Game Discussions Star Citizen Discussion Thread v12

Lovely stuff is lovely stuff. (In the space sim in my mind ships slice spray across the water as they pass, for sure ;))

But the question for me is: Isn't this exactly the kind of 'duplicate the real world' intent that is impossible in to achieve in totality, and which should be faked wherever possible? The type of bonus physics load that you have to cull down the line when your computer is panting over the wildebeest dappling the veld, or whatever? When the full feature set is in. Or else risk seeing your game run like garbarge... (I can already imagine ungainly judders as the river figures out it's supposed to have wash running over it, 2 seconds after the blast ;))

Essentially: Isn't this the sort of thing you do last? ;)

There is a client hardware budget limitation they work towards, they talked about this in the last Calling All Devs, the difference between what they would like to do and the reality of what the hardware can reasonably render.

Having said that CIG are different to most game developers in that most game developers aim for consoles, CIG are making a desktop exclusive game where the hardware is much more powerful, So Star Citizen will never run on an Nvidia 1050TI, even with the lowest settings, but at a minimum it will run on a GTX 1070 or even a 6GB GTX 1060, that's a mid range GPU from 2016, i don't think that is unreasonable.

Its also ever evolving, it looks very different today vs how it looks now, it is also a live test environment, they are testing the limits of what they can do graphically, as an example the planet tech is on its 3'rd major face lift, the overall graphical quality is improved little by little from patch to patch to patch.... some time its hardly even noticeable, but it is there.

This is what it looked like in 2016, its Port Olisar, that's the Gas Giant Crusader and the ship is the Agis Gladius. i liked the way it looked and sounded back then, it had a certain charm to it.

Source: https://www.youtube.com/watch?v=NJ2hm09NHho
 
Having said that CIG are different to most game developers in that most game developers aim for consoles, CIG are making a desktop exclusive game where the hardware is much more powerful, So Star Citizen will never run on an Nvidia 1050TI, even with the lowest settings, but at a minimum it will run on a GTX 1070 or even a 6GB GTX 1060, that's a mid range GPU from 2016, i don't think that is unreasonable.
GPU - DirectX 11 Graphics Card with 3 GB Ram
Someone should tell CIG :LOL:
 
There is a client hardware budget limitation they work towards, they talked about this in the last Calling All Devs, the difference between what they would like to do and the reality of what the hardware can reasonably render.

Sure. But does Chris care when he's issuing these missives? It seems more like the devs have to make the feature regardless. And then strip it out, or pare it back, if/when it hits the cap.

Working like this can get fairly wasteful. Not only in dev time spent on 'simulating everything' to no end, but in the sense that you can end up with clunky leftovers in the code. Poor old John Pritchett's physics sim for the ships seems to be sitting their spinning away to no end, as designers override its outputs for numerous pragmatic reasons, for example.

Having said that CIG are different to most game developers in that most game developers aim for consoles, CIG are making a desktop exclusive game where the hardware is much more powerful. CIG are making a desktop exclusive game where the hardware is much more powerful, So Star Citizen will never run on an Nvidia 1050TI, even with the lowest settings, but at a minimum it will run on a GTX 1070 or even a 6GB GTX 1060, that's a mid range GPU from 2016, i don't think that is unreasonable.

Two things:
  • That's now. With only, what? 10% of the features implemented shall we say? So load is likely to go up.
  • They're also attempting to keep pace with the visual side of the coin, as you mention
The question becomes: How narrow can they make the window of playability really? Would running acceptably on the top 10% of machines be a viable proposition? The top 5%?

For an MMO?


Its also ever evolving, it looks very different today vs how it looks now, it is also a live test environment, they are testing the limits of what they can do graphically, as an example the planet tech is on its 3'rd major face lift, the overall graphical quality is improved little by little from patch to patch to patch.... some time its hardly even noticeable, but it is there.

Oh man I am well aware of both their constant push to keep pace, and the absurd amount of re-work it involves for the art teams. (Because they already pushed to keep pace with 2014. And 2016. And 2018. During an alpha...)

And that's on top of all the crazy re-hashing imposed by game design changes being made downstream, or finally being put into practice. (Oh did we make Lorville's cityscape too small for gameplay to occur inside? Whoopsies. Oh did we make the Retaliator corridors too small for AI to navigate through? Oh my...)

---

The thing is, in total, this all just plays into the narrative that Chris actively wants to push beyond the limits of 'now'. To deliberately go past them and overload his existing systems. On the understanding that that's groundbreaking. And that the future will save him.

But as the dev there notes, if you don't at least pick an end point and stick the landing, you still end up behind the curve. And still over your limits. Just getting lapped by the latest next console gen too ;)
 
Last edited:
This whole thread is quite good...


"SC is literally progressing amazingly fast. It's so weird to see people complain about the time it takes to do this stuff.

Everyone needs to relax and realise it's going to need another decade and another $500m to get anywhere with this, that's just how game development works."
 
"Yeah it's weird to make up some arbitrary number and then complain that it took longer than that.

Like "ooh I decided it should take exactly 13.4 years to make a single player space game and so as soon as it takes 13.5 years I'm going to tell everyone how ridiculous that is."

Cathedrals in Europe in the middle ages often took over a century to build, SC is no less ambitious, if not more so."
 
@VR Golgot I think you are correct in pointing out the absurdity of it taking so long they are having to go back and rework things to suit new objectives creep for the game, its not just that the retaliator interior is too cramped, almost every ship will need a rework to some degree for the new physicalized components system, Hurston is getting a rework, at some point the entire ArcCorp Planet will need redoing from scratch as its based on older no longer compatible tech, they even said this in one of the Calling All Devs.

But its not just a case of pushing the engine further and further beyond its limits, they are employing the latest technologies, for example moving the entire particle system off the CPU and on to the GPU, to use one of many examples, the engine its self is continuously upgraded, there is barely anything at all left of the original Cryengine.
 
There is a client hardware budget limitation they work towards, they talked about this in the last Calling All Devs, the difference between what they would like to do and the reality of what the hardware can reasonably render.

Having said that CIG are different to most game developers in that most game developers aim for consoles, CIG are making a desktop exclusive game where the hardware is much more powerful, So Star Citizen will never run on an Nvidia 1050TI, even with the lowest settings, but at a minimum it will run on a GTX 1070 or even a 6GB GTX 1060, that's a mid range GPU from 2016, i don't think that is unreasonable.

Its also ever evolving, it looks very different today vs how it looks now, it is also a live test environment, they are testing the limits of what they can do graphically, as an example the planet tech is on its 3'rd major face lift, the overall graphical quality is improved little by little from patch to patch to patch.... some time its hardly even noticeable, but it is there.

This is what it looked like in 2016, its Port Olisar, that's the Gas Giant Crusader and the ship is the Agis Gladius. i liked the way it looked and sounded back then, it had a certain charm to it.

Source: https://www.youtube.com/watch?v=NJ2hm09NHho
Ahh...but do you remember this from April fool's day 2015 ;)

 
Cathedrals in Europe in the middle ages often took over a century to build, SC is no less ambitious, if not more so."

That one made me do a double take. Can't decide whether they are joking or not.

Nobody in their right mind would be ok with a game taking over a century to develop. PCs probably won't even be a thing by then.
 
@VR Golgot I think you are correct in pointing out the absurdity of it taking so long they are having to go back and rework things to suit new objectives creep for the game, its not just that the retaliator interior is too cramped, almost every ship will need a rework to some degree for the new physicalized components system, Hurston is getting a rework, at some point the entire ArcCorp Planet will need redoing from scratch as its based on older no longer compatible tech, they even said this in one of the Calling All Devs.

But its not just a case of pushing the engine further and further beyond its limits, they are employing the latest technologies, for example moving the entire particle system off the CPU and on to the GPU, to use one of many examples, the engine its self is continuously upgraded, there is barely anything at all left of the original Cryengine.

Well, thankfully its still early days. Development only really started in 2020. They are still developing the tools, but once done, development speed will really pick up. There are cathedrals in Europe that took over 100 years to build that are smaller in scope than SC!
 
Nobody in their right mind would be ok with a game taking over a century to develop.
If you can play and enjoy it more than a lot of released games during the process, it's ok for a lot of people. Like planting a forest in a desert and look at it growing. It's just better every years.

If you take the actual problem with console version of ED, the game is released but not in development anymore. I guess some of those console players will find some appeal to the idea of a playable alpha on heavy development with a lot a fun already here to play and promises of more fun every quarter.
 

dayrth

Volunteer Moderator
If you can play and enjoy it more than a lot of released games during the process, it's ok for a lot of people. Like planting a forest in a desert and look at it growing. It's just better every years.

If you take the actual problem with console version of ED, the game is released but not in development anymore. I guess some of those console players will find some appeal to the idea of a playable alpha on heavy development with a lot a fun already here to play and promises of more fun every quarter.
I can't see the sense in being OK with waiting 100 years or more for a video game you already paid for, although liking CIG's development pace to watching trees grow is very apt.

I also don't understand the logic behind thinking that people who have a game on console that is no longer being expanded will switch to a game that is not yet released, may not be for 100 years and won't be coming to console even then.
 
I can't see the sense in being OK with waiting 100 years or more for a video game you already paid for, although liking CIG's development pace to watching trees grow is very apt.
You skip the part where you can play the game. No need to wait.

I also don't understand the logic behind thinking that people who have a game on console that is no longer being expanded will switch to a game that is not yet released, may not be for 100 years and won't be coming to console even then.
That's not logic. I read it from console ED players. Some were waiting the EDO experience on console. As this experience will not come and they have no other alternative choice on console, they plan to get a PC but will use it to play SC instead of ED.
Some also talk about the fact that transfering your account from console to PC will make you loose a big part of what you have grinded (engineer) and they don't have the will to do it again and prefer testing something new with SC.
Or the fact that some guys think that it's the start of the end for ED and prefer to invest their time on something that seems to have a future.
 
You skip the part where you can play the game.

You skip the part where it's a horribly buggy alpha...

Source: https://www.twitch.tv/videos/1425424424?t=1h18m38s


Source: https://www.twitch.tv/videos/1425087868?t=2h4m46s


Source: https://www.twitch.tv/videos/1424712722?t=3h38m41s

 
@VR Golgot I think you are correct in pointing out the absurdity of it taking so long they are having to go back and rework things to suit new objectives creep for the game, its not just that the retaliator interior is too cramped, almost every ship will need a rework to some degree for the new physicalized components system, Hurston is getting a rework, at some point the entire ArcCorp Planet will need redoing from scratch as its based on older no longer compatible tech, they even said this in one of the Calling All Devs.

But its not just a case of pushing the engine further and further beyond its limits, they are employing the latest technologies, for example moving the entire particle system off the CPU and on to the GPU, to use one of many examples, the engine its self is continuously upgraded, there is barely anything at all left of the original Cryengine.

Yep, they are constantly updating the engine, and there will be savings there as well most likely. Gen12/Vulkan will free up CPU capacity etc. But for every current gen saving (forced into an older architecture), there's also every ray-tracing-style extravagance for Chris to eye too ;)

(And as the engineering guy noted recently, all of these tech changes hit the art departments along the way with the rolling bonus reworks...)

Ultimately though, the '10%' thing doesn't go away. How will that river physics (sweeping networked items along etc) stand up alongside wind physics, and animal physics, and destroyed ship physics, and so on, all in one location? What would be the first thing you'd chop, if you had to make a choice amongst all that? (It'd be the wave effects from blasts, and the ability for a box to float downstream. Surely?)
 
Last edited:
You skip the part where it's a horribly buggy alpha...

Source: https://www.twitch.tv/videos/1425424424?t=1h18m38s


Source: https://www.twitch.tv/videos/1425087868?t=2h4m46s


Source: https://www.twitch.tv/videos/1424712722?t=3h38m41s

In before: "You see bugs, I see emergent content generation". The mess would be pretty bland without all these bugs, even more for the car-crash-beholders watcher that I am :)
 

Viajero

Volunteer Moderator
As this experience will not come and they have no other alternative choice on console, they plan to get a PC but will use it to play SC instead of ED.
This still does not make sense unless your goal is to shill for SC or you were actually in CIGs payroll directly or indirectly. If a console player that wants to play EDO decides to buy a PC why wouldn’t he play EDO?
 
Last edited:
Yep, they are constantly updating the engine, and there will be savings there as well most likely. Gen12/Vulkan will free up CPU capacity etc. But for every current gen saving (forced into an older architecture), there's also every ray-tracing-style extravagance for Chris to eye too ;)

(And as the engineering guy noted recently, all of these tech changes hit the art departments along the way with the rolling bonus reworks...)

Ultimately though, the '10%' thing doesn't go away. How will that river physics (sweeping networked items along etc) stand up alongside wind physics, and animal physics, and destroyed ship physics, and so on, all in one location? What would be the first thing you'd chop, if you had to make a choice amongst all that? (It'd be the wave effects from blasts, and the ability for a box to float downstream. Surely?)
This is the key point. Building wider streets does not reduce congestion. It brings more cars. It is a well known paradox of efficiency. It has name - https://en.wikipedia.org/wiki/Jevons_paradox
 
Back
Top Bottom