Game Discussions Star Citizen Discussion Thread v12

They answered this too, DLSS and FSR yes, Vulkan is about 70% done, Gen12 is what they call their own propitiatory API, its not DX12, They started working on this in 2014 before DX12 and Vulkan existed, they felt they needed their own propitiatory API because DX11 was far too limited for what they wanted to do, parts Gen12 are actually already running in the game, they continue to develop it but the focus has switched to Vulkan.

Ray Tracing, unlikely, they are not ruling it out but the game is far too big for Ray Tracing to be viable, the hardware capabilities need to improve massively for it to be viable for Star Citizen, IE by many multiple factors.

They are instead working on their own propitiatory lighting and shading system, including a much better screen space reflections system.

Different names, same issues though Trep. Some gains and efficiencies. Some new shiny with costs ¯\(ツ)/¯

They're naturally seeking to stay relevant. But they're also naturally moving at a slower rate than you'd see in a dedicated engine refresh alone, due to shonkier in house tools and working with a live build etc. Always destined to be behind that curve over time, and dragging beleaguered artists in their wake...

But the key points is: There is no magical saving out there that suddenly accommodates all the crazy stuff CIG want to do. (And again, the engineering fella doesn't claim there is. He thinks it's up to the gameplay guys to magically streamline everything at their end ;). While adding Jav interiors and sandworms or what have you...)
 
On the frontier forum. Never said that no console gamer wanted to play EDO on PC, just some of them.

Sidenote : it's really stupid from FDEV to announce the end of console development without giving in the same time a solid plan to transfer all assets, reputation, engineers, etc from console to PC.
No argument from me.
 
SC supports neither Mantle nor DX12. Speak about the game, not fantasies.
Unfortunately, this does not work with SC. Alpha game in very active development, you'll have to talk about "fantasies". And let's be real, the thread would be insanely dry if we just talked about the game.
 
and yet you play it quite a bit. interesting.
Is that interesting that I periodically want to look at the progression of a thing i spent a few hundred dollars on years ago ? Maybe so then, also helps me understand what's working in there and what's not.
 
Last edited:
He only said optimisations must be made by every dept involved in the game ... like in any game in development 😅

He says way more than that:

The one part that I have no control over is just the rest of the engine in the game code, which we talked about earlier... The particularly challenging area is a game where we get a ton of players or, you know, a city and whatever else on screen. Or things like Expo halls sometimes should be challenging like. This is where we, uh, where the other rest of the company will have to take their, or they are taking their, type of part of the problem.

Here's the earlier conversation he's referencing. (Mentioned in the previous breakdown I linked to :))

And when it comes to Xenothreat that will be the game code, there will be a lot happening there, there'll be a lot of lasers firing, a lot of AI, there'll be a lot of all sorts of things happening, and the optimization there is largely going to come from the game teams, I think. Um, obviously, as we're moving as much as we can from the rendering side of it from the graphics, the CPU cost of that will free up more time for them, uh, but... ultimately I think we'll get to a point where we will get better, but then it'll be onto their teams, to type of, then take the mantle and push it further.

TLDR: Ultimately, if big intensive gameplay scenarios perform badly (as they do now), the game teams are going to have to figure that out...

REMINDER: The topic under discussion was: How does CIG add all that they plan to add? When as it stands they struggle with what they've got...
 
CIG are not inventing DX12 they were creating an alternative to DX11 because it was too limited for what they wanted to do, CIG are not the only ones who felt like this, the CPU - GPU vendor AMD created a brand new Open Source API in 2013 to replace DX11 for the same reason, it was called Mantle, it was later given to the Khronos group (creators of OpenGL) who renamed it "Vulkan"

AMD did partner with a bunch of game developers in 2013 to promote Mantle, DICE was one, CIG was another.

I think you're missing the sarcasm in the post. How CIG seem to love setting themselves crazy goals and then having to reinvent the wheel to achieve it, but by the time they achieve it, the world has moved on.
 
CIG are not inventing DX12 they were creating an alternative to DX11 because it was too limited for what they wanted to do, CIG are not the only ones who felt like this, the CPU - GPU vendor AMD created a brand new Open Source API in 2013 to replace DX11 for the same reason, it was called Mantle, it was later given to the Khronos group (creators of OpenGL) who renamed it "Vulkan"

AMD did partner with a bunch of game developers in 2013 to promote Mantle, DICE was one, CIG was another.
I am waiting for CIG to build CPUs and design network protocols to handle QuePasa and Smashing.
 
I think you're missing the sarcasm in the post. How CIG seem to love setting themselves crazy goals and then having to reinvent the wheel to achieve it, but by the time they achieve it, the world has moved on.
In this case, reinvent the wheel = get rid of dx11 for the last gen of Vulkan.
They are not reinventing but updating the renderer to use a new API.

Something that FDEV should do with dx12 as the console version is gone (perhaps they do it in secret, we can just guess as they don't have an explicit roadmap).
 
Last edited:
Back
Top Bottom