There is a very well written article in the German PC games magazin "GameStar" with some interesting and exciting news. I will try to translate the core content of it:
page 1 - 2: describes some history of the founding campaign, that Chris Roberts tried to start it via own servers which collapsed during the first hours, that following to that he went over to kickstarter, collected 2 Mio. Dollars during the next 30 days plus 4 Mio. Dollars via his own page.
page 2: gives an outline of the basic goals of Star Citizen, that it was planned as a readily comprehensible space sim project with a single player campain, just like the good old "Freelancer". But as the funding wouldn't stop, Chris Roberts more and more came to the conclusion, that he now could be able to do this project on his own, leaving possible investors outside (which he wanted to involve in the beginning). Plus he needed to install a studio structure, which means: the project officially started end of 2012, but this was not the beginning of the development as it is today. In fact, in parallel to founding the Austin studio, he took external studios under contracts like Behaviour Interactive, IIIfonic(*) and Moon Collider. From the beginning, Chris Roberts has had huge problems to recruit developers who had experience with the CryEngine. Good luck bad luck story: when Crytek UK went down, in April 2014 CIG hired every related professional they could get.
(*)...white box: they were contracted to build the shooter mechanics before CIG founded the Frankfurt studio because IIIfonic had some CryEngine experience.
page 3: starts with the fast growth of CIG which results in structural problems by end of 2014. At the same time, it came to a catastrophe: the report points out that IIIfonic delivered bad quality because their maps didn't match the specifications of CIG because they were based on a wrong scale; thus were too small. Roberts states that it seemed that they were already ready to go, but at the end that had to do it again, starting from the beginning. This crash influenced the whole development of Star Citizen. As a result, CIG canceled all external development contracts for making most of it on their own.
Very helpful for becoming an ooperational studio structure was the experience of Erin Roberts and some of his colleagues who had worked on the Lego(tm)-games before.
White box: a short view onto the CryEngine and if it was the wrong choice. At the point Chris Roberts started to realize his vision, the CryEngine was the most potent graphics engine available. The Unreal engine [UE4] was not released at that time, and to create an own engine would had cost at least 2 years. Plus in 2011 the size of the project was not even imaginable (pls. refer to page 2 "basic goals").
page 4: continues to explain the way of a "basic game idea" to a huge multiplayer space sim which never has been done before. Star Citizen came to the point, where the 32bit engine couldn't handle the (planned) size of the instances anymore. It was good for doing i.e. Arena Commander maps/sessions. But it wasn't possible to build a "map" like the today known Crusader. Realizing this, CIG began to re-develop the engine into a 64bit based engine, which is ongoing since today. To meet precisely the size of space (keep in mind: a map of Star Citizen can not only include square kilometers of space like recent shooters but quadrillions(!) of square kilometers) it is of essential need to get the precision of 64bit calculations.
End of 2016, approximately 50% of the re-development is done. Currently the "Star Citizens" can explore around 400 quadrillions square kilometers of space. Its empty space for sure, but it shows even today the potential of the new engine.
But all this work costs an enormous amount of time and effort. This, in turn, leads to the (partially rude) critics of and by the community.
White box: explains that the 64bit engine works right now and distances are calculated precisely. A user has proven that, pls refer to youtu.be/t51EDw-mdHo
page 5: Roberts states that he probably should have stated the problems and current tasks in a more transparent/direct way. He excuses by saying that he for himself would very like to getting things faster, looking to the development speed. But the whole team of CIG is working really hard, and the delays are not risen because of unmotivated team but of the before outlined "start-up troubles".
2.0 was the most challenging patch up to now because it was the start of the 64bit-based engine and the first step away from single gaming modules towards a persistant universe.
Furtheron the report gives a short look back onto GamesCom and CitCon 2016, which made it the most successful year in the history of Star Citizen.
White box: informations about 2.6, new ships and Star Marine as well as the drama around server pings (foremost in Europe), balance and flight model.
page 6: speaks about Squadron 42, which is kind of "victim" of the before described problems. Again, the focus pans to the community and the continuing criticism of the delays and intransparency of the developing process. Chris tells us that the community is great and so on (you can read between the lines that he sometimes is a little bit annoyed): "People are sayin: give it to us, give it to us now. We don't care about bugs". And then you are giving it to them and they say: "This looks like shit and is full of bugs."...
Aside from the KI, which is planned to operate and act as independant as possible, the level of detail is a huge challenge. Chris Roberts wants that it looks everytime like a cutscene in a movie or scripted scene, but not scripted and in real time. There are problems with the illumination: CR wants it to look like a cinema movie, and for instance the Idris has thousands of light sources which have to adjusted manually.
Another point are the so-called "object containers". Within the huge maps, masses of data have to be stored and loaded. Since this data cannot be saved and loaded all the time (at once), CIG is planning to do this with a streaming technology: they are streaming "object containers", which is working in the background during the gameplay. This technology, currently in development phase, is needed for patch 3.0 as well as for SQ42. Furtheron, even procedural generated planets are part of SQ42, which explains the delay of the release of 3.0.
White box: continues the information about and around 2.6 from the former page, now describing the new camera "directors mode camera", ending with excited words about the level of detail and atmospheric gameplay.
page 7:
white box: discusses the point of "does the development of Star Citizen take too much time"? Its been again explained that CIG ot only had build an AAA-game from the scratch, not only to hire some developers but to install an international studio structure. References are given to GTA 5 (development time 5 years), Starcraft 2 (7 years) and Diablo 3 (first planning around 2000). Result: everything okay and in time.
Furtheron, 3.0 is now focussed: there is no schedule yet. But its being worked on the core development goals like air/space traffic control especially above landing pads.
page 8: Chris Roberts states that this is not really a challenge, but its one of the things that has to be done next, as well as elements like boarding and security (to lock up your ship, for instance). Every time, they are looking on the content and gameplay elements of 3.0 the team is saying: "Damnit, thats more like any other common game contains". But they are keeping focussed plus working in parallel on the performance, net code (aka number of players in one instance).
Talking about the netcode, Chris Roberts states that this is the most critical challenge in the whole development process, not least because there aren't that many good netcode programmers on the market which are available. Going over to Lumberyard shows the importance and the willingness to realize the vision of Star Citizen. This contains, as a sidenote, the (I don't know how to translate this) "physical grid-in-grid technology" which i.a. is responsible to hold the position of the space craft when you going EVA or back into it (that it does not spin around!).
Furtheron, Chris Roberts talks about "Item 2.0" or, as he names it, the "Entity 2.0". Every entity has its own components, which can be a physical, graphical or a radar component. Every compenent of the "entity system" can be manipulated specifically, some of them will be "refreshed"/updated by the server every second, others every couple of minutes. This results in a maximized flexibility and a most efficient data transfer. Entity 2.0 will be the basis i.e. for ship components and user HUD and integral part of 3.0.
page 9:
Its all about procedural planets, eco systems and buildings and so on. They are finishing currently this automated systems therefore CIG will be able to design a planet within a week. Some parts will be made individual, most of them automated. On every planet and some stations there will be special missions and "super missions".
Also based on Entity 2.0 you will be able to build your own hangar, houses and even bases (for corporations). You can land on some planet, build a powerplant, secure it with a jamming station and then start to build a base, do farming or whatever. But: CIG will limit this gameplay element somehow for that not every player can build his/her own mega-city.
page 10:
White box: all about the money. Chris Roberts states that he has a good sum of money as backup. As long as the money flows, the development will continue. Even if there will be no more funds starting by tomorrow CIG would be able to finish and release SQ42. He finally states that Star Citizen is a live game and that they are planning to continue with the development constantly, adding new features and content.
This page speaks about sandbox gameplay, in which corporations are meant to fight each other like it is in EVE.
The report closes with a big thumbs-up towards CIG and that the community should give the team of Chris Roberts the time they need and that should not worry because everything is in time and ongoing.
