|
Jobbo_Fett posted:Isaac Newton's laws are in Star Citizen. Checkmate Heh, speaking of Newton's laws, I was recently gifted an Aurora in an attempt to get me into the game, but I find it far more fun to hover in EVA over the top of someone's connie and wait for them to take off. Nine times out of ten (so far), the connie hits me and starts glitching (with bits falling off) or spinning wildly out of control with damage states activated. Last time it crashed into the station and exploded. 11/10 GOTY :-D
|
#
¿
Dec 24, 2015 19:06
|
|
|
|
| # ¿ Apr 26, 2024 06:08 |
|
|
Mirificus posted:
Good lord - if you have to wait for your monthly pay packet before you splurge $1500 in virtual spaceships for a game that is 95% incomplete, then you probably cannot afford to be spending that money in the first place!
|
|
#
¿
Dec 24, 2015 19:09
|
|
|
redwalrus posted:
Hang on, does CryEngine even provide a basic quadruped animation rig? I know it is a bitch to get looking right (Ubi spent years getting it to look good for Far Cry), so if the engine doesn't support it natively.... well... let's just say even CIGs biped animation looks rear end IMO.
|
|
#
¿
Jan 5, 2016 02:26
|
|
|
Nyeehg posted:I know nothing about art. Yet for reasons I can't explain, this poster offends me. I count at least 4 or 5 different perspective vanishing points in that image - looks like it was thrown together from several different concept pieces rather than composed as a whole... given how CIG appears to work, that seems likely.
|
|
#
¿
Jan 8, 2016 02:14
|
|
|
Chalks posted:Why are the ring rotation animations so glitchy as if they're suffering from network lag? I've always noticed this and it doesn't make any sense. Why aren't the ring animations client side? Are they really trying to stream full dynamic structure animations directly from the server? What's the point of that? The ring animations aren't client side because you are able to fly through the rings, so there is a moving collision volume that needs to support physical interaction with a network replicated object, meaning it too needs to be replicated. Nothing would break "immersion" (heh) so much as a connie flying full pelt through a pylon like it doesn't exist because the rotation is at a different phase in a peer's simulation. 'Course, there are other ways of approaching those types of problem (e.g. synchronized rotation based on elapsed time with drift correction, fixed physics time-step, etc), but this is likely a naive first pass.
|
|
#
¿
Jan 10, 2016 20:28
|
|
|
Chalks posted:It's a very typically CIG problem. The rings rotate because it looks good, it has no gameplay reason, but because of "immersion" they need to have a server overhead for the once in a million years when you fly into them. Usually you'd have static rings then if you find yourself with spare server resources (lol) you can add additional network packets and server overhead to cope with them rotating - but no, it's CIG, so we're going to have a fully animated and network synchronised dynamic station structure purely because it looks cool when 90% of the mechanics that will cause actual server overhead haven't even been written yet. I was simply explaining why the animated rings incur a network simulation cost, but yeah, given the fact the rings have no gameplay associated with them whatsoever they could have simply avoided the problem and left the rings off to save their engineers unnecessary headache. For those that might not necessarily know, unlike Elite, SC appears to be set in a universe where we have artificial gravity. In Elite's fiction a station's rotation actually has a purpose (to provide a sense of gravity) and provides a gameplay element (getting through the slot and (if you have rotational correction turned off) landing / navigating in the interior of a station).
|
|
#
¿
Jan 10, 2016 21:43
|
|
|
Lime Tonics posted:https://www.youtube.com/watch?v=aaoGxOxzAwc&feature=youtu.be&t=1h50m50s Hilarious awkward giggles mixed with silence when bearded hobo says, in spite of not reading it yet (because NO SPOILERS!!), the script for Sq42 is already starting from a better place than a AAA script "such as Skyfall" - you can just hear the other guys going "um, yep, um... ok... moving on". Also, how is ship stealing / owning going to work... "dunno - too far away". You have entire packages based on pirating, how do you have no idea, even at a very high level, of how the basic mechanics of stealing and ownership are going to interact? I've worked at 5 different game studios on 9 different AAA games and I have never seen such disorganization nor lack of idea of how the very intrinsics of the core meta-game mechanics of your game are going to work after so many years of working on the "vision"... as has been said so many times already, this is either all exclusively in CRs head, or they are making this poo poo up as they go along. This will not end well.
|
|
#
¿
Jan 11, 2016 22:08
|
|
|
Scruffpuff posted:That's another thing - Chris doesn't have a loving "vision". His vision is "everything." "I want a space game that has everything." Numbnuts, that's not a vision. Anyone can sit there and say they want a space game with space pirates and space rape and boardings (but not pvp god-forbid) and immersion and just say "yes" to every loving idea brought up on 10FTC. Anyone can do that. That does not comprise a "vision." ^^ this. I totally agree, this is pretty much the only way CRs decisions so far can be rationalized.
|
|
#
¿
Jan 11, 2016 22:27
|
|
|
A Neurotic Jew posted:10 for the developers Featuring such humdingers as... "Did you correct Mark Hamill's nose?" "When did UEC move away from paper money?" "Will citizens have different clothes from civilians?" Talk about cherry picking the non-questions.
|
|
#
¿
Jan 12, 2016 01:20
|
|
|
Amarcarts posted:I think the funny thing is that they don't even technically have 1 solar system yet. They have one planet and it's moons. And be careful engaging jump drive by accident with no valid target - I got "reaching end of known universe" about two seconds after leaving from whatever the starting station is called (right in front of the gas giant) before dying and respawning... so their "large" playspace isn't even the "one planet and it's moons".
|
|
#
¿
Jan 13, 2016 02:25
|
|
|
MeLKoR posted:Lowtax has warned you to stop posting creepy poo poo about Sandi already and private stuff like this is never cool. Can someone PM me what this is about? It sounds terribly exciting, but I can't see it and am having a terrible case of FOMO 
|
|
#
¿
Jan 14, 2016 00:06
|
|
|
Beer4TheBeerGod posted:Apparently adjusting the Freelancer's speed is a balance tweak, but adjusting the Glaive/Scythe is a fix. No, see 'cos [some handwavium about Freelancer top speed being tweakable, whilst Glaive/Scythe have deep ship pipeline dependencies on top speed]. Now we're nearly finished ship pipeline v.34g, these types of changes can be made far easier. Two weeks.
|
|
#
¿
Jan 14, 2016 01:26
|
|
|
Toops posted:Is Sandi "acting" in this intro, or is it me? Because she sounds like a floating wood block clomping against the pier. Not completely caught up yet, but do these AtV episode always follow this format? Is Sandi auditioning for a role as a news anchor or something? It's like "Sandi soundbite, Erin fill in, repeat x5 and ad-lib some witty banter" - I was half expecting Sandi to shuffle her notes when the camera cuts back to her and then at the end she'd turn to Eric for a muted conversation with lights fading out.... Hollow, bloated, over-produced, soulless and totally unfulfilling, much like the way the game seems to be going.
|
|
#
¿
Jan 14, 2016 22:08
|
|
|
|
|
#
¿
Jan 14, 2016 22:16
|
|
|
G0RF posted:She already has that role--both in fiction ("Beck Rossum") and in reality. Watch her early Wingman's Hangar interviews and you'll get a sense of how little she knows/cares about games and the gaming industry. One interview (with Jason Spangler, former CTO) was so frustratingly shallow and awkward that it was followed by Wingman interviewing the same person and asking useful technical questions and both of them having a jolly good time talking tech. The role of the interviewer is to drive the interview - Eric has excellent conversational interview skills, keeping the interview flowing and interesting whilst maintaining the focus of the conversation on the interviewee... these are skills that Sandi sorely lacks, and the interviews suffer hugely as a result. e: Just off the cuff, she could have name dropped some of the games that he worked on rather than asking him to recount his working history, then asked questions about aspects of those games and how they'll relate to Star Citizen... falling back to "Did you pledge? How much did you pledge?", totally shows what Sandi was about from the get go. sanctimoniousqfd fucked around with this message at 22:50 on Jan 14, 2016 |
|
#
¿
Jan 14, 2016 22:40
|
|
|
Nomenclature posted:Regardless of how much heat is being dissipated, having more barrels means that for a given number of shots fired, each barrel of a multi-barreled gun will accumulate less heat than a single barrel firing the same number of shots, so having a multi-barreled space gun might be useful. Still just brings with it other problems; mechanical lifetime is also a pretty major problem in space, due to things like cooling of lubricant, pumping of lubricant, evaporation of lubricant (liquids boiling off so fast is more a function of the surface exposed to vacuum than the relative temperature of the liquid to space).
|
|
#
¿
Jan 15, 2016 20:23
|
|
|
The last two days of Dilbert seem to apply really well to SC   
|
|
#
¿
Jan 15, 2016 21:50
|
|
|
BIG HEADLINE posted:So - silly question...but I remember in the hangar that the same way to get *out* of a cockpit was to press and hold the "F" button...why is that not working anymore? Now Ctrl-F, for... reasons...
|
|
#
¿
Jan 15, 2016 21:56
|
|
|
Incombibulator posted:You clearly do not understand game development. I'm beginning to suspect as much, which is highly ironic (since I am a game developer  )
|
|
#
¿
Jan 15, 2016 22:02
|
|
|
Matlock Birthmark posted:Oh great it's another of Derek Smart's alt accounts. Can't fool us so easily. Nah - you can tell because I haven't said "glorious" or "two weeks" yet.
|
|
#
¿
Jan 15, 2016 22:18
|
|
|
|
|
#
¿
Jan 17, 2016 07:42
|
|
|
Colostomy Bag posted:I'd rather have a cockpit view of that other ship when it happened. There is one - the video interlaces the cockpit view and shows the ship flying over part of the building in the center of the pad and completely clearing the first set of landing pads (flying over the "building" in the center). The shot only cuts away after it is clear of the second set of pads (and almost parallel with the third ring)... the on-foot player is standing on the building the pilot believes he is flying over (mentions as much in the comments) on the first set of pads, where it looks like the client simulation appears to correct the orientation of the ship (first time the on-foot player sees the ship, it is apparently flying backwards) which I guess applies some sort of physical impulse to spin the ship that in turn sends the ship careening into the ring in a flat spin.
|
|
#
¿
Jan 19, 2016 19:08
|
|
|
Mendrian posted:Maybe someone who actually programs can give me some insight into this: Not caught up yet, so entirely feasible someone has already answered this, but... Basically it all boils down to architecture and data budgets. When designing an engine, the architecture of the engine is determined based on a set of requirements drawn up very early on and taking into account technology now and (e: *possibly*) projected technology a couple of years out. That determines how much run time a given thread gets, how quickly file IO requests will be serviced, how data pipelines will be organized to guarantee a particular memory fill and drain rate and reduce CPU/GPU stalls, at what frequency inputs are processed in order for them to be translated, serialized and sent to the simulation (whether server-side, client-side or some combination of both) to make the game feel responsive. How gameplay events are processed (e.g. interrupt, single-, double- or even triple-buffered) by the various gameplay systems, what inputs must be authoritative vs extrapolated for the physics and animation systems to do their work, what order gameplay events are processed by the gameplay components that consume them, etc, etc. As a result of laying down all of these specifics, an engine builds up thresholds in its tolerance to changes in the underlying assumptions made in order to turn the engine into a viable product... "cruft", if you will. For the CryEngine, evidently one of those thresholds was that it can handle simulation inputs for 16 players, but going above that starts to create instabilities in other systems - perhaps important messages are now dropped by the networking systems due to buffers not being big enough to handle the additional influx of data. I am not overly familiar with CryEngine, but this may or may not scale linearly, resulting in a geometric or (eek!) exponential increase in resource requirements - in any given instance, the simulation server needs to be able to simulate a possible X*(n*(n-1)) interactions in a given timestep (where X is the number of interactions possible by a player in a given timestep, and n is the number of players), and the clients need to be able to handle the increased event load that comes with it. Going from 16 to 24 players means more than doubling the number of possible interactions in given timestep, all of which is sent to and from the server with varying guarantees of delivery (lost packets may be re-requested, resulting in a delay in processing, or dropped altogether and so never known about). Now, there are steps that can be taken to greatly mitigate the increased traffic, such as defining interaction "causality bubbles" where fine-grain interactions are only sent to peers in the general locality, but the engine still needs to be able to cope under the strain of all players choosing to be in the same place at once. That's why engine limitations like number of supported players are usually set in stone and do not tend to fluctuate much over the course of an engine's lifetime without large-scale changes to how those interactions are implemented. From an engineering standpoint, depending on the the original specifications for an engine it can often be easier to throw it all away and start again... but that is enormously costly, in terms of time, money and lost experience. This is why COD still runs on a heavily modified Quake engine, why Bethesda games still run on Gamebryo, why the latest Splinter Cell still ran on an extremely mature version of the Unreal Warfare engine that was used for the very first Splinter Cell, etc, etc. Wow, my first effort post on SA... hope you enjoyed it :-) sanctimoniousqfd fucked around with this message at 05:49 on Jan 22, 2016 |
|
#
¿
Jan 22, 2016 05:28
|
|
|
ZenMaster posted:Why does the rift need 2x the processing power? Couldn't you just have the left eye monitor take a photo of what the right eye is seeing and display it at the same time (or fast enough to not notice) or is our eyeline that offset that you need two slightly different views? Couldn't you trick the brain? I assume they thought of this, but it seems simple enough! 1. Because it is displaying two slightly different view perspectives, which, as far as the GPU is concerned, requires two complete renders - occlusion, visibility culling, shadow volumes... everything needs to be recalculated. 2. a) No, because that would be perceived as 2D by the brain (and might even make you a little disoriented) b) Yes, the spacing between the eyes (called the interpupillary distance) is what grants a significant enough perspective shift that the brain can interpret a scene as three-dimensional 3. We are already tricking the brain... we are presenting two completely separate 2D renders simultaneously to trick the brain into perceiving it as a three-dimensional scene ;-) The concept is simple enough... the tech required to actually pull it off convincingly is not. The Rift represents the first device approaching the mainstream market to deliver a solid VR experience that doesn't make you blow chunks as soon as you move your head.
|
|
#
¿
Jan 23, 2016 10:11
|
|
|
trucutru posted:To be fair it's something that has never ever been tried before so you cannot expect them to figure in advance all those little details. Like, how can you prevent someone else from stealing your ship? it's not like keys are available in the future. "Once we see how Elite does multi-crew, we'll copy that, but ours will be better because of FIDELITY, see!! We'll... we'll even show it off the day after E:D multi-crew releases, so nyah!"
|
|
#
¿
Jan 26, 2016 04:20
|
|
|
I am the 12 second section of the 16+ minute MVP video where Froogle "compliments" the yaw rate over that of Elite Dangerous and says how realistic the physics are with g-effects causing pilot-only blackouts (because a 30th century space vessel with artificial gravity has no means to counteract inertia for the person flying it). gently caress you CIG, that's low. https://www.youtube.com/watch?v=MOoinEmmVLI&t=3102s
|
|
#
¿
Jan 29, 2016 00:33
|
|
|
peter gabriel posted:In a staggering twist of idiocy CIG have just said G forces won't affect anyone but the pilot, so we are a situation where a pilot could be fainting from GFOOOORCE and you could be stood at the side of him giving it 'LOL@U' totally unaffected LOL, yeah, before ultimately crashing into a station and taking everyone in the multi-crew ship out because you took so long to recover whilst in the pilot seat and nobody else could take control to avert catastrophe... Idea for another vid? ;-) sanctimoniousqfd fucked around with this message at 00:53 on Jan 29, 2016 |
|
#
¿
Jan 29, 2016 00:41
|
|
|
ryde posted:This is a promise that would make sense to walk back, since having ship-external physics impact the ship-internal physics could be pretty tricky. They should just remove the blackout altogether and straight up say "Its not working out, we'll try it again some other time." Its not as if blacking out is a thing in other space-shooty games. Why do they want to die on this hill? Because CIG have built themselves up to their backers on the promise that they are doing things different from traditional game development, i.e. the land where gameplay ideas are tried out and anti-fun things are determined to be anathema for a product that needs to sell itself... Because EA, Acti, Microsoft, Ubi, et al, are "broken" so the way they do things must be equally "broken". I'm actually starting to think that ideas that Chris himself has doubled down on are untouchable... they stay in whether they are fun or not, because it shows that CIG is doing things different (even if they are not necessarily the "right" things) and he is pursuing his perception of "fidelity" even if it flies in the face of "fun" (or, you know... actual fidelity).
|
|
#
¿
Jan 29, 2016 01:33
|
|
|
Ladies and gentlemen, for pure nightmare fuel I give you the Coconut Crab
|
|
#
¿
Feb 11, 2016 21:30
|
|
|
xanif posted:Just drink some low calorie beer. Try this No good can come of this.
|
|
#
¿
Feb 23, 2016 20:29
|
|
|
Xaerael posted:I thought consoles were the devil, and all need to be burned in a big heap as a silicon sacrifice to the almighty PC god, cRobinkles. Well obviously until SC releases there is nothing at all to play on PC, so, even though they are forced to give themselves lashes and tighten their cilice for it, console games is all they have until the glorious vision of The Christ CRobberts is realized...
|
|
#
¿
Feb 28, 2016 18:58
|
|
|
Scruffpuff posted:https://www.youtube.com/watch?v=DFkTG9YungQ Hah, "back-end server mesh tech" @ https://www.youtube.com/watch?v=DFkTG9YungQ&t=270s... I imagine he means that they are looking at making the 16/24-player instance servers peer with one another to share their client/server simulation state in an attempt to increase supported player numbers per "instance" in the PU... I have full confidence that CIG will pull off this *extremely* complex and overly ambitious method of network simulation with as much skill and to as high a quality as they have the released modules so far.
|
|
#
¿
Mar 1, 2016 00:49
|
|
|
Fortuitous Bumble posted:According to reddit it will involve nested layers of self-organizing networks that split and merge based on some sort of spatial bubble system, which to me sounds like a larger project than actually making any sort of space video game. Ha... Haha... hahahaha.... hahahahahahahah! <snort> Oh, this game is the gift that just keeps on giving. Yes, that's a huge technical undertaking, and if past performance is any indicator it'll end up never happening and in the minds of the shitizens that will be double-plus good.
|
|
#
¿
Mar 1, 2016 01:08
|
|
|
peter gabriel posted:Just like that LOL  e: needs more fez ;-)
|
|
#
¿
Mar 1, 2016 01:20
|
|
|
Beet Wagon posted:Nope, it's been this way since forever. It's just bad design, through and through. The problem is they felt the need to try and automate a bunch of stuff for you to make your ship feel smart (as opposed to Elite's "put your gear down and then strafe down onto the pad" solution which works perfectly) and the end result is that regular flying is loving ruined, any kind of close manuevering is loving ruined, you can never tell where you're going to actually land or if your ship is just going to drift away when you get out, AND there's like three different kinds of assisted landing mode to help you out - all of which are completely loving unusable unless you've played their tutorial and learned how to use them. That tutorial is broken, by the way. So have fun with your brand new super-futuristic pile of poo poo. E:D goes one further than button combos - when free-looking you can look through the systems menu on the right pane and find landing gear, lights, cargo scoop, all toggle-able items within the in-game UI. So you have bindable buttons, button combinations and an in-fiction ("immersion") means of doing such a simple thing as lowering landing gear instead of the "landing gear lowered, landing gear raised, landing mode engaged" BS in Star Citizen.
|
|
#
¿
Mar 1, 2016 04:20
|
|
|
D_Smart posted:whoa!!! which word? I have many! All of which are perfectly sackable offenses! I assume he means the word "smart" ;-) e: Beaten to the punch...
|
|
#
¿
Mar 2, 2016 01:42
|
|
|
Tippis posted:…beaten by another Braben game before it's even out. Someone really needs to step up and stop that guy. Notice how Braben has a hand in the vision for Frontier's games, but leaves the details of designing and building the game to those better equipped to do so... That is the sign of a competent studio head.
|
|
#
¿
Mar 7, 2016 19:58
|
|
|
"Tattoo-er-wee-in"... The self-proclaimed Star Wars fanatic who cannot pronounce Tatooine (taa-too-een)
|
|
#
¿
Mar 8, 2016 00:35
|
|
|
Ash1138 posted:I wish there was an easy way to find video of Chris Roberts talking about a specific technical thing and John Carmack talking about the same technical thing. Just to see the difference. Ask [Google] and ye shall receive. John Carmack at an AMD Mantle roundtable session: https://www.youtube.com/watch?v=3exPJu_F8xk Chris contributing (as far as I can tell) precisely zilch at an AMD Mantle Q&A: https://www.youtube.com/watch?v=sSY2KXBoro0 Videos published within a month of one another. Chris shilling AMD at E3: https://www.youtube.com/watch?v=jDrIyOjEYRk Note how much more comfortable he seems with the marketing BS than he is at the technical Q&A event... not to mention how bored he looks at the Q&A! sanctimoniousqfd fucked around with this message at 02:07 on Mar 9, 2016 |
|
#
¿
Mar 9, 2016 01:56
|
|
|
|
| # ¿ Apr 26, 2024 06:08 |
|
|
orcinus posted:I've just scrubbed through the vid, but from what i've seen, he didn't say a single word? @17m43s is the only time he speaks (I think! I stepped through in 5 second intervals until I got bored of watching him picking his eyes, fidgeting, etc. in stop motion). It was a junior level observation that didn't take into account certain fundamental issues when multi-threading a game engine... almost as you might expect from this project! Like a kid trying to say something clever at the grown-ups table to get some attention.
|
|
#
¿
Mar 9, 2016 05:46
|
|