Why Odyssey has so many issues? The problem is the game engine
- Thread starter MrShinyTorterra
- Start date
Epic, Crytek, DE, Croteam, DICE,... doesn't rewrite their engine with every iteration. It's the same code
FSR is done. The game already included an upsacler and they just added CAS to compensate for the blurring of the upscale process. But I don't really see how this would improve the quality of the graphics (like DLSS). It's more solutions to allow companies to invest less in optimizations.
Vulkan and Dx12 are not supported by the Xbox 360 and PS4, which would require support for several APIs within the engine. How to say...
Given the current state of the game, I highly doubt this will run on the toybox that is the 360 in any capacity. Unless they like cinematic 24 fps.Epic, Crytek, DE, Croteam, DICE,... doesn't rewrite their engine with every iteration. It's the same code
FSR is done. The game already included an upsacler and they just added CAS to compensate for the blurring of the upscale process. But I don't really see how this would improve the quality of the graphics (like DLSS). It's more solutions to allow companies to invest less in optimizations.
Vulkan and Dx12 are not supported by the Xbox 360 and PS4, which would require support for several APIs within the engine. How to say...
That's why they are looking on how to properly optimise the game (mostly in settlement and station) and how many time it would require to include it. Well... i hope. If it is to lay us another pile of sh|t and cheat with the tools tech to ensmoke us one more time, not even in dream.
I was here for the Horizons launch and the "improvements" so don't have to tell me twice. At this point I'm just watching, as I refuse to play at 30fps, or to downgrade my internal resolution to something from 2000.
From what I understood, their new mesh system somewhat removes the need for multiple levels of details as has traditionally been done - although how this stands up outside of carefully curated tech demos is yet to be seen.
What people seem to be misunderstanding is what impact this will have on the development cycle. There's a common idea that it turns game development into a plug-and-play style affair, just using the new lighting and geometry systems provided by the engine. What this notion fails to account for is the fact that video games are not movies. If you just had to drop some assets into a scene to make it look pretty and have a lighting artist fix the indoor areas, then sure, UE5 would basically sort most of that for you, but what about everything else? I mostly blame Epic for this, with their approach of jamming as many buzzwords into their presentations as they possibly can, and then refusing to explain what any of them mean. For example, what exactly does "virtualised micropolygon geometry" mean?
UE5 will no doubt have a large impact on the artistic side of game development, but it's not some magical solution that will make every other engine obsolete. The faster people understand that, the better.
They are, it seems. The engine, the code, the developer turnover exacerbating the problem...
I think even with the original core devs well versed in the code they'd be limited in what they can do because of the engine.
I did enjoy the replies from posters claiming that the OP is nonsense. There are clearly issues with getting the cobra engine to manage Odyssey which might explain why they released it in it's current state. It was either cobbled together last minute or they had been plagued by performance issues throughout development and just gave up trying before release.
You can't have proper conclusion if you don't use the proper facts. No matter the logic behind. The OP's fact are basically hearsay. They are not facts.
I do not know if the Cobra engine is unfit or not, and I'm fairly certain nobody here knows.
What I know is that Edo was rushed. And any game, on any game engine, will be bugged and perform poorly if it's rushed.
I wish they had done this when making Odyssey, since they were rebuilding the game from the ground up anyway.
Problem with Unreal - licensing fees. I'm not sure how much it would cost year to year, but I'm betting it's more than Frontier wanted to pay on a non-subscription game. Pity. I bet it would've looked glorious. It's even possible Epic might've cut them a deal because of the other groundbreaking stuff (1:1 scale milky way) ED does... or maybe not.
Please...
1/UE is terrible at large scale stuff. ED is all about large scale
2/UE is less flexible than an in house engine, unless you are epic, I guess.
3/
Sees the thread is still spouting rubbish, and leaves...
Simpler explanation because it is absolutely typical of software development in general: developer writes code with comment "will fix in performance pass later". "Later" gets pushed back for "reasons" (which can range anywhere from "oops, forgot" to "management declared these new, surprise requirements higher priority"). Release day arrives ahead of "later". That performance pass was always going to be a problem--it was put off because it wasn't going to be quick and easy--and the intervening time and code changes only ever make it worse.
I mean we can cook up all kinds of theories about why things are what they are, but this situation isn't remotely novel.
which is all well and good but I'm not in a court of law or science lab looking for hard facts, I'm quite happy with a bit of hearsay etc. I'm quite content to consider it's likely a number of problems;
management
staff
game engine
resources
fixed deadline
Some of the above would be likely causes for a rushed release, lack of integration with ED (EDO feels very separate) and a potential lack of content, and some would explain the dreadful performance, delayed console release, lack of official VR support and no plans to further develop VR - the engine can't handle it.
It was rushed because the fiscal year was ending a few days later. I think that's by far the most plausible explanation. With EDO numbers in the balance, they were able to push a "growth through internal license only" which is very good for shareholders.
Any game that is rushed like that will end up full of issues. And that kind of issues takes a while to fix, not to mention being harder to fix in a live environment. VR being dropped is simply because they focus on the "essential" (IE having the game run, and release it on console).
Also, they probably shifted some team around already, since they are working on JWE 2 and a warhammer game, both big licenses and money making.
I'm frustrated to. I want the game fixed. But throwing everything and anything until something sticks is not the proper way to do it.
Unreal Engine is bad at large scale stuff? That's news to me. I'm not disagreeing... I just find that surprising. I really don't keep up on all the game engine details like I did 20 years ago. I was just under the impression UE was the overall best... but ok.
However, it is curious that Star Citizen is switching from modified Crytek to UE... but then, one should never take anything that game does as "the right way to do things". lol
Remake in a new engine? No. Maybe try to bring the engine up to snuff? DX12, Vulkan, FSR, DLSS, all these could help alleviate the current problems.
I mean even ETS2 is getting engine upgrades under the hood, but somehow they manage not to break the game
Sure, and that is the way they should go, not a complete remake like some suggest.
They switched to Lumberyard years ago. It's still a Crytek based engine, but they've done extensive rewrites to it.
