Making a puzzle game in CE is very easy, in fact people have made all sorts of things like sidescrollers etc.
Making a simple space shooter game like AC (and before you call me a hater, of course AC has some very nifty damage effects and the whole thruster articulation, but right now it's really just a space shooter where all you do is blow up wave after wave of AI ships, or play team deathmatch) is relatively easy, because you're still just working within the confines of the engine - i.e. your "game space" is limited to the maximum permitted by the engine.
Adding double precision touches everything. I've already told you, it's not just the game engine itself, but you also have to touch the editor and the toolchain. The editor alone is very fiddly to modify. Alternatively you could keep the editor single precision but then you'd need to have CE support both single and double precision.
And ah yes, I recognize this mindset. "If we just throw enough money and resources at this situation we've got ourselves into, we're sure to overcome it." - I've watched a 200+ million-euro project go down the toilet. No amount of money and people helped them overcome the very basic erroneous decisions made at the start, because they were dead set on those. And they had more money and people at their disposal than SC and ED have combined.
It's great that you still believe in their ability to deliver on their promises. But you see, while I totally support you there, I don't share that optimism. I'm gonna go out and say it:
I would totally support SC dropping CryEngine today, and I firmly believe that while it'd be a setback initially, it'd be worth it down the line.
Also, it is coming together? Like, super late, with buggy multiplayer that doesn't even HAVE any of the complex stuff like, say, docking at a space station? I mean heck, Elite may have problems with their netcode too, but they're like a kitchen team working on a slightly botched dessert in a big menu with all kinds of starters and main dishes already done and smelling awesome, while Star Citizen just finished a mug of milk in the microwave.
I'm gonna say this one last time and then never again because you apparently ignore what I wrote before:
Why would they test the single-precision collision detection? It's already built into CryEngine and works. They gain no useful information about double precision collision detection. They have to rework everything from netcode, collision, physics, etc. including all the ship performance values.
Also, there's enough literature on that subject. It's not like SC is the first game ever to plan to do this, even though they'll no doubt be happy to claim they're the first space game ever to pull it off.
What proper space physics? That things have inertia and retain their momentum unless force is exerted? Wow that's super detailed, I'm amazed. Apart from the thruster articulation (which is pointless but looks cool) and the damage effects (which for me are the absolute highlight of AC and where they wipe the floor with ED right now), there is nothing detailed or amazing.
That's an assertion you can't prove, just like I could not prove mine. I can only point to past projects that took on more they could chew, as well as point out the complexity of the task, having access to the engine code.
Self-validating statement is self-validating. The fact of the matter is, CryEngine is one choice among many, with it's own benefits and drawbacks. Saying it's the "best" is an apples vs. oranges thing.
Well I'm not so sure CR knows what he wants. He wants everything plus the kitchen sink, but there seems to be no coherent plan to go about it. And no, believing doesn't make it so.
But even if CR had a clear, precise and detailed idea of what he wants this game to be, that doesn't mean anything with regards to engine choice, the ability of his team, or how this project turns out to be. The rain doesn't go away just because you really want it to.
I'm looking at the source tree right now and quite honestly there's not a lot of stuff I'd reuse. CryEngine is a value proposition only if you take it for what it is, and use it to best effect. As soon as you meddle with the inner workings, there's a real danger that you generate more work than if you built your own.
If you disagree with that last statement, fine, I probably can't convince you. I already poked you about certain interna of CryEngine in an earlier post, which you then conveniently ignored before repeating your same statements again. So I'm not going through this again.
Also, CryTek has their own set of problems. I doubt they can contribute a huge amount of people for long stretches of time right now.
Well said my friend I almost agree 100% with all above except the one thing and that is the dmg model.Is it good? yes it is, is it something new and spectacular no it's not.Is it better than ED?I can not claim that as I am not ED backer yet but even if is it better that is not a big impact as FM on a game play.As you can see I consider myself as a Realist and even if I am not ED backer i am not blind and don't have any problem to accept the fact that ED is by mile better in almost all of the aspects that should make a BDSSE...