Summer Surprises: Behind the scenes with the iniBuilds Scenery Team!

[iniBuilds]

Hi everyone, 👋

I’m Greg Jones and I’m the Head of Scenery and Geospatial Development here at iniBuilds.

Greg looking very happy at his desk! 🙂 

iniBuilds CEO, Ubaid, explained his vision for the iniBuilds Summer Surprises event this year, and said he wanted to take things beyond just another excellent scenery release. I always love talking about scenery development and this seems like a good enough reason to reveal some of what we get up to "behind the curtain"!

I wanted to share with you a little bit of background about our scenery development pipeline and some behind-the-curtain insight into how our vision and efforts find their way into your sim.

My journey with flightsim started way back in the 80s when I first caught sight of Sublogiic’s flightsim on my cousin’s Apple IIe iirc. Just a few years later, when I got my hands on a PC, I came across a copy of Microsoft Flight Simulator 1.0 on a bunch of 5.25” discs I was given. I was immediately hooked.

Fast forward to 2006 and the advent of FSX; at that point, several tools became available that made scenery development a much more open experience.  I was a member of the old AussieX community flight group and an avid follower of the freeware development there. I’d heard that a new version of my local airport, Moorabbin, was being developed by Ken Hall for Orbx and I offered to assist with site photography as I grew up virtually in its shadow.

Ken offered to teach me some basic modelling in gmax and together we completed what became YMMB Moorabbin for Orbx. It remains my favourite scenery and the one of which I am most proud. Ken and I went on to produce other small airports together - namely  Hobby Field and Longview Ranch.

At that point I relocated to England, where I still live. I began working as Project Manager in 3rd party scenery development for a local company. At that time, Goodwood was in production and we would go on to complete a number of small UK airports; Shoreham, Popham, Compton Abbas and Caernarfon to name a few. It was there I first came into contact with development  legends like John Venema, Russ White (The Texture Wizard), Bill Womack and Jarrad Marshall.

An image of YMMB - Moorabin in FSX - Greg's creation he is most proud of.

Sadly, that company did not last but, along with Russ Linn, we took the opportunity to start our own company, Turbulent Designs. At TD we were laser focused on developing the best quality airports possible, within the constraints of the ESP architecture. As well as creating a total of 20 new airports, we were also responsible for updating most of the existing Orbx library to support changes in P3D and porting that library to X-Plane. We also worked with a number of large organisations in aviation, marine and defence industries on various visualisation projects. We even did some work on the short-lived DTG Flight School (remember that one!?).

After 7 years with TD, it was time for a change and I went to work, full time with Orbx where I stayed for nearly 3 years. It was during this time, at the end of the last day of FSExpo in Orlando, as we were sitting in the hotel bar with other developers, that the MSFS announcement trailer dropped and it seemed that our little flightsim corner of the world went crazy. We set about re-tooling our development processes to include the new sim which soon became our primary focus.

Microsoft Flight Simulator sunsets looking stunning as per 🥰

In 2022, the opportunity to work with a relative newcomer to the industry, iniBuilds, arose. In my first conversations with Ubaid, I was struck by his ambition and vision to achieve the highest levels of quality content available in the market. 

The iniBuilds team had already released the original version of Heathrow along with Kerry, Shoreham and Queenstown when I came onboard. Buffalo was in its final stages and LAX was well on its way.

iniBuilds Los Angeles International Airport - KLAX - one of the first major projects Greg led at iniBuilds 🌴

iniBuilds consists of a team of incredibly clever and creative people spread far and wide across the globe. Being a cloud-based company means that we aren’t limited to only those artists who live locally, but can make use of the very best people available. The downside is that it can be difficult to overcome varying time zones, cultural and language differences. This was my biggest challenge, having spent most of my professional life in an office surrounded by a team.

Another challenge we faced was that everyone came from a different level of experience and exposure to flight simulation and aviation in general. It turns out saying “make it low poly” means very different things to different artists! It was essential that we needed to establish a baseline and a set of standards in order to maintain the level of quality and performance we wanted to achieve. This is a process that continues to this day, but thanks in no small part to the efforts of art team lead, George Cass, the team continues to grow and progress in the same direction.

The last big difference for me personally, was that I had only worked on a small handful of what could be considered ‘large’ airports. Coincidentally, the largest in terms of the number of models and groundpoly complexity was my good old Moorabbin.

An airport like LAX is a massive undertaking for anyone. Not only did we need to produce one of the most iconic airports in the world, it would be only the first fully inhouse iniBuilds airports under our new team. The technical goalposts in flightsim development are always changing and we encountered many headaches and stumbling blocks. Not only are there around 150 jetways all needing to be created and animated, but the area is covered by detailed photogrammetry. 

Photogrammetry (PG) is both a blessing and a curse; no longer do we need to worry about the laborious autogen process (yes all those trees and buildings in FSX were placed by hand long ago!) and the environment looks incredibly lifelike at the right distance.  However, the terraforming tools available in the sim that we would have once given our right arm for are rendered almost useless by PG data. Fortunately at LAX, after many hours of trial and error, we were able to get a satisfactory result, if not exactly what I had envisioned.

An example of one of the tools we have for performance enhancement - LOD's! 😲

After the successful release of LAX and armed with the confidence of being able to produce large airports at high quality in a reasonable timeframe, we began to expand the team and broaden our outlook in scope of sceneries.

By most sim development standards, our scenery team is very large. Currently we have 28 artists, specialising in various disciplines. Over the years we have honed the team’s skills and we have tried to foster talent within the group. I’ve always found it amazing, and at times frustrating, that there’s no one right way to achieve a development outcome. I’m constantly surprised at the ingenuity of our team to adapt to an ever-changing platform.

Many of the comments I’ve read over the years have been disbelief at how quickly we are able to release a new airport product. To the outside observer, it may seem that we never sleep, that we don’t focus on detail and cut corners wherever possible. I can assure you that that is most definitely not the case (except the sleep bit!). As mentioned, we have a very large team spread across the globe. 

Depending on the size of the project, a large airport generally takes about 6 months to complete once production gets underway. On top of that there’s a lot of planning, site visits and testing to arrange. A small airport such as Palm Springs, could be done in about 10-12 weeks.

Having a large team means that we’re able to have multiple projects running concurrently and that some projects may be partially completed by one person, ready for someone else to take up the mantle some time later. This allows us to have projects running more efficiently without any downtime and to not have the team burning out.

At any given moment, someone, somewhere in the world is working on an iniBuilds airport & maybe enjoying some pizza on our monthly virtual pizza parties!! 



Another often-mentioned issue with our airports is performance. Yes, there have been some issues with some of our products and we’re always working on bug fixes and performance improvements.  Most of the issues stem from the sheer scale of the projects we work on and the amount of detail and content that we include. Whenever I see a comparison between an iniBuilds scenery and another developer, I always make it a point of checking out the competing product. Usually, there are fundamental differences in complexity and detail that form the heart of our airports.

Optimisation is an over-used and perhaps misunderstood term to describe performance of flightsim addons. The guidelines for MSFS optimisation are very clear and well documented but often not adhered to. This is something that we’re very proud to say we follow to the letter, despite what a few nay-sayers might suggest! Combined with a few of our trade secrets, we generally keep things running well. That’s not to say that we always get it right, or that the guidelines are infallible. Some of our best selling airports have been problematic in terms of performance on lower end systems, and this is something that we’re always striving to rectify.

We recently completed an overhaul of our largest and most ‘top-heavy’ airports, introducing a more finessed method of both instancing and lodding in our models. This has vastly improved performance in most cases, particularly on lower end systems. As with everything though there’s always room for further improvement and we’ll continue to review and update the entire catalogue.

iniBuilds Dubai International is one of our recent larger hubs that has had an extensive update a few weeks ago! 😎

From a technical perspective, our focus has been on finding just the right number of assets that can be instanced while keeping the node count at a reasonable level. As all of our sceneries include detailed interiors, this can be a tricky and time consuming balancing act. In addition, we’re streamlining drawcalls wherever possible while reducing texture resolution across each LOD level.  We’ve invested a vast amount of time and energy into striking the right balance and we’ll continue to do so.

Airport terminals, as a general rule, don’t lend themselves to lodding very well. LODs and LOD values work best when applied to mostly uniformed shaped objects such as a car or a cuboid building. By necessity airport terminals are long and thin. This means that you end up with a very large LOD radius. You can be both very close (at either end) and very far from the building (in the middle) almost at the same time. This creates a nightmare for performance when considering LOD values. 

LOD values are calculated (in MSFS at least) using an invisible sphere which will be the size of the largest dimension - in the case of a long terminal model, the ends. The idea of using a sphere is to create a standardized shape around the object that remains the same no matter what perspective you’re viewing it from.  This is where it gets complicated. LOD values are based on the vertical height of your display. If a cuboid object is taking up 50% of your vertical screen space, it effectively has a value of 50 or thereabouts (this varies depending on your object draw distance setting). If that object is very long and thin, although it might appear to be taking up 50% of screen space you’ll actually be inside the LOD radius where LOD values are no longer effective.  There are ways to overcome this and that’s a large part of what we’ve done as part of our overhaul. Fortunately, in this regard, the future looks much brighter!

Our second in command in the scenery team George hard at work reviewing the old Heathrow and identifying assets that need improvements for performance 💻

A lot of people ask us what are thoughts are on Microsoft Flight Simulator 2024! 

We’ve all seen the announcement trailer for 2024, and the thing that struck us most is the amazing ground details that will be available to us. In the meantime though, we hope you enjoy iniBuilds Dubrovnik International - our latest scenery release. It was a really fun one to work on for the team and we’re very proud of the result! A lot of people might be asking what our update policy is going to be? 

iniBuilds Dubrovnik International available today!  🎉

We have decided that most of our library will be updated free of charge; this will be all products released in 2024 and most in 2023 but there are a select few products that will have entirely new versions made from scratch. Rest assured all current owners will receive a discount on the new versions (which will be significantly better than their predecessors). 

Well that's for the future, but what about from now till the new simulator? We have a few more updates to do on some of the current library and a couple of new airports to go! 

Right now, we’re working on a project that I’ve been wanting to do for many years and I hope you’re going to be as excited about it as I am!

Can you figure it out!?

Thank you for taking some time to read our behind the scenes update!

Many thanks

Greg Jones & the iniBuilds Scenery Team ❤️ 
 

[Manu85]

Finally,  F A O R  !

[arwasairl]

What a small world Greg! I worked with Orbx briefly for their KSJC product for P3D, and who would've thought, you were also involved with that particular airport too.