-
Hey Guest. Check out your NeoGAF Wrapped 2025 results here!
Star Citizen Alpha 2.0 | The 'Verse Awakens
- Thread starter Raticus79
- Start date
Dictator93
Member
Yeah I much prefer that thanks. I don't really know what interlacing is but imo it makes the one I was watching kinda bad.
If we are talking about final game, i expect 1060 to run it at 30-40fps at max setting in 1080p and 60fps in Med/High. The problem is your CPU, 40fps is pretty much all it will be able to do.
What? Star Citizen is really that demanding on the CPU? The i5 6500 seems pretty capable in every other game today. In any case, if my PC can't keep up, I'll just run it at 30FPS and enjoy the max settings visuals. According to benchmarks, my PC seems capable of running most games at 1080/60 at max settings, though.
Star Citizen is not like most games. Its not a game that have set 30fps ceiling for 6 Jaguar cores. Its a sim, so its physics driven in every aspect of its design.
Now that i think about it, you wont probably max it at fully stable 30hz. But High should be definitely possible.
Damn, we will see, I guess. I have yet to actually even run any games on my PC.
So if I want a dual stick setup, do I go with those updated Thrustmaster sticks coming out next month or the ones he has in his video?
You definitely don't want to start with Thrustmaster's Warthog HOTAS (that's what the guy is using). It's 400$ for the stick and throttle control and you'd need two. You can find the stick alone for about 200-250$ on ebay, but it's still a shitload of money for something you don't even know will suit you. Also, you'd have to modify the second one for left hand operation.
Start with something cheap(er) and ambidextrous like the new T16000M and move up from there (or not). Eventually.
Do they sell the sticks seperately? Because that website mentions only the HOTAS setup.
AFAIK, no.
Buggy Loop
Gold Member
Been a long time since i fiddled in SC, but this is the most un-assisted flight mode right?
Still so heavily assisted :/
Ðuelist
Member
Heavily bottlenecked by servers.
i7 5930K
GTX 970 OC
1TB SSD
32GB RAM
And I get about 45-50 FPS on Medium...
For any potential future dual Warthog stick owners, this guy from a dedicated thread on SC's official forums is making left hand grips. Well, will be making them soon, still WIP. They'll be made in cast resin and will likely be available in different / custom colours.
Default black
"Avenger" variant
Looking good indeed. Read more and subscribe at Dualsticks.com to get notified when they go up for sale.
Also, TheNOOBIFIER1337's video on how he modded his stick should be up later today.
Default black
"Avenger" variant
Looking good indeed. Read more and subscribe at Dualsticks.com to get notified when they go up for sale.
Also, TheNOOBIFIER1337's video on how he modded his stick should be up later today.
Rephin
Member
I'd wager that any large ship to large ship interactions would be done via the ARGO or other similar shuttle-type craft.
Zalusithix
Member
All this talk about control interfaces got me thinking about how I need to work on my 3/4 axis throttle. On my way over to the pile of parts, however, I passed by my Vive controllers. Got me to thinking... Each of these things is technically a rock solid 6DoF controller with haptic feedback and additional flexible input mechanisms. They're also shaped very much like joysticks already. Technically I could write a program to interface with Open/SteamVR to interpret their movement and then feed a joystick emulator program like vJoy.
Dual 6DoF joysticks. The sky's the limits when it comes to how you want to interpret the movements and map input to output. The lack of centering could be an issue, but having used the controllers for a some time now, I'm relatively confident it's a solvable problem without needing physical rigs.
Controller angles for pitch/yaw/roll are relatively easy to return to the zero state even without force, as it's just the neutral hand position. Set the top dedicated button to zero the current position as the neutral state just in case.
Throttle would be more difficult, but you could make it so the input is relative to when the grip button is pushed. Want to speed up? Hold the grip button and move the controller forward and release to set cruise. Stop? Hold the grip button and move the controller back until a haptic buzz. Make it so the strafing thrust resets to zero after releasing the button.
The touchpad could control the gimbals. Make it normally act as a series of buttons, but then when the stage one of the trigger is engaged, switches to an analog gimbal control.
I really think I'm going to have to do some experiments with this...
I have my questions about docking collars in general given the position of some of them on ships being highly recessed vs other ship geo (like say, the Constellation) and the relatively thin wall cross sections. I just can't see a segmented rigid tube extending far enough to make the collar work without positional ballet. A thin flexible plastic/fabric-like tube could potentially work, and would solve the problem of needing two potentially massive ships having to be perfectly in alignment.
Dunno. I don't even think CIG knows to be honest. Lots of talk about landing, EVAing, snub ships and things like the ARGO, but very little about docking. Let alone docking between larger ships.
Dual 6DoF joysticks. The sky's the limits when it comes to how you want to interpret the movements and map input to output. The lack of centering could be an issue, but having used the controllers for a some time now, I'm relatively confident it's a solvable problem without needing physical rigs.
Controller angles for pitch/yaw/roll are relatively easy to return to the zero state even without force, as it's just the neutral hand position. Set the top dedicated button to zero the current position as the neutral state just in case.
Throttle would be more difficult, but you could make it so the input is relative to when the grip button is pushed. Want to speed up? Hold the grip button and move the controller forward and release to set cruise. Stop? Hold the grip button and move the controller back until a haptic buzz. Make it so the strafing thrust resets to zero after releasing the button.
The touchpad could control the gimbals. Make it normally act as a series of buttons, but then when the stage one of the trigger is engaged, switches to an analog gimbal control.
I really think I'm going to have to do some experiments with this...
Something like the Aquila isn't going to have an ARGO, and the P52/72 is barely useful as it can only carry one person. Frankly we just don't know much about docking in general at this point.
I have my questions about docking collars in general given the position of some of them on ships being highly recessed vs other ship geo (like say, the Constellation) and the relatively thin wall cross sections. I just can't see a segmented rigid tube extending far enough to make the collar work without positional ballet. A thin flexible plastic/fabric-like tube could potentially work, and would solve the problem of needing two potentially massive ships having to be perfectly in alignment.
Dunno. I don't even think CIG knows to be honest. Lots of talk about landing, EVAing, snub ships and things like the ARGO, but very little about docking. Let alone docking between larger ships.
Yup, I've thrown the idea around a few times. The best solution I could think of for the feedback problem was to have an easy way to set a new home position as needed and have your movements from there be your inputs, similar to how it works with the Steam Controller's gyro (like you're talking about here).
For throttle, I'd still go with pedals. I think it would be too hard to avoid accidental movement on that axis during other movements.
Edit: I've been thinking of it as one controller for all movement, then the other for mouse/aim. If you use both for movement, you could do some crazy stuff like controlling a virtual yoke for very precise inputs defined by relative position and orientation between the two.
Deckard Chapel
Banned
So wait a minute, this game has been in the works since 2011, has four studios and over 300 people working on it - and it's still in Alpha?!?
Does that make any logical sense, or am I just being naive?
Does that make any logical sense, or am I just being naive?
No, development started in January 2013 with like 12-14 people. They finished kickstarter campaign in mid November 2012.
Zalusithix
Member
Yeah, it's not the first time it's come up, but the previous times were before we actually had the hardware. Getting a feel for what's possible and what's not before you even have your hands on something is rather difficult. Too many variables in play. Even now when I'm relatively confident on some fronts, the proof will be in the actual application. The one benefit is it's all software, so iteration can be rather rapid.
As for using pedals as throttle...Traditionally they're yaw, though you could use them as roll. The angular push is for toe brakes which admittedly have little use in a space game, but can't be mapped to throttle as they're separate axes. Using the linear movement of the pedals for forward and backwards thrust seems... odd. Doable, but odd.
I'm trying to sidestep everything aside from the controllers with this concept though. If I'm going back to physical interfaces, I might as well continue down the multi-axis throttle path.
Ðuelist
Member
You have been lied to I'm afraid.
Technically they finished funding in 2012. They had maybe 10-15 people working at the time.
The original kickstarter was JUST for a single player campaign. Then the money kept coming in, so the scope increased and so did the fidelity of the assets and mechanics.
Depends on what you really consider it to start. I personally consider development starting late 2014 when they had a couple of hundred employees and the structure to produce the assets effectively.
Oh yeah, there would definitely be interest in it if you could put something together.
For my CH pro pedals, I ignored all convention and have front strafe / throttle on the right toe brake, back strafe on the left toe brake, and roll on the rudder axis. Sounds goofy at first but it works well with the Steam controller.
Honestly its too difficult to predict what hardware would run SC comfortably at this time. The game is still a year+ away from release and the network code is really unoptimized and holding back the performance. You'll notice the game runs much much better in Arena commander without the persistent universe aspects. Hopefully the new network code will show marked improvement quickly from what it is now (slated for initial release in Alpha 3.0).
If you really need to upgrade now for other games however, Lando mentioned they were using 980ti's in the booth so something like a 1070 should be fine.
You've been given the wrong information sir.
The thing is, it's absolutely bizarre that the framerate is tied directly to the netcode. I can't think of many (any?) games like that and I feel like it's going to be a big problem in the end.
In other games, a poor network connection results in all sorts of glitches (teleporting players) but the framerate is unaffected.
staticneuron
Member
Iirc they had 7. And no it was not only for SP.
https://www.kickstarter.com/projects/cig/star-citizen/description
The PU was always apart of the pitch. The only difference is that they had planned to do S42 first. With increased funding, they decided to develop them both alongside each other.
Yeah, I've seen very little talk about docking--and yet there are airlocks/docking collars on every large ship and sometimes multiple ones. If the Endeavor has a docking collar so the Carrack, Aquila, Dur, etc. can dock to it, it's a sure buy for me next sale or in-game. I'm getting more and more friends into this game and it would be great to have a central and mobile base of operations for all of us that's still manageable to operate with 10-15 persons. It's probably to early for any of this but I blame the hype of 3.0 for this.
Because they need to sync way more information then other games. And because many calculations need to be done on server side to prevent hacking, client need to wait for response from server to proceed with further computation.
They are syncing way to much stuff right now and they are working on a system to eliminate that and improve physics calculations on server side.
Ðuelist
Member
I could've sworn that PU was a milestone rather than the original pitch.
Either way, that's good to know.
This is helpful, thanks.
Things have changed a lot since I first started following the game so this is good to know.
RubberJohnny
Banned
I thought there weren't any checks against hacking in the game yet? You can get rid of stuff like criminal timeouts by changing your system clock, and there exists a client hack that allows a fully offline PU, so I don't think any of the security or AI or whatever is actually done on the server like with MMOs.
EDIT: Actually even if it did do those checks, it wouldn't be a reason to tie client performance to netcode, plenty of games do those checks without doing that and in theory they should mean better client performance not worse as stuff is offloaded.
EDIT: Actually even if it did do those checks, it wouldn't be a reason to tie client performance to netcode, plenty of games do those checks without doing that and in theory they should mean better client performance not worse as stuff is offloaded.
AI is done on the server. It has to be. There are checks for damage, ammo etc.
They also track to many things right now in too high refresh rate and send it to everybody, not only ships in proximity.
At least this is what they did last time Chris talked about it in 104tC.
From what I understand because the server/netcode/etc was made for small arena type FPS games, it sends way too much data that is often unnecessary for a players immediate circumstances (for example things happening hundreds of thousands of kms away). In turn, the client has to always process all of this data, no matter how far away these things are happening from the player, hurting performance.
When you can see things such as this
https://www.youtube.com/watch?v=G29b0GKX6Sc
Though cool, the server is probably sending way too much info
For what it's worth, I tried some basic edits (value freeze) with cheat engine a while ago and it worked fine in single player, but not any of the online modes.
Similarly, playing Crusader offline was an edit of the offline mode.
https://m.reddit.com/r/starcitizen/comments/3uj93k/no_ptu_access_play_it_solo/
Do as you like if it works for you.
However the reason twist feels fine with yaw is because Elite gimps yaw control. Yawing and Banking are pretty much to a person's preference.
Xplatformer
Member
So this is how you get to GrimHex without committing a crime.
As soon as 2.5 goes live and raised some cash this is the first place I'm going.
As soon as 2.5 goes live and raised some cash this is the first place I'm going.
Sometime before Winter 2021.
superdeluxe
Member
So Chris mentioned csquad42 stuff at citizencon, are we expecting a trailer?
Xplatformer
Member
Persistent universe? 2018
We're talking fully customizable avatars (I'm talking The Division level customization...not the overkill type), full npc AI (copilots...npc shop owners that go home at night, come back the next day), in PU stories that involve stealth missions into Vanduul held territories.
I'm expecting a lot by 2018.