of those old sites and is now relegated to the graveyard that is the Wayback Machine and other archives.

The TRMA was a leading resource for Titanic modelers. While there are many more groups and resources today, its loss was still a major blow to the Titanic community.

In my earlier scale-modeling days, I spent hours browsing these pages. I referred to Mengot’s wreck model images when attempting my own cardboard wreck model. I had a yellow "Nye Labs'' binder full of printed pages from the various Titanic pages including Encyclopedia Titanica, another major repository and forum of Titanic information that's still thriving today. Those prints mostly weren't of much use, but I liked to have them for reference or for when I wanted to feel like some kind of big Titanic expert. I still have that binder, probably now a record of web pages that vanished long ago.

My Nye Labs binder full of old printouts. It's mostly pages from Roy Mengot's old site, probably printed sometime around 1999-2000. Some are marked as copyrighted in 1997.

Pixels and Dots

If the development of Adventure Out of Time is any indication, it was indeed possible to find somewhat better material online in the 1990s and 2000s (though that material may have also been subject to technological limits), but it was still extremely difficult for the average person to find that stuff. It got easier over the years as more things went online and search engines improved, but you were still lucky if you could find a Titanic or Olympic photo with pixel dimensions bigger than three figures.

Two examples of the typical images you'd find online one or two decades ago. The first was a 300-pixel cropping of some furniture in a stateroom. The high-res image we got in 2018 gives a whole new view of the details. The second was a low-res copy of an electrical panel. The high-res image is clear enough to read the labels for the switches. This level of clarity was often difficult, if not impossible, to find for most images that were easily available to the typical Titaniac. Most images in my early digital archives were 1k pixels and under, with only some getting near 2k and even fewer over 2k. Now, you can simply order many of these photos or view them for free at fairly high resolution online.

And if you wanted plans, you were out of luck. Bruce Beveridge’s General Arrangement plans were available early on, in paper form, for a price. Various other plans were so under-detailed or such low quality that they could easily be ignored. Iron plans were extremely hard to come by, usually hidden in private collections or buried and lost in various archives. The plans that were digitized in those earlier years by various organizations were of poor quality even at higher resolutions. Even today, many of the iron plans we have are only in black and white with a “crunchy” and extremely burned-out appearance, making them difficult to read. It was only in the last decade and a half that some of these plans were digitized in much higher quality. Some plans could only be found in old copies of books like the special Titanic/Olympic volume of The Shipbuilder which features amazingly detailed plans of the engine and boiler rooms.

No, your internet hasn't reverted to dial-up, nor has your monitor suddenly switched to VGA. This is just what we had to work with in terms of iron plans up until around 2020. Tantalizing, tiny, limited, low-resolution croppings of pieces of some plans - here all from Dr. Hahn's site. Without these plans, it was impossible to model the ship to the level we're doing now.

There were also plans created by researchers like Dr. Robert Hahn and Dr. Robert Read, both printed to-scale and useful in different ways for modelers, and both with their pros and cons. They also cost quite a lot and were only provided in printed form on very large sheets. It was only in the last couple of years that Dr. Read began offering his plans in digital PDF form AND for free.

Me with my copies of Dr. Read's Titanic CAD Plans in September 2012. They can now be found in PDF form for free, generously provided by Dr. Read. I had based my earlier Lost In the Darkness and THG hull model on these plans, which turned out to be something of a mistake due to the fact that I had to work with paper copies and other issues.

Beyond the internet, books were still your best bet for good Titanic reference material. Even then, what you'd get was pretty limited and usually at a quality determined by the size of the dots making up the printed pages. Even 2008’s Titanic: The Ship Magnificent, with its many photos and even some iron plan samples, was of limited quality and use beyond a certain point. The most useful book, at least for me, was Peter Davies-Garner’s ‘RMS Titanic: A Modelmaker’s Manual’ from 2005. It was full of plans for many parts of the ship’s exterior and was printed with excellent clarity. But even with that, there was only so much info you could glean from the material.

Yes, you could get books for their photos, but it didn't do much good due to the way they were usually printed. While these are actually two different photos from slightly different angles, the one scanned from a book (left) and the one scanned from a photo print (right) show the incredible difference between sources. Only seeing the book version, you would have never seen what the docking bridge buzzer looked like, or that it was even there.

And so, when some of us on the THG team began modeling Titanic and during the early years of our projects, we were still firmly planted in this "dark age" of Titanic information. If you wanted to have more than a few scraps of mostly low-quality reference material, you needed to be REALLY good at searching the internet, know who to call or who to email, needed a lot of money, needed plane tickets to travel to some overseas archive, or you simply needed to have friends in high places - usually established historians. And then there was the other side of this very annoying coin: Technology.

We're Gonna Need A Bigger GPU

My work area back in the day, probably around 2013-2014. I know, it's the height of tech.

When Adventure Out of Time was produced, it’s safe to say personal computers had profound limitations. The first 3D games were released in the 1990s, but even by the 1990s there was no way to properly simulate something as complicated as an Edwardian ocean liner in the kind of detail we seek. CyberFlix had to resort to pre-rendered still images for the incredible detail they needed, as did most other game developers using photo-realistic environments.

The 2000s weren’t a whole lot better. Games were getting increasingly 3D and more detailed, but they look pretty bad by today’s standards. Polygon counts and texture sizes reigned supreme, and you could forget about trying to recreate an entire ship. Though that didn’t stop some people from trying, such as the aforementioned Half-Life Titanic mod or the Mafia Titanic Mod, albeit with huge limitations on detail. Most attempts simply did not get far at all.

Early models made for THG and its predecessor were made with the idea of using as few polygons as possible. This was compounded by the fact that a lot of details just couldn't be made out in the low-resolution photos that were readily available to us back then. The result was blocky models with little detail, models that can be found dotted around Demo 401 now. Our relative lack of modeling skills back then didn't help.

Even by the late 2000s and early 2010s, as games got increasingly realistic and game engines got more powerful alongside CPUs and GPUs, you had to be careful where you spent your details, and you had better not forget those LODs. Poorly-optimized models could mean the death of a project or the need to go back to the drawing board, and there were still huge limitations with rendering lighting in real-time. Yes, there were more and more games with fantastic graphics AND great performance, but this still came at the cost of unrelenting optimization; judicious use of repeating assets, careful consideration for poly counts and texture sizes, and game worlds practically built around the limits of the tech.

Even on a personal level, it would have been difficult to deal with anything even remotely resembling Titanic: Honor and Glory as it stands today. It wasn’t until later in the 2000s that my family got DSL internet in the house. Until then, we were using dial-up. You plugged your PC into the phone outlet via a modem, the modem would make horrifying sounds, and you’d get connected with the incredible ability to download a 256-pixel-wide image in about ten minutes. That is, until someone decided they needed to make a call. In the early days of THG, one of our team often quickly reached the end of his rural internet data plan and would practically have to vanish for the rest of the month. It wasn’t until 2017 that our house finally got cable internet. Yes, there were absolutely faster speeds possible even back then, but it took a while for us poors to catch up. Fiber wasn’t even installed in my area until 2022-2023. There was also the simple reality of data storage and file sizes.

You would have needed around 160 of these things to store Demo 401. True, it would easily fit on the drives we were using in the late 2000s and early 2010s, but we would have needed to do some spring cleaning to make it fit.

In the days before digital distribution really took off, and before more people had faster internet, you really had to check yourself. This was not as much of an issue due to the fact that computer tech and game engines in general were enough to keep files sizes low, but it was an issue on the production side and for the sharing of information among the team. Making models takes up space. Reference material takes up space. This in turn limits what you can store, send, or buy. My main research folder for THG is over 1 TeraByte. An earlier archive of THG work was over 150 GigaBytes, while a newer one is nearly 350 GB. Demo 401 comes out to 40 GB while its raw project file is much, much larger. A single scan from my recent digitization runs well over 300 MegaBytes while the processed versions are still over 100 MB. A single Maya save file for the structural model is already at 1 GigaByte, and that will only get bigger. And, of course, games today are seemingly in a race to take up more and more of your drive space, with some of the biggest games reaching hundreds of GigaBytes. Thankfully, the abundance and cheapness of high-capacity drives and the ability to download stuff at incredible speeds makes all of this much easier nowadays.

All of this is to say: If you’re one of the many Gen-Z people working on Titanic projects now, be thankful for when you were born. Yes, you may have been born into a world that’s getting more chaotic by the week and subjected to a constant blast of all the worst information imaginable, but at least you have nearly endless options when it comes to recreating all of your favorite ships. You’re light-years ahead of where we were in the late 2000s and early 2010s, and it can only get better from here. That said, there are still limits. Many archives are still somewhat difficult or even impossible to access. We ourselves are unable to share our collections freely due to a whole host of reasons, including copyrights and the wishes of collectors who lent us material. And while the latest tech might allow practically unlimited detail, not everybody is building their recreations with that tech - there's only so much you can do in Minecraft or Roblox.

But despite all of the limitations of the time, it was in 2007 that a particular game entered the market, a game that pushed the boundaries of late-2000s hardware, a game that looked real even by today’s standards, a game that caused people everywhere to ask “can it run Crysis?”

End of Part 3...