I think i'm pritty clear about the glassdoor reviews not being 100% reliable but there are 2 other points in the thread you seem to have skipped.
Why Odyssey has so many issues? The problem is the game engine
- Thread starter MrShinyTorterra
- Start date
The one thing nobody is talking about... The new upcoming game, FAR: Changing Tides. Brings it to 4 games now I think. That's an awful lot of games to spread your experienced game engine devs around. Maybe the talent is a little thin? I'd say Odyssey is a showcase for it. over 2 years in "full" development and we get this?
OP makes good points. I
There is a reason why games are not written in assembly anymore, it simply does not scale enough to write complex software without becoming too difficult and error prone. It may, however, make sense to write performance critical parts in assembly. Not all code needs to be blazing fast if it isn’t run too frequently. Same principle applies for scripting languages in a game engine.
So, interesting take back i have from those glassdoor reviews is actually the one that says the Engine is hard to work with... unless you are a programmer.
And here we have a taster that FDev are actually doing programming, not simply scripting everything, there is a difference between actual programming and scripts... one takes a lot more effort than the other, they have advantages and disadvantages too but typically actual programming is a hell of a lot faster and solid than scripting.
What that review says to me is the following "I suck at actual programming because i probably only learnt python, maybe some perl or something and I had to get my hands on some actual C++ (or the like) and found it difficult... because i found it hard, FDev's engine sucks"
See lots of different ways to look at it.
If we want to be a bunch of pedantic ..... its also a true statement that if you know what you are doing, the fastest way to do it is to get as closed to the metal as possible and just go full assembly programmingThe higher the level of programming, the worse performance you often get.
There is a reason why games are not written in assembly anymore, it simply does not scale enough to write complex software without becoming too difficult and error prone. It may, however, make sense to write performance critical parts in assembly. Not all code needs to be blazing fast if it isn’t run too frequently. Same principle applies for scripting languages in a game engine.
The graphic engine have been modified for Odyssey. See the Horizon version as 1.0 and the Odyssey one as the 1.5
Is that the game I just received a spam email about, telling me to enter a contest to win BIG, except the contest (2nd contest in a row) is open only to europeans, and americans, (of which I'm neither)?
Tossed it.
Won't follow.
Sorry, not sorry for semi-rant.
The leaked roadmap is even less useful - it doesn't say why JWE "can't be fixed". If it was something to do with the engine, it wouldn't be any better in a sequel either, so it's probably bad implementation decisions elsewhere if they think starting over will help. That's evidence against your hypothesis, not for it.
The quality of the game in general ... well, "everyone agrees" that Unreal Engine is really good and there have been some absolutely terrible games released with that. You can program bugs into just about anything. The main issues with the game are not "engine" ones -
You've proven that Frontier have issues developing Elite Dangerous, definitely ... but a quick look at the game would show that, it hardly needs proving again. There's absolutely nothing to tie this specifically to the "engine" rather than to the bits built on top of the engine. Planet Zoo ... same engine, they don't seem to be having serious scheduling or quality issues with it.
Looking at the top 30 open voted bugs ... there's some graphics issues and some networking issues which might be the engine (or might not), but over three quarters of them are design or implementation issues which have nothing to do with the underlying game engine at all.
Even if there are serious issues with the Cobra engine, that doesn't mean a different engine would be better:
- Elite Dangerous: custom in-house engine (released)
- Star Citizen: uses a commercial engine, heavily customised (several years behind schedule)
- No Mans Sky: custom in-house engine (released)
- X4 (and previous iterations): custom in-house engine (all released)
- Dual Universe: a commercial engine, but a really obscure one that's specifically designed for space-scale environments (years behind schedule but might be released soon-ish)
There's a definite pattern there. I can't think of a successfully released open-world vaguely-like-Elite space game (constrained combat-only games, yes) which has used a commercial engine ... and Dual Universe would barely be an exception if it does get finished since they're not using a game-focused engine for it.
The problem is that the issues with producing a game set in space - where objects a metre away and objects a few AU away can both be relevant at once - or where entire planets need to be generated and simulated at once - is that this is a set of problems way outside the scope of conventional mainstream game engines. They're not optimised at all for that use case - because there's about ten games total which would need it, and optimising for that makes all the Yet Another FPS/RPG/RTS games much less efficient, which is not, commercially, a good thing for engine writers to do.
Which give these a bit of credibility it's that tt still matches quite a bit with the clues we already have (the inhomogeneity of the renderings of the pipelines, the inability of FDev to comment on the state of their graphics engine, their confession that they would not have launched the game if they had known of its real state, that they are not serene about the console release,...). There is no absolute proof, but we can't ignore that we felt that there was somewhere a loss of control of the game.
Unigine. It's not more specificaly design for space-scale environnement than any other commercial graphic engine
And there is also the abandonned Grav which run under UE3.
Last edited:
Not specifically for space-based games, no. But the 64-bit coordinate system it has is almost essential for space games and rarely necessary for anything else (which is why most commercial game engines don't). Whether or not it's suitable for the rest of the complexities of space games, I guess we'll see in a couple of years, but that's definitely a big advantage over most of the others.Unigine. It's not more specificaly design for space-scale environnement than any other commercial graphic engine
And for Grav ? How it handle it with UE3 ?
Since they abandoned the game before it got out of Early Access? Probably not very well, though there are plenty of other problems with writing space games which even a good engine can only go so far.
Certainly fits the pattern of "space games made with commercial engines fail entirely" / "space games made with in-house engines are heavily criticised on release but eventually fixed later" anyway. (Whether Odyssey continues that pattern or is actually unfixable we'll find out in a few years...)
The underlying pattern is that no-one sensible would set a game in space to start with. Space's defining characteristic is that it's really big and really empty, neither of which are particularly conducive to interesting gameplay. But no fancy game engine is going to fix that, either.
I only addressed the points worth addressing.
The only other point I want to address in this thread is the nebulous calls to consolidate game development into only a handful of engines (Specifically Crytek and Unreal) which would be the death of videogames. The best games on the planet consistently are developed with first-party/in House engines, because a "One tool to rule them all" approach lacks the freedom required to address very specific game design issues without overhauling the engine to a point where it's unrecognizable anyway.
Centralization of assets is never the answer, it's only a disaster waiting to happen. The real solution is to train more guru-level programmers, instead of script kiddies using Unreal Blueprint to spurt out some garbage on steam just so they can put in their linked in "Titles Shipped: 1"
Last edited:
The thing about the Cobra engine is, we don't really know what kind of a workflow it has. It is a custom egnine by FD, not a multipurpose engine like UE or Unity. We don't really even know how it follows the common practices of other game engines, i.e. in what comes to the art pipeline. It might do things in a similar fashion to UE or Unity, or it might have a pipeline unique itself.
What I'm saying is, that as it is a custom engine, it most likely requires the people using it to familiarize themselves with it. It might even be, that any existing knowledge with other engines helps very little with it.
Is the Cobra engine having technical issues? Obviously so, Odyssey is proof of that. Is the Cobra engine harder to use than other engines? Might be, but as only FD developers use the thing, we don't really know. All we can do is speculate.
What I'm saying is, that as it is a custom engine, it most likely requires the people using it to familiarize themselves with it. It might even be, that any existing knowledge with other engines helps very little with it.
Is the Cobra engine having technical issues? Obviously so, Odyssey is proof of that. Is the Cobra engine harder to use than other engines? Might be, but as only FD developers use the thing, we don't really know. All we can do is speculate.
Yup, the engine is outdated, the net code is is dated, the graphics are dated. Everything is dated and they're trying to shoehorn what they can into a dated system. Its at its limits. Hard to expect any further development on this title.
Its time for a sequel not more DLCs.
Its time for a sequel not more DLCs.
I don't doubt that fdev will release a working version the latest season, unfortunately they took the step to release an alpha as a full priced dlc. It's that bit I can't understand surely DB should have realised releasing a broken dlc not only would cause issues with the game but completely tarnish the reputation of the company he's the CEO of.
As they say money isnt everything
As they say money isnt everything
A post that needs a table of contents?!
Wow.
But ok, the engine is a problem. Great. FD should start over? Not going to happen. Same reason CIG will never start over on Star Citizen. Too much work already invested.
As for a sequel, i think that could kill the franchise. Think of all the people who bought lifetime passes for this game. Those who have invested recently in Odyssey. The only way to placate those people would be to offer the sequel to them for free, which would kill profits.
There were some of us who hoped the reason Odyssey was taking so long was that FD were using the time to rework the game engine... it appears that never happened.
But engines can be improved over time, and that has to be the way for FD to go.
Wow.
But ok, the engine is a problem. Great. FD should start over? Not going to happen. Same reason CIG will never start over on Star Citizen. Too much work already invested.
As for a sequel, i think that could kill the franchise. Think of all the people who bought lifetime passes for this game. Those who have invested recently in Odyssey. The only way to placate those people would be to offer the sequel to them for free, which would kill profits.
There were some of us who hoped the reason Odyssey was taking so long was that FD were using the time to rework the game engine... it appears that never happened.
But engines can be improved over time, and that has to be the way for FD to go.
Well, the pass is for the games lifetime, not the players. All games end at some point.
True, but you don't think the lifetime pass holders will feel a bit robbed if FD were to dump Elite 4 and start on Elite 5 at this point?
And if the reasoning was to make a new game engine (or use a different one) rather than working on the engine they have.
We'd also be back at square 1... unless some people here are imagining that FD would suddenly make a game equal to what we have now in just a few years? FD won't magically get better at developing stuff. They'd still take a few years to effectively give us Elite Danerous (redux) 1.0 and then more years to add landable planets, etc.
I suppose at least with a second attempt they could not do Powerplay and CQC
But this being FD, i suspect we'd just get a whole load of new things that don't work well.