Star Citizen - The E.L.E.

[AP]

What I don't know is what the game they release will be like, what features it will have or how big it will be. I suspect they will go the MVP route and we will see a lot of stuff patch in later.

Given what you have said how concerned would you be if it came to light that CIG was downsizing and staff were leaving?

[ConfusedMonkeh]

What can't be held in memory? Millions of kilometres of nothing with some POI markers and a few low LOD images of planets and a star? I think you're wrong. The 2.5 alpha which does exactly that, only over a smaller space, (which is still around five factors of ten larger than current maximum CryEngine map), thinks that you're wrong, Elite Dangerous, that does exactly that, thinks that you're wrong.

Source: Star Ctien alpha 2.5 and Elite Dangerous, both showing that it would, and does, fit in memory.

You're trying to tell me that the 2.0 SC alpha never happened. Well I'm sorry but it did.

I'm not quite sure what exactly you are trying to argueng
As I understand Derek Smarts argument, he is saying that the game cannot be "seamless"...that it is stitched together from a number of instances and zones. That it is not one big map...because having just one big map means having to devote server resourves to tracking each and every object within it.

That could be players...it could be ships...it could be planets and moons and stations and it could be every single asteroid in an asteroid belt.

It isn't just POIs that a server needs to keep track of...it would be every single object within the instance.

I would assume clever programming tricks could reduce this number to some degree, but one of those tricks would appear likely to be instances and zoning so that the load can be shared on the servers. And of so, then there would need to be some way to stitch these instances together...two of which Derek Smart outlined.

And we have evidence that CIG does uses instances as it creates dedicated orbit instances and battle instances...so it musr have a way to stitch these together into a whole.

Either that or its developed a revolutionary new system that does away with the need for instancing...in which case it can easily license that technology for additional funding.

So I'm not ecactly sure what point you are trying to make since Toberts own description of how SC works tallies pretty much with what I umderstand Derek Smart is saying

CIG have rewritten parts of CryEngine that deal with the amount of space available for a single 'map'. The so called 64 bit conversion. What this has enabled is maps millions of kilometres wide. No stitching. You can fly from one side to the other seamlessly. Once more systems are added then of course they will be stitched when we travel from one to another but when within a single system there is no stitching. The game just loads up any assets as you aproach them. This is how the 2.5 alpha works. Derek seems to think that the POIs are stitched together via a QD loading screen which is patently false.

CIG may well license off their engine when it's finished. They may have already done so.

Derek asked me to provide an example of what CIG have done that nobody has done before and this is exactly that. No one has made a CryEngine map bigger than 8 by 8 km before. CIG have one hundreds of thousands of km wide. No stitching within a system. Just authoring of assets as you approach them.

[Kyrt]

Given what you have said how concerned would you be if it came to light that CIG was downsizing and staff were leaving?

It would depend on why this was happening.

Even if it were due to a lack of funds, CIG should IMO have enough cash to feature lock SC and put out the MVP they talked about.

My main concern is SQ42.

I think we are close to the point CIG have to put something out to show some actual progress not just on underlying mechanics but actual gameplay. Otherwise they risk losing support from very generous backers.

But...Infinite Warfare has beaten CIG out. Other games are following. SQ42 might get lost in the crowd
CIG hasn't locked down the flight model yet. For a game like SQ42, that seems   to be quite an oversight.
Marketing is going to be expensive....but I haven't seen any. That's going to affect sales unless CIG are going to rely on word of mouth.
SQ42 IMO needs to be good and successful. Its primary goal will need to be incentivising people to keep contributing to SC and providing an alternate revenue stream for CIG. I also have to wonder if a release would mean developers permanently split off from SC to develop new missions, whether changes in flight model, animations and so on will be carried over between the hames, etc.

A bad SQ42 release could really scupper SC.

But right now I'm wondering if SQ42 will be out in 2017. I think it will be but I admit to having wuestiond about the lack of marketing, previews, etc...everything I would expect from a major launch

[Kyrt]

CIG have rewritten parts of CryEngine that deal with the amount of space available for a single 'map'. The so called 64 bit conversion. What this has enabled is maps millions of kilometres wide. No stitching. You can fly from one side to the other seamlessly.

You can do that in other games as well, such as Elite and these games require "stitching". ED even uses math to work with a de facto 128 bit system and its own maps are mich bigger in scale than that reported oso far in Star Citizen.

Once more systems are added then of course they will be stitched when we travel from one to another but when within a single system there is no stitching. The game just loads up any assets as you aproach them.

So....SC does use a method of stitching scenes together just as Derek Smart suggested.

This is how the 2.5 alpha works. Derek seems to think that the POIs are stitched together via a QD loading screen which is patently false.

Derek Smarts analysis seem to pretty much correspond with the explanation put forward by Chris Roberts.

Derek asked me to provide an example of what CIG have done that nobody has done before and this is exactly that

And your example was one which is used by other games. Some even on a larger scale

No one has made a CryEngine map bigger than 8 by 8 km before. CIG have one hundreds of thousands of km wide. No stitching within a system. Just authoring of assets as you approach them.

You just said they did stitch the system together...just by streaming in data as it is needed. Which is one of the options Derek Smart suggested. He also suggested the use of instances and we know from what Chris Robets said that the game makes extensive use of instances. Dropping out of QD to join a battle or investigate a wreck will create a battle instance. Which again will need some sort of stitching to mesh into the rest of game.

So...I am not seeing where you and he disagree.

He is saying 'stitching is needed' and you are saying 'that is right but you are still wrong". A game world which streams in data and which creates instances as needed is one which requires and has the stitching you say SC doesn't have.

[ConfusedMonkeh]

CIG have rewritten parts of CryEngine that deal with the amount of space available for a single 'map'. The so called 64 bit conversion. What this has enabled is maps millions of kilometres wide. No stitching. You can fly from one side to the other seamlessly.

You can do that in other games as well, such as Elite and these games require "stitching". ED even uses math to work with a de facto 128 bit system and its own maps are mich bigger in scale than that reported oso far in Star Citizen.

Once more systems are added then of course they will be stitched when we travel from one to another but when within a single system there is no stitching. The game just loads up any assets as you aproach them.

So....SC does use a method of stitching scenes together just as Derek Smart suggested.

This is how the 2.5 alpha works. Derek seems to think that the POIs are stitched together via a QD loading screen which is patently false.

Derek Smarts analysis seem to pretty much correspond with the explanation put forward by Chris Roberts.

Derek asked me to provide an example of what CIG have done that nobody has done before and this is exactly that

And your example was one which is used by other games. Some even on a larger scale

No one has made a CryEngine map bigger than 8 by 8 km before. CIG have one hundreds of thousands of km wide. No stitching within a system. Just authoring of assets as you approach them.

You just said they did stitch the system together...just by streaming in data as it is needed. Which is one of the options Derek Smart suggested. He also suggested the use of instances and we know from what Chris Robets said that the game makes extensive use of instances. Dropping out of QD to join a battle or investigate a wreck will create a battle instance. Which again will need some sort of stitching to mesh into the rest of game.

So...I am not seeing where you and he disagree.

He is saying 'stitching is needed' and you are saying 'that is right but you are still wrong". A game world which streams in data and which creates instances as needed is one which requires and has the stitching you say SC doesn't have.

Are you deliberately misunderstanding? In the game as it is right now there is no 'stitching of scenes'. There is only one scene that players fly within.

When other systems are added then we will see stitching of scenes via jump gates.

Derek said they are stitching now in the alpha. They are not. At all. It's one, very large scene that doesn't require stitching.

Elite has done very large maps but not in CryEngine. That is what CIG have done that nobody has done before. A CryEngine map bigger than 8x8 km.

I'll say it again just in case. Derek says they are stitching now in the alpha. They are not. They will stitch systems together once multiple systems are in place  but the Crusader map is a single scene that doesn't require stitching.

[Kyrt]

Are you deliberately misunderstanding? In the game as it is right now there is no 'stitching of scenes'. There is only one scene that players fly within.

Which according to Chris Roberts own description of how things work, makes extensive usage of instances and zoning not least because the game can only handle so many players in any one aspect.

So I am not certain where you get this idea of only one scene. Even then, your one scene can potentially contain hundreds or thousands of objects which must be tracked...not just POIs and players but NPCs and objects such as asteroids.

You say such objects are loaded in as needed but not only is that a form of stitching but I can't see how the game would know when to load in such assets without maintaining some data and information about it. Which then takes up system resources. Of which there is a finite amount.

When other systems are added then we will see stitching of scenes via jump gates.

And why do you not consider the dataloading that takes place via this streaming or the use of onstances and zoning "stitching"?

Derek said they are stitching now in the alpha. They are not. At all. It's one, very large scene that doesn't require stitching.

If they are making use of instances the way they say they are, then they ARE stitching. Tying different scenes together in order to appear as a whole. Even in the alpha.

Elite has done very large maps but not in CryEngine. That is what CIG have done that nobody has done before. A CryEngine map bigger than 8x8 km.

So...the achievement is not creating 128 bit precision a la ED, it isn't even in creating 64 bit precision....

The achievement is in doing something several other games have done but doing it with a specific game engine no other company saw fit to do because they had better options available.

To put another way....Star Engine is not the only 64 bit engine around.

I'm not certain I would call that a major achievement. They weren't the first to develop such an engine, they won't be the last and converting CryEngine to a 64 bit system might have been a worse option than building their own custom engine from scratch. ED uses a custom engine that developers state allows for 128 bit work via SIMD.

And while CIGs maps are bigger than 8x8km, they still state they use instances and zones. They still have a limit on the number of objects and players they can have in any one scene. They still have a need to manage transitions between different scenes as players and objects and NPCs move.

Hence...stitching of some sort is necessary. Even in Alpha. A point which appears to agree with what BOTH Derek Smart and CIG are saying.

So again...I am not certain why you think that Star Citizen, even in Alpha, has only one scene.

Once you leave a planet...you get put into an orbit instance. Stitching needed.
You set a course and transition to QD. Stitching needed.
Data is streamed while in QD. This is stitching.
The Galaxy Server determines if you have a random encounter. If you do, you transition to a Battle Instance. Stitching needed.
The Galaxy server keeps track of you and other players and even if you don't have a random encounter, might determine that you need to transition to another instance. Stitching needed.

This info is from CIG....so I really do not understand why you think there is only one "scene"

[dsmart]

*I don't even know where to begin*

The "stitching" thing is 100% factual and isn't even open to discussion because I am 100% correct. That's how it is being done, it's how to do it because it would be impossible to fit an entire world into memory. Why? Because it's not procedurally generated.

When you have two scenes (S1, S2) which are different, and part of the game "world", they have to be stitched somehow. It's not only efficient, but it also gives more options for scene and player management. As big as Elite Dangerous is, they do it.

The 'entire world' doesn't need to fit into memory at the same time. The memory intensive POIs are only 'authored', via LODs, as players hit the trigger points. I presume the jump points between systems will be 'stitched' via loading screens, but inside each system, there is no loading screen when in quantum drive. You're moving through the large world space. I've seen a video that has caught it happening, a ship, QD screaming past another moving at SCM speeds.

I can, right now in the small Crusader map, QD at 0.2c ish for as long as it takes for me to hit the edge, in any direction. I don't need to go to a POI. I can fly at 1000m/s or so for an hour or four in the same direction, in an engine that used to have a limit on the size of it's maps of around 8x8 km

That's new for a CryEngine game.

Wait! What!?  :psyduck:

I know you're not a developer; so using buzz words and terminology that you don't understand in the pursuit of defending a game, while trying to give the appearance of knowing anything, is a waste of everyone's time.

Disagreeing with someone is fine, but argument for the sake of argument - when you have nothing worthy of note to add - is just noise.

It's amazing to me that you guys will argue - and look foolish doing it - just to defend a position because it's inconceivable that the "other" guy (especially Derek Smart), could possibly be right, or know more than you do.

You clearly didn't understand what I meant by this statement, and your comment (already addressed by JHow) proves it.

"The "stitching" thing is 100% factual and isn't even open to discussion because I am 100% correct. That's how it is being done, it's how to do it because it would be impossible to fit an entire world into memory. Why? Because it's not procedurally generated."

NOTHING you wrote makes ANY sense. NONE. The fact that you said "The 'entire world' doesn't need to fit into memory at the same time" is clear evidence that even though you agree with me that they have to stitch scenes together in order to give the illusion of expanse, you still chose to argue about it. If the "entire world doesn't need to fit into memory at the same time, please explain to us how exactly it is possible for them to build a world with the proposed scope seen in their starmap.

FYI "streaming in data" is completely irrelevant to this discussion because it's just a method of "loading" the data" in that you either stream it in chunks or you demand load it in one shot - which is where "load times" come from.

Also, the size of the scene/map bears no relevance to whether or not it is stitched or contiguous. They modified CE3 to give them larger scene/sizes. That's no marvel of engineering. They had the source code that allowed them to increase the scene sizes, and which they also modified the engine editor to support. But it still has nothing to do with the illusion of expanse. It's not like they built the entire world (as per the starmap) already and it's all one massive map; hence no stitching required.

In fact, right after I posted about this thread on FDev, one of the devs (Ben Parry who works with rendering tech only), chimed in. And as is the norm with CIG devs who simply opt for obfuscation rather than clarity - which is why backers theory-craft their way through dreams - he still wasn't clear until I pressed him. Here, start reading from this page. My first post is mid-way down the page, and Ben's posts start thereafter; and the discussion flows for 2-3 pages thereafter.

Agreed. But nobody does think that. That's a preposterous proposition. Each system will be 'stitched' via jump points, but inside each system is a single 'map' millions of km wide.

So what exactly are you arguing about then?

I don't have a 'developer contact'. I messaged a developer and they replied. I don't believe I am a clueless fool. Opinions on that will vary I'm sure. Que sera sera. Brian Sean doesn't need to disprove what you've stated in that video, he's just discussing what they've done. It's there to be played, right now, in 2.5. A CryEngine map, hundreds of thousands of km wide. In the engine that can only create 8x8 or so.

Brian's missive bears NO relevant to this discussion which is about "stitching" in order to give the illusion of expanse. You can't possibly be arguing that they've already built the entire starmap, that it is one contiguous map - so everything I stated is rubbish. Try not to change the subject. Stay focused.

A tier one engineer must have a long list of popular and critically acclaimed games to their name! Cool.  Let me go have a look at the absolute Metacritic scores for your back catalogue. Be right back. I could do with a new game to play.

That didn't take long, did it? This is the thing with you guys, you hate me so much that you find it inconceivable that I could possibly know more than you; let alone more than your dreamer.

The fact is that, degrees aside, I am vastly more qualified than Chris Roberts; and dare I say, every single one of his engineers struggling with this project; by the very fact that I've been built tech that they are still struggling with; and also built the sort of game they still can't build, even after all this time and money. Through the years, I have built various tech from scratch, while gaining a lot of experience doing so. That is an indisputable fact. You don't have to like it.

Granted none of you have a clue what a "Tier 1 Engineer" means in software development, let alone what the qualification for that is. But FYI, it means someone who has experience in various fields, while mastering in one or more; and who is technically experienced and qualified for a CTO position. That's why that position exists. The CTO is the lead for the company/project technologies and is the highest position in that sector. When a Tier 1 software engineer is in a technical engineering meeting with other engineers (programmers, data scientists, artists etc), there isn't a single subject that should be foreign to him, or which he doesn't have experience in. e.g. if an artist and programmer are discussing how to handle/implement PBR in a rendering pipeline, though that's not his job, the CTO should know precisely what they're talking about. When an AI engineer is talking about an FSM, pathfinding, RBFN, Hopfield or similar, and how it's used in the game, the CTO should have a working knowledge of what is being discussed. Now you get the idea.

As someone who has been in every facet of software development for over 30 years, be it AI (FYI several academic articles and university thesis have been written about the AI which I developed for BC3K. Yeah, Google it), graphics, data, art etc, I know everything there is to know about every single subject in the field; even if it doesn't pertain to game development proper.

Your infantile jab, is just that - infantile. If your weight of a person's accomplishments is based on third-party opinions, and by that admission isn't worthy of merit, that means a) you're shallow b) every single person of accomplishment - whether it be art, science, tech, movies etc - is somehow demoted and stripped of their knowledge and experience because the reception of their works, as it pertains to a system of metrics that is devoid of factual data, is subject to being rigged etc, is worthless. I'm sure all the scientists, engineers, actors etc all of whom have had some success and failures, are all worthless because somehow that's a credible metric according to you.

Jabs from people like you are meaningless to me. They always were. I know who I am; what my accomplishments are, what I have gained and lost from them. A Metacritic score for a game (or any creative work for that matter) bears no relevance - nor credibility - to the accomplishments or talents of the creator. Out here in the real world, success isn't a hard metric; especially as it pertains to what a "success" is. You look at a Metacritic score (at this point, I'm laughing) for my games, and somehow in your head, the technical and creative achievements, not to mention the financial gains (obviously to have a career that spans 30+ years, across several projects and millions in revenue has to count), are somehow of no consequence because "Oh, your game got a low score! har! har!".

Chris Roberts hasn't made a game in almost two decades; was booted off a high profile one for exactly the same reasons playing out here, has a string of horrid D movies; but somehow, I'm the failure because of metric scores. Yeah, alright then.

And the fact that, in the middle of an argument that you're cluelessly out-classed in, rather than making the case for your argument, you somehow decide to devolve to the lowest common denominator that you Shitizens on Reddit and RSI forums exist in; and which makes it OK for you to needlessly make personal attacks as a way to either shut down an argument, or to detract from a losing discussion and/or position.

I'm the better person. You don't have to accept or like it, but there it is, Monkeh. You're in my house now; and here (unlike Reddit or the RSI forums where you mobs tend to shout/vote people down when you don't agree with them), we're going to engage you in a manner that not only completely strips you of any semblance of "group think", but also puts you on a platform that serves to illustrate to you exactly where your "place" is; because clearly you've forgotten.

Tell me again, how much my arguments are without merit because of an artificial subjective metric which has no relevance to anything other than to convey your inferiority complex and ignorance.

I would particularly love to know how that score somehow takes away from the fact that, since day one, my technical missives about Star Citizen have not only been proven to be correct (to the extent that everyone is now parroting the "they chose the wrong engine" line which I was the first to point out back in July 2015 in my very first blog), but to this day, they remain undisputed with regards to what they have delivered 4-5 years and $134+ million later.

I would also love to know how that metric somehow comes into play in the discussion of how Star Citizen isn't - and never will be - game, let alone one with a cohesive world that isn't "stitched".

[dsmart]

Guys asks another guy if he's been an idiot to give him money, 2nd guys says no.

I think Derek Smart brings up some good points and a number of potential technical issues that the Star Citizen developers might be or have faced.

However....I would take his assertions of 100% certainty with a pinch of salt. CIG might have developed solutions to the issues he sees, or developed their StarEngine to address them or like FD does, might be using technical or mathematical tricks to work around the issues.

I personally think development of star citizen is going better than Derek Smart suggests...but also a lot worse than CIG are letting on.

Why yes of course that's a perfectly reasonable argument to make because we can only go by what we see and experience, based on what they have developed and/or delivered. At the end of the day, I know that I am 100% right - and that's a hill that I'm willing to die on. All we can do now is wait and see what happens; and whether or not people will accept the end results, instead of dancing around them with excuses. Not to mention that the very notion of an MVP has already proven me right, in that they can't build the game pitched. Aside from the fact that, even with 4.0 (which barely has 30% what was promised to backers) claiming to be coming out in Q4/17, isn't the complete game they promised.

[dsmart]

What can't be held in memory? Millions of kilometres of nothing with some POI markers and a few low LOD images of planets and a star? I think you're wrong. The 2.5 alpha which does exactly that, only over a smaller space, (which is still around five factors of ten larger than current maximum CryEngine map), thinks that you're wrong, Elite Dangerous, that does exactly that, thinks that you're wrong.

Source: Star Ctien alpha 2.5 and Elite Dangerous, both showing that it would, and does, fit in memory.

You're trying to tell me that the 2.0 SC alpha never happened. Well I'm sorry but it did.

I'm not quite sure what exactly you are trying to argueng
As I understand Derek Smarts argument, he is saying that the game cannot be "seamless"...that it is stitched together from a number of instances and zones. That it is not one big map...because having just one big map means having to devote server resourves to tracking each and every object within it.

That could be players...it could be ships...it could be planets and moons and stations and it could be every single asteroid in an asteroid belt.

It isn't just POIs that a server needs to keep track of...it would be every single object within the instance.

I would assume clever programming tricks could reduce this number to some degree, but one of those tricks would appear likely to be instances and zoning so that the load can be shared on the servers. And of so, then there would need to be some way to stitch these instances together...two of which Derek Smart outlined.

And we have evidence that CIG does uses instances as it creates dedicated orbit instances and battle instances...so it musr have a way to stitch these together into a whole.

Either that or its developed a revolutionary new system that does away with the need for instancing...in which case it can easily license that technology for additional funding.

So I'm not ecactly sure what point you are trying to make since Toberts own description of how SC works tallies pretty much with what I umderstand Derek Smart is saying

He knows quite well what I'm talking about, but with these guys, it is completely inconceivable that I could be right about anything. Their worst nightmare is not that Chris Roberts was wrong; it's that Derek Smart was right about anything.

ps: No, they haven't developed any such tech. If they had, it would be in the game by now, instead of them using tech demos in order to fool backers into giving them money. Think about it.

[dsmart]

As for Derek Smart...I think he is too caught up in this. I think he makes some very good points about development wrt Star Citizen. But I also think he can be his own worst enemy in this debate. Too certain he is correct. Too certain that if he can't do something, noone can and therefore CIG must be wrong.

I think Derek Smart is correct when he suggests CIG is doing worse than many suppose. That they are having problems. That they may even have funding issues.

But I don't think things are as bad as he is portraying.

That's the whole point of a "debate". It's not about being right or wrong; it's about discussing the merits of the topic, the nuances within, the issues that are intrinsic to the position being held in the debate, as well as putting forth an argument which, popular or not, serves to move the discussion forward with statements which are fuel for thought - regardless of merit.

I believe that, not only am I right in every single regard related to the state of the project and the fact that they simply cannot build this game, but also that I will be proven to be right. The fact that we're still having this debate, even as CIG continues to flail, alienate backers etc, proves that point, even as it moves the discourse forward.

But neither do I think things are going as well for CIG as they say. Their new sales seem to indicate a grab for as much money as they can get. That isn't indicative of a business rolling in cash. They've had numerous delays, they've split SQ42 off in a blatant attempt to create a new revenue stream, the demos they've shown have had issues, they've allowed expectations and hype to get out of hand, they've had poor communications with their backers, we still have seen next to nothing of SQ42 - worrisome for a game supposedly due to be released in 2017 - and so on.

All these are signs which most backers, caught up in Cognitive Dissonance and Sunk Cost Fallacy, refuse to reconcile, let alone acknowledge.

But i do believe they are developing the game they want...and I think they'll get the $150 million Derek Smart suggested it would take to realise their vision.

They are at $134 million as of now; at an average of $33.5 million per 4 years of funding. Which means that, without accounting for investor amounts, bank loans etc - which aren't in that crowd-funded chart, they've already blown past my estimated $150 million mark. And if you go back to that first July 2015 blog in which I came up with that number, this is specifically what I stated:

Without disrespect to anyone, I’m just going to say it: it is my opinion that, this game, as has been pitched, will never get made. Ever.

There isn’t a single publisher or developer on this planet who could build this game as pitched, let alone for anything less than $150 million.

The original vision which I backed in 2012? Yes, that was totally doable. This new vision? Not a chance.

The technical scope of this game surpasses GTAV, not to mention the likes of Halo.

Do you have any idea what those games cost to make and how long they took?

Do you know how many games which cost $50 million to make took almost five years to release? And they were nowhere in scope as Star Citizen?

This means that they still couldn't build the game with $150 million. And if you ignore investor and bank loan amounts, which we're not privy to, and instead just go straight to the crowd-funding chart (which sources have shown evidence that it's not accurate), it's easy to see that even if (LOL!!) they somehow manage to release 4.0 by end of 2017, if the yearly funding remains constant, they would be at around $165 million by end of 2017. And with a 4.0 build that's not even 30% of what was promised. Think about that for a minute.

Also, it's now the end of 2016, and the lies about 3.0 being in dev, coming end of the year etc, have already been made known. So if 2.6 goes live in Q1/2016, with 3.0 to follow mid-year, well, guess what, assuming that CIG survives 2017 intact, we're now talking about a 2018 time line for 4.0.

[Kyrt]

I would particularly love to know how that score somehow takes away from the fact that, since day one, my technical missives about Star Citizen have not only been proven to be correct (to the extent that everyone is now parroting the "they chose the wrong engine" line which I was the first to point out back in July 2015 in my very first blog), but to this day, they remain undisputed with regards to what they have delivered 4-5 years and $134+ million later.

I'm not sure "chose the wrong engine" would be entirely fair to CIG.

Given the budget at the time, developing a custom engine was probably beyond their reach, given the budget and self imposed timeline. So one had to be licensed.

Without knowing what criteria they were aiming for, I can't say with 100% certainty that the choice of CryEngine was a mistake.

If CIG had $130 mollion four hears ago, then they very likely would have developed their own engine....and likely saved themselves a lot of time and effort.

CryEngine was probably a suitable choice for what they were planning. But you and I and everyone else knows feature creep crept in.

As such, it can't really be said to have been a mistake. But it does look like it turned into one given the time effort and money pumped in to adapt it to support the vision they now want.

To their credit, it appears they have largely succeeded, albeit significant issues such as the NetCode still remain. And such adaptions would likely have been needed anyway, regardless of which engine they chose.

But I for one question first, whether an engine primarily designed and intended for single player FPS action was entirely suited for a space sim, even with a FPS component and secondly, whether the amount of changes they needed to make anyway means a custom engine would havf been a more viable choice.

But I ain't in the industry so I can't really comment. All I can say is that with hindsight, and IMI, CryEngine looks to have been the wrong choice.

But hindsight is 20/20.

[dsmart]

I would particularly love to know how that score somehow takes away from the fact that, since day one, my technical missives about Star Citizen have not only been proven to be correct (to the extent that everyone is now parroting the "they chose the wrong engine" line which I was the first to point out back in July 2015 in my very first blog), but to this day, they remain undisputed with regards to what they have delivered 4-5 years and $134+ million later.

I'm not sure "chose the wrong engine" would be entirely fair to CIG.

Given the budget at the time, developing a custom engine was probably beyond their reach, given the budget and self imposed timeline. So one had to be licensed.

Without knowing what criteria they were aiming for, I can't say with 100% certainty that the choice of CryEngine was a mistake.

If CIG had $130 mollion four hears ago, then they very likely would have developed their own engine....and likely saved themselves a lot of time and effort.

CryEngine was probably a suitable choice for what they were planning. But you and I and everyone else knows feature creep crept in.

As such, it can't really be said to have been a mistake. But it does look like it turned into one given the time effort and money pumped in to adapt it to support the vision they now want.

To their credit, it appears they have largely succeeded, albeit significant issues such as the NetCode still remain. And such adaptions would likely have been needed anyway, regardless of which engine they chose.

But I for one question first, whether an engine primarily designed and intended for single player FPS action was entirely suited for a space sim, even with a FPS component and secondly, whether the amount of changes they needed to make anyway means a custom engine would havf been a more viable choice.

But I ain't in the industry so I can't really comment. All I can say is that with hindsight, and IMI, CryEngine looks to have been the wrong choice.

But hindsight is 20/20.

That's not the point here. The point - as I stated these past months - is that CE choice was perfectly fine for the smaller scope game they pitched on Kickstarter. Yes, they would still have to make certain modifications to CE3 to build parts of it. But it was after increasing the scope, that the engine core became completely out of scope. And that was the mistake that has contributed to the demise of the entire project.

And four years ago, they weren't asking for $130 million to build a game. They were asking for the $2m that it would take to build the game and for which they already took into account their CE3 license.

And I fail to see what they have "largely succeeded" at. Maybe I am missing something; so please explain it to me. For one thing, by all accounts, the netcode is the least of their problems. Even if they managed to resolve that - which it highly unlikely to give them an MMO - there are clearly a lot of other areas that are lacking; even as none of the promised game mechanics are even in yet.

[dsmart]

Monkeh, you probably want to read this. Pay close attention to Ben's statement. For context, read this first.

[Kyrt]

And I fail to see what they have "largely succeeded" at. Maybe I am missing something; so please explain it to me. For one thing, by all accounts, the netcode is the least of their problems. Even if they managed to resolve that - which it highly unlikely to give them an MMO - there are clearly a lot of other areas that are lacking; even as none of the promised game mechanics are even in yet.

While there are still isdues to be resolved, the demos they have put out do show progress in development in many areas of the gameworld. They have flight and FPS and combat amd more so it appears to me that the basic structure of the engine is workable.

More  does need to be done of course, especially if they are to include the features promised and I'll admit I'm not overly impressed with what they've produced so far but this is still what they describe as an Alpha release.

Although tech demonstrators would IMO more accurately describe what has been publicly released.

The point is that their CryEngine is now doing a lot of what they want it to do. It thus appears to me that their changes have largely succeeded in giving them what they want. I am hopeful...skeptically so...that this will allow CIG to speed up development.

I am however still very much aware that we have not seen SQ42, that they still haven't finalised their flight model,  that they missed their 2.6 date, that 3.0 isn't going to be released this year and that a late 2017 release date for 4.0 is unlikely.

But I don't expect CIG to hold to their release dates anymore. And of course, there is a huge amount of work still to do. Fixing the netcode, implementing mechanics for the various roles, adding new dhips and systems and so on. A lot of work.

So it  may be that you are correct and that I am being overly optimistic in my reading of the interviews and statements, such as that provided by Kotaku.

But as I see it, it looks like a lot of work on the engine has been finished. More needs to be done of course, and I don't seriously expect any sort of release until about 2019 but it appears the foundations of the game are largely in place.

My opinion anyway

[Mehlan]

"While there are still isdues to be resolved, the demos they have put out do show progress in development in many areas of the gameworld. They have flight and FPS and combat amd more so it appears to me that the basic structure of the engine is workable."

 I'd offer/suggest... consider those 'demos' to be vaporware until such time as whatever they claim 'exists' in those 'demos' is actually patched to the live 'server'.     

  Consider the 'demos' of  Star Marine 2015 and where we are now, well over a year later.     


  Heck consider Arena Commander/DFM...   'delayed a couple months' which in reality turned out to be 5+
   It was delayed to produce a 'multiplayer experience' where people could fly and fight with their 'own 'ships'.
   What was released, was a single players vs Bots... and 2-3 months to 'open' multiplayer to all the backers.
      IF DFM was really so 'close' in december and really meant to be only a 'couple months', then why wasn't the 'staged mp access' mentioned @ that time?     

     Recall what we were 'told' about the SQ42-illfonic by CR vs what we found out in the Kotaku UK article.

   Reality, CIG & Chris have not been all that 'open' in this 'open development' of theirs.


   The only 'code' or functions that you can count on, are what exists on the live server.  The rest is just a carrot on a stick.