It used to be a modified version of CryEngine, it's now an equally-modified version of Lumberyard.
The Star Citizen Thread v5
- Thread starter Brett C
- Start date
- Status
It used to be a modified version of CryEngine, it's now an equally-modified version of Lumberyard.
So LY is identically modified to CIG's frankenengine? Every bit of the code is exactly the same so all they had to do was load assets and bam there's star citizen in LY? How does that happen?
Sean Tracy or someone claimed they'd modified over 50% of CE. They made a huge deal out of 64-bit conversion. If this stuff were true, how do you go about just switching to a different branch of the original base engine without a huge amount of work?
Regardless of what's happened, the fact of the matter is CIG has spent 5 years at least claiming they were doing a ton of engine modification to get things working how they want. Now apparently it's not good enough, but this other iteration of CE is? And it's no work to port over what's been so far?
Shoving this news out two days before Christmas is not a good look either. Friday afternoon releases are the first trick in the PR handbook for handling bad news, around Christmas is even better. If they were that confident about this move it would have been announced a year ago. So much for open development.
Anyone who can't see the glaring warning signs at this point is blind.
Last edited:
I have to agree. I understand Ben's position, but it doesn't add up with importance and urgency of CIG own modifications in engine. Either those aren't as big or unique as they claim, or something else is going on.
Anyway, back to poking funny bits in my PC for Christmas Eve.
Essentially this. That's why it is useless to care too much about it. Let them have it.
I'm not sure what I'm screwing up in the explanation here...
It's not about how far SC's engine is from base CryEngine, it's about how near base CryEngine is to base Lumberyard. Switching to Lumberyard doesn't mean undoing any of CIG's changes.
Imagine you do this the dumb way, generate a massive patch containing everything that CIG has changed relative to CryEngine, then apply that patch to a near- (or completely-) identical codebase, the patch would apply with very few (or no) problems. Even better, you don't have to do it that way, you can calculate the differences between CryEngine and Lumberyard at the appropriate point in the versioning history, and apply them to the SC codebase directly. For the majority (maybe all) of those systems, that'll be zero changes, so you go from "5 years of modifications on CryEngine" to "5 years of modifications on Lumberyard" in a pretty easy patch.
I mean, the network team probably have more work, because they want to update forward enough to get all the AWS stuff, presumably. When I get in after Xmas I'll probably have a dig through their changes to see if they spotted any stuff to fix on the rendering side that I agree with, I imagine other teams will do likewise
I'm afraid you can't make a silk purse out of a sow's ear, no matter how much money you throw at it, but I'm sure you already know that at this stage.
Last edited:
OK, so had a "quick" go of 2.6.
Is it better than 2.5? - short answer, yes.
Things what I did like:
Loading times - much better
New game interface - choosing what to do (PU/Star Marine/Arena Commander) - big improvement over 2.5
Ship customization for AC - much much better than before - having a series of screens to configure the ship is much easier than before - still needs work, but a good step forward.
Flight Model for AC - seems much better to me - more "playable" than previously - again, needs polish, but a good step forward, I now feel like I am in control (I understand this may be subjective and down to personal preference)
I've only had a go at Vanduul Swarm in AC
Not tried PU yet
Not tried Star Marine yet
Controls customization for my Warthog is still pretty sucky - I need to go over it a few more times before it's just "right"
Some bugs in AC with HUD (not showing weapons/shield/power status, and also problems with view after redout/blackout occurs)
Radar is obscured by helmet chin bit.
Still no TrackIR support
(one of my personal bug-bears)
Is it better than 2.5? - short answer, yes.
Things what I did like:
Loading times - much better
New game interface - choosing what to do (PU/Star Marine/Arena Commander) - big improvement over 2.5
Ship customization for AC - much much better than before - having a series of screens to configure the ship is much easier than before - still needs work, but a good step forward.
Flight Model for AC - seems much better to me - more "playable" than previously - again, needs polish, but a good step forward, I now feel like I am in control (I understand this may be subjective and down to personal preference)
I've only had a go at Vanduul Swarm in AC
Not tried PU yet
Not tried Star Marine yet
Controls customization for my Warthog is still pretty sucky - I need to go over it a few more times before it's just "right"
Some bugs in AC with HUD (not showing weapons/shield/power status, and also problems with view after redout/blackout occurs)
Radar is obscured by helmet chin bit.
Still no TrackIR support
(one of my personal bug-bears)
Yeah that's pretty sucky.
Whether the modifications are good or sufficient wasn't the original topic of discussion.
I can't fight your characterisation of CryEngine as a sow's ear, but it was still the best option on the table when the project started.
Well, it's not their first lie... I mean, first little misunderstanding.
No, the best option as has been borne out thru this whole mess would have been a in house developed engine, but you also need someone competent to manage the project. It's shagged on both counts.
Aye, like there were never any little misunderstandings about Elite Dangerous!
Yeah, there was one, 3 years ago, how horrible of them.
Edit: The funny thing is, FD can't lie about the number of paying customers because of them being a public traded company.
Last edited:
I still laugh when people suggest that.
Even knowing from day 1 what the scope and budget would be like, even with razor sharp management and an experienced team who had all worked together before... no. Not unless you wanted three years of pre-production just making the editor, debug interfaces, thread management, maths libraries, all the other boring vital junk. And for your troubles, you now have an engine no one else is publishing updates for, and none of your new hires have any experience using.
Even if you're planning to ship-of-theseus your way to a whole different engine with none of the original code, at least having that engine to begin with lets you test and profile each newly replaced system as it's done. While it's being done, even.
So basically you're saying that they were winging it from day 1 and had no plan, nice. How does any project ever come to fruition with such an outlook.
You have never played blackjack it seems
Not with other people's money.
It was horrible, yes. Also dishonest, and utterly trust destroying.
Wasn't horrible, was dishonest, wasn't trust destroying. It's not lying about the number of your customers, after all.
- Status