|
Bolow posted:Trackballs being on a thing you're moving around constantly seems really loving dumb. It's for MFDs (multi-function displays). It allows you to stay HOS (hands on stick) but still manipulate a cursor. In Star Citizen it'll get a similar use for radar display/target designation/EWACS. Supposedly the next preview will include the new GUI for multicrew ships, including the updated radar display.
|
#
¿
Nov 4, 2015 18:27
|
|
|
|
| # ¿ Mar 23, 2025 06:39 |
|
|
Troll Bridgington posted:Roger delta niner over Comes from a sim background. Helped a friend out with this for DCS - http://bit-shift.com/icontroldcs/ So you end up learning a lot of crazy acronyms.
|
|
#
¿
Nov 4, 2015 18:57
|
|
|
alphabettitouretti posted:Seems like it'd be way more fiddly than a ministick, and also potentially useless for thrusters, something which would be far more helpful in a "game" like Star Citizen. If they haven't put in some kind of thruster specific controls they've already made a HOTAS that's worse than the 20+ yr old design of the CH Throttle. MadCatz/Saitek tends to solve this via game-aware mode switching. Certain games expose data to the joystick drivers which allows them to switch modes automatically. For example in Elite, the T.A.R.G.E.T. drivers (ug) from TM update the controls for Flight Assist off movement. The hat can be updated to provide thruster inputs. You are correct about trackballs being terrible for thrusters. Generally you do not want to use an input that requires polling (ask an android developer their thoughts on detecting trackball events) for a state-ful control like a thruster. It is better to use an impulse event (hat up, hat down), which is likely what this stick will do. The trackball is useful for emulating mouse movement without having to move hands from the controls. As a real world example, this is used on MFDs to do target selection. You do not want the pilot to have to move hands off the throttle/stick to make a selection. In Commercial aviation you run two or more pilots (PF and a PNF) so this is less of a concern. Good example of why militaries use a more complex HOTAS setup vs Commercial aviation having multiple pilots: http://aviationtroubleshooting.blogspot.com/2010/12/qantas-flight-qf32-captain-david-evans.html
|
|
#
¿
Nov 4, 2015 19:39
|
|
|
JakeP posted:The marketing is all word of mouth among a community of spergs and their friends. There aren't new people getting involved in this. Its the same goofy morons buying new jpegs every time they come out. Sandi isn't doing anything to motivate these sales, she just happens to hold the title of the person who is usually responsible for marketing. According to Derek she was the one who designed the Concept Sales (aka, jpeg ships). So, I guess that's going for her.
|
|
#
¿
Nov 4, 2015 21:04
|
|
|
SirPhoebos posted:I've mostly been lurking in this thread, so I'll let goons that have been involved with SC respond to your points. But one thing I'm curious about : as someone that still believes SC will be released, what do you think the game is going to be? I'll answer this one from my perspective. I'd like to see a take on commercial aviation in a space genre. I'm an edge-case, as I come from the Flight Simulator world and what I like to do is commercial flight (transporting passengers from location to location). For awhile I was on VATSIM, but eventually lost interest as further developments post 2004 (when FSX launched) stopped. With the death of the ACES team at Microsoft, there's been a lack of development in the flight simulator market. DCS is still around doing their combat games, and I was slightly involved with that, but nothing new has come. On the space sim side, we've had Elite: Dangerous, which took a fascinating route to release. Instead of trying to build complex, they went with polishing a simple flight model and releasing a basic simulator. It is a good example of a well designed product, though people quickly discovered that large world sims suffer from grind/lack of depth if they're released without enough interacting systems. They've been steadily adding bits, like the faction system, but their core gameplay (the verbs of their design) have not changed. That's why it still feels like a grind despite new features being added. They have not increased the gameplay depth yet. Once they do, I'll be looking into it again. My favorite release was the AssPirates in Space spin-off, Rebel Galaxy. Though short, I finished it with the largest ship in around 10 hours, it is exactly the kind of snack-sized thing I like between big games. I'm hoping that SC, at a basic level, will be as delightful to play as Rebel Galaxy, but with the depth of a full sim like Flight Simulator so that I have something to spend my time getting to learn the nuances of a complex system.
|
|
#
¿
Nov 4, 2015 21:39
|
|
|
Orcs and Ostriches posted:Drive around in the buggy listening to sick tunes? This and doing the Arena Commander mode. Interest spikes and then falls off after each update as people try out whatever small changes have been made. For my part, I'm waiting for the baby Universe (AC 2.0, I guess?). Much like in Rebel Galaxy, I love traveling around big maps and exploring. It's the same reason WoW pulled me in when it first came out.
|
|
#
¿
Nov 5, 2015 00:04
|
|
|
illectro posted:I like to fly around in AC in an Aurora and get kills, mostly to troll the whales in their Gladiators and Super Hornets. The Aurora right now is hilarious.
|
|
#
¿
Nov 5, 2015 00:11
|
|
|
tooterfish posted:You never watched the Sopranos? A RICO case requires the justice department make a case that there is a conspiracy of criminal intent. You have to show the participants were part of an organization and had intent to commit a crime. RICO allows you to charge all of the participants of a criminal conspiracy with the same charges, instead of having to build a case for each individual. They're generally used for taking down large syndicates or networks. This would not happen for a game developer as there is no need to put in the work. You can charge the participants individually and have the result you need. If everyone involved at CIG was part of a criminal enterprise, and remember this requires not only showing that their actions were criminal, but that the intent was to commit a crime, then you'd have grounds for a RICO prosecution. This is at best a small part of a foreign subsidiary of a large developer. A lot of younger people remember RICO took down the MOB!! And forget the reason RICO was created, and used, was to take down large criminal enterprises when the leaders were not committing the crimes. Mr. Smart has insisted that he has knowledge that at least some of the leaders are directly involved in criminal activity, so again, no need for RICO in a case like that. ... Assuming that all is true, of course.
|
|
#
¿
Nov 6, 2015 21:59
|
|
|
Kazy posted:Is it me, or is PC Gamer the only place with regular articles on Star Citizen? There is little reason for most places to report on it. A lot of them investigated some of Derek's allegations or wrote about the Escapist article, but as there has not been a major update since Arena Commander/Racing and the 'Social' module came out, there's no reason for game sites to update about it. It's kinda the same tack they take with Early Access. You don't see a new post on every Rust or Ark update. Only for major updates.
|
|
#
¿
Nov 6, 2015 22:15
|
|
|
Wafflz posted:but $50 is roughly what a video game costs, paying the normal price for something isn't the issue here. It really depends on your perspective. Just as an example, out here in the valley, one of my friends just had his third company purchased. One of the things he likes to do is spread that income to kickstarters, he tosses $5-10k donations to whatever his favorite projects are at the moment. When you have enough money to no longer worry about acquiring income, your perspective shifts on what you consider 'wasting' funds. I occasionally think about $100 purchases, but I have friends who have difficulty getting $5 together for lunch. It's all a matter of perspective. (For those curious, yes, I do help out my friends.)
|
|
#
¿
Nov 6, 2015 22:23
|
|
|
Unfunny Poster posted:Yes. I've been in production meetings where people get super loving mad about being given "Producer" credit rather than "Executive Producer" credit. It's a hilarious thing that never ceases to amaze me. That's because they know that Producer is the real poo poo and Exec Producer = kiddie gloves tag. Which is why someone like Bonnie Arnold (Producer on Tarzan, Over the Hedge and How to Train Your Dragon) ends up co-president of feature animation at Dreamworks.
|
|
#
¿
Nov 6, 2015 22:51
|
|
|
Matlock Birthmark posted:Well good, now they can build another Mocap set for that ship. Yeah, now that they have the studio, they love doing custom animations for all of the accesses. It makes things look nice (when it is finished), but it is annoying to players if the canned animations take too long. I'm hoping they'll drop back on it eventually. One of the most impressive details of a studio like Naughty Dog is they know when to use canned animations, and when to use environmental IK tagging. It's why you see Joel (in last of us) reach over to brace himself on certain objects when he is crawling. It is not an animation you cannot break out of, just a nice touch players notice. I'd like to see more of that, but Crytek does not have that capability, to my knowledge. (That is, it has IK and pre-IK, but not IK tagging on objects). In my opinion, it's better for western games to focus on interrupting player movement as little as possible. Here's where I note I'm not a fan of Monster Hunter or Dark Souls. 
|
|
#
¿
Nov 6, 2015 23:03
|
|
|
Chalks posted:This is Bounties have been known about for awhile. It's part of the Mercenary job set. (Escort or Attack missions, NPC created or PC created). quote:Here's another question, about LTI in case anyone's been making a mental note of the crap the come out with. Has anyone explained how the game can distinguish between someone with LTI getting their ship stolen via boarding piracy and them just giving the ship to a buddy then claiming a new one through LTI? They say the market systems will account for this. The NPC insurance companies will deny claims if you end up having too many. In addition, the hull ID is flagged as stolen, so if you're giving it away to a friend, they'll be using it in the equivalent of EVE Null space. Also, LTI only applies to what the hull came with, not to the equipment or cargo. I'm sure people will abuse the heck out of it and have a good time. CIG is not too worried about this kind of friendly fraud. If you have followed along from the beginning, their plan was to release ships on a regular basis much like car companies do. So you'd have the 300i, followed next year by the 302i, etc. Assuming they pull it off (or release the game), it'll be used as part of the inflation mitigation. Much like how new mounts were in WoW.
|
|
#
¿
Nov 7, 2015 00:17
|
|
talking out of his rear end isn't it. I mean, if you can put a bounty on me because I ran away without giving you X, I can put a bounty on you because PARRRRP. So either bounties actually mean nothing or they're great for griefing...? Oh poo poo looks like another feature being made up on the fly literally while answering the question.
|
Chalks posted:So not actually a thing you can just set on someone because they didn't do X, but rather you have to pick a specific job to randomly spam bounties on people? Not too clear how that avoids griefing or indeed why I can't just have friends cash in my bounties for no penalty. Bounty systems only really work in games with severe death penalties. My understanding is that you can offer a bounty on another player. It's been awhile since I heard (Ex-CIG member) Rob Irving talk about it. But the intention was at that time in PVP areas you could flag another player and then offer a bounty contract on them. Other players could collect. It's a bit different from EVE as the areas are not universal, it's instanced. If you're wanting to track down another player, you have to flag them in game (so the server knows to match your instances). You'll only find them in a PVP zone. Supposedly areas that are in heavy control, like UEE areas, will not allow this via some method. This is the kind of stuff that has to be worked out. I recall how Age of Conan launched and how much it changed from their hardcore philosophy. I'm sure, assuming SC ever launches, there will be all kinds of changes in the gameplay design. quote:So if you blow someone up 3 or 4 times the game will refuse their LTI? That seems like a fun grief. There's no set limit on it that I'm aware of, but there are factors the influence. The larger the ship, the more material is required to build it. Much like EVE, to build ships, the actual required materials have to be transported to the various factories, components shipped, and all assembled. Only then do you get the replacement. Small ships are easy to replace, they're common and all over the place being built. Large ships take more. A critical difference between EVE and SC (again, all of this is proposed because of course there is no SC out right now), is that the economy is almost entirely run by NPCs. There's a lot more of them than player characters in the economy. However, those large ships will still take time to build. That's one of the ways of discouraging LTI abuse. And yup, money money money. Though supposedly they will not be selling ships at launch. At that point, you'll be getting them for in-game credit only, so no more $5k idris-Ps. -- Again, something said, so who knows how it'll be by the time SC launches, if ever.
|
|
#
¿
Nov 7, 2015 07:07
|
|
|
Aesaar posted:This too. Directional shield controls sounds cool in theory, but in practice, no one is going to be loving around with them in the middle of a fight. At best they'll press an equalize shields key (which, in Freespace 2, is the only shield control anyone cares about). What's more is that if shields block 100% of all damage until they go down, there's no point to controlling shield distribution. Equalizing shields across all quadrants works just as well. It'll be mapped to a hat on a HOTAS. In the case of shields, usually these get mapped with a shift key. (Joy button 10 + Hat 1 Up, as an example). I've also seen a lot of people using Voice Attack for those kind of controls: http://www.voiceattack.com/ It's not a difficult problem to solve.
|
|
#
¿
Nov 7, 2015 11:07
|
|
|
SirTagz posted:From the monthly update from Frankfurt: Small note, the AI was not going to be in the first release of Star Marine. It was going to be PVP with players only. This is an addition for the AC 2.0 release. Of course, as Derek mentioned, Star Marine is still delayed. It's not landing with AC 2.0.
|
|
#
¿
Nov 8, 2015 00:26
|
|
|
Justin Tyme posted:Isn't the texture resolution mostly just dependent on how varied a texture map is and not really related to how "sharp" textures are, like the only reason you need 4k textures is if for some reason you have a gigantic model that is all one piece with multiple different objects on it? SC doesn't use 4k textures. Most of them are 512 or less. They're used more efficiently because of the deferred render technique. More info here: https://www.reddit.com/r/starcitizen/comments/3ogi3o/im_an_tech_artist_in_the_industry_and_id_love_to/ - They use an atlas of smaller textures so you can get the details without needing to use large textures and a lot of VRAM.
|
|
#
¿
Nov 8, 2015 00:33
|
|
|
OhDearGodNo posted:That graph is also completely wrong. For example Fallout 4, GTA V, etc didn't began after the previous iteration. You're wrong on both Fallout 4 and GTA V. You can read about FO4's development here: http://www.gameinformer.com/b/features/archive/2015/11/06/the-making-of-fallout-4.aspx?PostPageIndex=1
|
|
#
¿
Nov 9, 2015 13:59
|
|
|
CrazyLoon posted:Is the black supposed to symbolize their inner depression, that they try to stave off by buying spaceship .jpegs? It's to match the Centurion card: https://en.wikipedia.org/wiki/Centurion_Card
|
|
#
¿
Nov 9, 2015 14:03
|
|
|
EminusSleepus posted:if they can't simply check out the history and proper balance of the transaction with just one query then I say their database design is really hosed up and Lando might be correct that due to their shifting minds/rules with regards to packages and accounts their tables in their database right now is like as horrible as the monster that Dr. Frankenstein created. It's pretty common for MMOs. You should see the terrible mess World of WarCraft's oracle DB was before the cleanup after Wrath. When they did account investigations it took weeks because they'd have to load old backup (dbs) to their test RAC to query them. They didn't track the transaction history. It's why account recovery before 3.0 took weeks. It's no surprise that SC is going through the same pain, especially given their complex melt and gifting setup.
|
|
#
¿
Nov 9, 2015 15:53
|
|
|
Wafflz posted:I don't know how anyone's brain can see how janky everything they have produced so far is and still say "this is going to be the glorious savior of pc gaming." If you're curious, I can answer this from the perspective of someone who is a supported, but got concerned after Derek's reveals in July, but still is hopeful.
|
|
#
¿
Nov 9, 2015 17:10
|
|
|
Wafflz posted:I am actually. Being concerned and hopeful is nowhere near the guy I posted though, he thinks it is going to be a revelation. Skip down to  if you just want the SC stuffI'll try to keep this shorter (note: I failed) and skip most of the financial side. I can address that later if you're interested. I've been a simulation fan since childhood. My first real gaming experience was Chuck Yeager's Air Combat (on an IBM XT!) and Camel.exe. I loved Microprose's work with JANES (F-19 Stealth Fighter!), Papyrus's Indy Car, and all those classics of the day, but my true love was always space combat. Somehow, I missed Elite. My first experience was 1990's SNES port of Wing Commander, which is still one of my fond, if retroactively bad, memories. X-Wing in 1993 was the perfect game for me, but I always felt the want to go outside of the ships. I wanted to have a mission that involved multiple warps. From that time on, I was a fan of sims. Post I-War 2 and Descent: FreeSpace 2, the Space sims were essentially dead. Now an adult with all of the income I wished I had as a kid, I had nothing to spend it on that fit that love of space simulators. Nexus: Jupiter Incident was okay. Evochron Mercury was empty of story. Jumpgate Evolution was supposed to come out, but through all this time, I only had Flight Simulator to really get deep into as I used to. By this time, I had enough money to get involved in smaller investments and got to evaluate a number of proposals, but the overwhelming majority of them were tiny productions looking to deliver an experience like Rebel Galaxy. $3-10 million type of productions. I was pretty disappointed in the projects being offered, to the point I mentioned what kind of thing I would actually like to fund in 2010: http://www.shacknews.com/chatty?id=22187877#item_22187877 "I'd budget the following: Space Sim MMO, FPS and Space Sim 1. $40-50 million for an FPS that features Euphoria engine based on my favorite comic. 2. $40 million for a space combat sim in the style of Independence War II. This engine would be used for: 3. $70 million ($40 up front, $30 for continuing support) for an MMO based upon #2. It would feature an ongoing NPC economy and piracy system that players and guilds can interact with." I had gotten a good idea what the costs would be to deliver on the kind of project I was interested in after Earth and Beyond blew up due to EA's shifting priorities and joked about what I would do when a lottery. So, that's the history. I wanted a space sim that was detailed. When I was first approached, the presentation felt similar to Earth and Beyond ++ which is something I am not interested in. From my post in 2010, you can see where I was predicting the costs to go, I passed on the investment opportunity and then on the Kickstarter. There was not enough to interest me and the goals seemed unrealistic for the income. I knew it would take around 70+ million to get to that point. -- Let's fast forward to today. One of the things I've observed over the years is that younger gamers tend to have short memories. We tend to not see improvements. When you've been involved in a few projects you get a sense for where things are going, especially when you've invested in a failure and success or two (Maginatics and Apple, just as examples). Part of that income I spend on games. Before Fig I was doing a lot of Kickstarter funding (FTL, Wasteland 2, DFA, Replique, Shadowrun, PA - RTS, Edge of Space, Void Destroyer, Lacuna Passage, Project Rain World, etc) and one of the delights is seeing what used to be exclusively shown to publisher management get shown to the 'public.' So here we are, Star Citizen, which has one of the most updated development programs around, and I'm getting more access than I ever got as an investor. Because of our short memories, we tend to forget progress and only see the now. When I made the choice to fund was a little after their first modal ship reveal. During a livestream they demonstrated their modal ship* the Constellation. There was nothing released at this point, but I saw that the direction they were going is much more in alignment with what I was looking for. At that point I started to follow the development and made my first donation (for a 300i). I signed up for the Goonfleet with my essentially idle account because you guys tend to have the most fun when gaming and don't take poo too serious. * Modal ship is the equivalent of a vertical slice for a more traditional game. When you're making a sim, you want to get a complete all-up model of what your primary development goal will be. You spend the effort early to build this, rig all the simulations into it, and set that as the goal for your development. 'This is what all ships must reach to be acceptable.' Sorry for the big post. When you say 'didn't read' I totally understand  tl;dr Since that point, I've seen a steady release cycle. A baby hangar mode followed by a renderer redo (to go to a deferred method using PBR/texture atlas with normals). A basic (sim-system free) flight mode with basic AI. Then a race mode. That's essentially where we are on the surface level. What I've been interested in though is the background engine development. If you have not invested in a game before, you might not realize how much work is essentially behind-the-scenes engine work, and in 2012, what they proposed was not possible, which a lot of us recognized. It would take a massive engineering effort and years of work. What I look at is what improvements have happened to the engine since launch: The deferred renderer was improved in 2013 with the addition of support for chamfering, Physical-based rendering and the use of texture atlasing for decals. That's meaningless gibberish unless you've worked in the game industry and realize that saves you a huge amount of space on texture memory. It's critical for performance. If you'd like to read up on it: https://www.reddit.com/r/starcitizen/comments/3ogi3o/im_an_tech_artist_in_the_industry_and_id_love_to/ - That same year the CryEngine multiplayer code was rewritten*. In 2014 basic multiplayer was delivered (AC update 1.0). Ship outfitting and the first of the simulation systems (energy usage) came next which means simulation design was in the delivered engine. In 2015 the next major fix is shown, with the switch from 32-bit FP positioning to 64-bit. Non-clamped nesting zoning is added. Multiplayer Large is released (player counts over 15). * Did any of you play MW:LL? (Mech Warrior: Living Legends). That was another CryEngine release and one of the things the development team had a lot of complaints about is how terrible CryEngine's native multiplayer code is. Like everyone else who works with CryTek, they had to do a rewrite of the network code to make their game playable. It's a thing gamers see as a failure (CIG is terrible they cannot even get multiplayer working), but if you know your engine development, you realize that's a good delivery. -- Basically, that's what I'm looking at. When Derek's first blog came out, I decided to sit back and reconsider further donations. I wanted to see if CIG was still able to progress, as the FPS portions looked terrible. Not bad, terrible. We're talking Halo: Master Chief Edition bad. Since that time, they've demonstrated that they've moved to a 64-bit FP, while retaining 32-bit addressing for render (this, BTW, is the most impressive thing I've seen. Keeping the optimized render path is a massive issue, and I'm impressed any company has managed to deliver this). Proper nested zoning is in. My one goal I've maintained as being my measure of success since 2014 is 'supposedly' coming, the AC 2.0. That's kinda what I've been waiting for all along: http://www.shacknews.com/chatty?id=31750914#item_31750914 Sorry for the long poo. Loiosh fucked around with this message at 18:24 on Nov 9, 2015 |
|
#
¿
Nov 9, 2015 18:21
|
|
|
I kinda failed at TL;DR, so I'm going to apologize again. Sorry about that, all.
|
|
#
¿
Nov 9, 2015 18:23
|
|
|
Vire posted:Can someone explain to me how doing all this back end optimization still getting a crappy result would be any way better then just literally dumping the whole thing and starting over with unreal 4? Like how is that encouraging when unreal is basically giving away its engine for free to indie devs and has a lot of that stuff like physical based rendering and the like working by default in the engine, amazing built in tools for devs and artist, non lovely MP code and probably more things. I can answer this from my perspective, but I'm not a game developer. The issue with starting over or moving to a new engine is one of capability. The requirements for SC's goals are not available. From the sidelines we look in and see vehicles in (Halo or Unreal) for example and go, Look at that, it's right there! If you dig into it though you begin to realize limitations and how well hidden they are. It's why you (In Halo 5 as an example) still see the same jump height/floatiness while supposedly jumping 'down' the side of a large ship. The engine only supports a single zone (which includes a universal gravity vector). Unreal 4 is the same. (Tech note: 'zones' in Unreal Engine are self-contained area. They're not nestible). Even in engines with zoning, like Uncharted 3) they tend to be heavily restricted. Uncharted 3 has nested zones, but they're clamped on the X/Z axis. It's used for the sinking ship level where 3 different zones are nested to do the physics simulation, the only zone rotation is on the Y axis (bank) as the ship rolls. Unreal 4 does have excellent multiplayer code, but it is not optimized for zoning which would require the engine to functionally be re-written to support it because the entire engine thread makes assumptions like each level having a universal gravity. I have not worked with it, but I believe it does allow you to specify different gravity vectors, but they're static and baked into the level. (This allows you to do something silly like make a level inside of a sphere, but the limitation is that the sphere cannot rotate). You can see how that would be a problem for an engine that wants to have ships with internal gravity sources moving and allow the player to do EVA outside of a mobile ship. It's actually a hard problem to solve, and besides Derek's assertions that BattleCruiser 3000AD and LoD support nested moveable zones, I'm not aware of another engine that does (please feel free to correct me).
|
|
#
¿
Nov 9, 2015 18:50
|
|
|
Vire posted:Cool thanks that is actually interesting to know. So they have actually implemented the zone nesting in their frankenstein engine already? Is this why everything that should be simple is now broken because it's so ridiculous to do something like that having every ship having its own gravity while dudes outside of it float freely? Well, there's a couple of parts to that depending on what broken you're talking about. The zoning is relatively nice in the slices we've seen. It works well and is multiplayer. One of the funny notes about that is they just got player counts back up (it was down to 2 players stable for awhile as they fixed some major desync bugs). The biggest jank is something that has always plagued CryEngine, and that's movement. Movement refinement is hard, the reason Destiny, CoD/Half-life/Quake-engine games, and Unreal feel good to move in is because they were designed from the ground-up to be great shooter games. Everything followed from a core engine designed to make movement and shooting feel excellent. Bethesda games feel like poopoo when you are moving because their engine was designed with a different goal (wide expanses and heavy scripting). CryTek came from the FarCry engine, and was primarily designed as a demonstration of streamed asset loading (to make the world feel seemless, without loading). The movement has never felt fluid (Ubisoft would eventually rewrite /all/ of the movement code after they got FarCry 2 for their Dunia engine). Flying vehicles and zero-g were bolted in with Crysis and never well optimized. If you remember Crysis 1 MP the vehicles were always jerky and weird because of the lack of effort to clean things up. For Star Citizen, Croberts knew they needed to fix this, so they brought on a dev who they thought could do something similar to the Dunia engine for them, basically do a rewrite of the animation, IK and pre-IK engine, and movement code. Unfortunately that team (IllFonic) was entirely not up to the task (I'm guessing something like Batman: Arkham Knight by Iron Galaxy bad). If you saw those early presentations when they were first showing off the FPS stuff it was really jittery, lots of horrible animation interruptions and stuttering. It kinda looked like... well let's not go there. IT was bad. Bad enough that Chris, who is a huge fan of FPS games, pulled it away from them and scrapped their work to redo it in house. That's kinda when I took a step back and got concerned about the direction of the game. At the time I did not realize how bad the work coming from IllFonic was. I've seen a number of early releases (you should see what Frostbite looked like during its first two years), and recognize jank being part of it, but SC's combat was -bad-. The other part of broken is the animation transitions. Normally in game engines the animation clean-up side of things tends to come in later, after the engine is alpha'd so that the animation team can have a good idea of what their target is (Will we have IK? Will we need pre-IK? Blending? Tagging?). They'll have a target height and a nailed-down collision system to work with. Previous to that, they use bare animations and work on things like walk/run cycles and incidentals (death animations, opening doors, picking up items, etc). The polish comes in supporting clean transitions, where you add in the code and animations that link between various stands (walking -> stand. Running -> stand. Running -> against a wall. Reversing direction in run, etc). SC recently went through a rebase on their core skeleton, that is, they changed some of the properties of the essential base for all of their animation rigs (I recall this is to support a more robust item system), and when you do that, you have to redo all the rigs and then either recapture or tweak every single animation to fix where they break. This, combined with tossing IllFonic out, is why the FPS combat stuff is -so- delayed, even beyond AC 2.0, and why the presentations look so janky in the animation department. When it's fixed, the devs will be able to 'tag' entry and exit points (in fact, this is going on right now), so that the animations will correctly trigger when the player, for example, approaches a doorway during an EVA. If you watch the first Multicrew reveal vs the Citizon Con AC 2.0 reveal, you can actually see a few of these tagged animations get added into the engine. Generally this is the stuff that is added later, but because of how SC is being designed, you actually got to see what happens when an animation system is scrapped and a new one added in.
|
|
#
¿
Nov 9, 2015 19:25
|
|
|
Vire posted:Is it common for video games to use mocap for everything these days even as mundane as climbing up ladders and walking? I thought most of the time it was just used for cut scenes to capture an actor's performance because it has to be cleaned up so much anyway I would think it would be easier just to have an animator do a walk / run cycle or climb up a ladder. Yes, it is very common because it is generally quicker (to get nicer results) than traditional key animation. For those who wonder, generally you want to do incidental animations via mocap (things that are unique and quick) and cinematics (Last of Us is a great example of this). You want to save things that are repetitive and players will see a lot for your best Key Frame animators. So things like run and walk cycles. What happens now days is a studio will spend a day or more doing walk/run, climb, and jump up/down, ladder, mantle, etc animations with each main character actor. Then this is handed over to mocap cleanup where they fix issues with the animation curves and missing data. Once the basic cleanup is done, for those critical motions that will have a lot of use (walking, running, going into cover, standing in cover, crawling in cover, etc) you hand those to your key animators and let them refine them so they fit in your animation system and make them look alive. Meanwhile, on the mocap stage you do cinematic work. And then later do pick ups for incidentals (picking up a glass, picking up an item at rest, leaning, that kind of stuff) when you figure out things you can add into the engine without a big impact. Usually this is done well after you have a skeleton and have rigged it for mocap. In the case of SC, they had to redo this a few months ago, which is why things looked so bad at Gamescom, but looked (only barely) better at Citizencon. Between those times they did a redesign on the player skeleton to support a better attachment system and then had to rerig the skeleton and -then- do mocap capture for the new rig and retarget whatever animations could be scrapped. It's one of those things like 64-bit level space, that flies under the radar of casual observers, but is a big tell that there are fixes going on under the hood. You generally never see this kind of information come out in the public. Most people don't even know that for Far Cry 2 (and 3!) the entire animation system was tossed out in the middle of development to fix it because Ubisoft didn't like the feel.
|
|
#
¿
Nov 9, 2015 19:34
|
|
|
Regarde Aduck posted:If by dying you mean we get cool games but miss out on "AAA" poo poo i'm also ok with this. Darksouls 3 on the ps4, Elite: Dangerous on the PC. Everything as it should be. Rebel Galaxy, Else Heart.Break(), SOMA I wish more people would play Else. It's a weird little gem.
|
|
#
¿
Nov 9, 2015 19:50
|
|
|
Sarsapariller posted:First of all- thank you for the interesting perspective on all of this. I really do appreciate reading analysis by someone who knows what they're talking about development wise.  - but my knowledge is limited to just playing and reading.quote:Some questions/follow-ups: I was under the impression that most of the rework for the skeletal system was being done because of the weird requirement that first and third-person use the same animations, as opposed to interrupts/jitter? And it doesn't seem like they've done a good job of eliminating jitter, by recent videos. If anything they've made it worse, as you can see in the first minute here: https://www.youtube.com/watch?v=6ccKp1n13jQ Edit: Ignore the second part, you addressed this in a follow-up post. I believe you are correct on the major reason. Just as an example of why anyone would care: https://www.youtube.com/watch?v=hXmZX8iz2SE Some classic GAME WTF. Another great one is Trespasser, of which the Let's Play is basically why I signed up on SA. CIG is probably a bit away from getting proper fixes in for those animation issues you are seeing. One of the smallest (but most obvious) is animation/movement disagreement. It's why the steps 'feel' wrong in those gameplay segments. In addition to the weird looking animation, the distance traveled (for the animated step) is not in sync with the distance traveled by the character model. The jerkiness I suspect is a mix of both lack of animation transitions -- which they spoke about in their last two monthly updates -- and the network code for player movement state/animation state being in an early phase. Most game engines handle this by letting the client do the animation state calculations, IE: the network code hands a vector with velocity and the game engine calculates the correct animation, with some tagging to identify when to trigger a transition (from running to stop or running to reverse). This is the kind of thing that gets polished later as MP testing is going on. Because of SC being forward with the progress of the updates, you get to see what this looks like when it is unfinished. Compare that to the Gamecon footage of a similar scene and you can observe some improvements, but at the fundamental level there's still a great deal of work to do.
|
|
#
¿
Nov 9, 2015 20:12
|
|
|
Warcabbit posted:https://www.youtube.com/watch?v=oLB28JvEQAg Heh, exactly. If you'd like a little trick, when your player is 'in air' (state jump) to fix the weirdness jumping from one gravity to another, cast a ray each tic that follows the local gravity and a second that follows the global gravity vector (hopefully that makes sense). You can use angle between them to blend the character's 'down' vector and fake gravity changes. Based on the character distance from the local ray, you can blend gravity from local to global to look like the player is jumping from one surface gravity to 'ground'. It'll take a bit of work, but it's a fun trick to try out. Also, make a mobius strip!
|
|
#
¿
Nov 9, 2015 20:22
|
|
|
Tijuana Bibliophile posted:Have they actually admitted to scrapping Illfonic's work? The quotes I've read have all been "everything is aye-okay/weeks not months" etcetera In the way game companies do: http://www.gamesindustry.biz/articles/2015-08-17-illfonic-sees-layoffs 'Cloud Imperium Games has provided GI.biz with a comment on its plans for Star Citizen going forward, explaining the move away from IllFonic as a way to consolidate more development internally.' Aka, we scrapped their work, which is why they did not pick up most of the team and bring them in-house. Bungie did something similar with Destiny, though internally. They never said why the story changed or why Marty O'Donnel was out. Kotaku broke that they completely scrapped their story 6 months before their original launch (and pushed the game back a year), and that Marty was fired because he refused to play ball with Activision's music changes and decided to pout at work about it. Generally the game industry doesn't air their dirty laundry publicly. 
|
|
#
¿
Nov 9, 2015 20:28
|
|
|
Hav posted:Perhaps you'd like to express your feelings on scope creep, or the overall plan/road map? I don't want to look like I'm spamming the thread, so please forgive me if I'm answering a lot. One of our datacenters went down and I cannot do much work :/ - I did not join in during the kickstarter because I felt the game as too small for what I was looking for. The original investment pitch was along the lines of Earth and Beyond/Freelancer ++ You can imagine that my response to creep is  because a deep sim is the kind of thing I'm interested in. I totally get why people who saw the original Wing Commander + small MMO idea and wanted that game and not what SC has become.quote:What number are they aiming for? Bearing in mind that we already have models for this kind of thing. 'Whatever they can get' but I suspect the realistic number is 50 per server instance which is what they've given as a basic goal. 50-100 is what they're hoping to get when this finishes. I think it'll start very low. Also, yes, shifting sands. It happens all the time for better (FarCry 3) and for worse (DNF). I'm not convinced it will either fail or succeed, but I am hopeful that something called Star Citizen, the MMO, will come out. It's the kind of game I've wanted to play since I was younger (see giant wall o text earlier D: )
|
|
#
¿
Nov 9, 2015 20:36
|
|
|
Hav posted:This is where I was. Giving to the original pitch for Squadron 42 was a hail mary willingness to throw $35 on something that may not happen, and remains the only reason why I won't ask for a refund. I *already* considered it dead money the second I forked it over. it was the creeping feature list that convinced me that they didn't have a solid plan going in. Rogue System looks like a neat idea, I love the complexity, but I hope he addresses the technological progress. Basically, being able to do a difficult manual landing is cool, but lacking an autodocking mechanism is a bit silly. I like the compromise E:D made in this regard (autodoc computer), and I hope he'll do the same. I haven't seen any combat yet. One that I've been very interested in is Enemy Starfighter: http://enemystarfighter.com/videos/ Best feature? It takes a realistic look at the future, pilots remote in to control drone ships and you, as the player, can shift in and out of different ships while giving RTS-like orders to the fleet (and it's multiplayer!)
|
|
#
¿
Nov 9, 2015 21:36
|
|
|
Hopper posted:Are you sure Ilfonic delivered bad work? Some people seem to favour the idea that Illfonic delivered an FPS module that Croberts was unable to implement into his game because in a move of retardedness he switched to 64 bit whatchamacallit after handing off the work to Illfonic. I have no inside information, so that's me reading between the lines. I could be entirely off-base and what you said above could be the reason.
|
|
#
¿
Nov 9, 2015 21:46
|
|
|
Enchanted Hat posted:I mean, yeah, it's pretty obvious that having this card is going to get you laid, but it's still a dumb thing for them to be selling. It was a suggestion made awhile back in the Concierge forums. Some of the fans really want to show a crazy amount of support. Accounts reach Concierge access at $1,000 spent, I believe.
|
|
#
¿
Nov 9, 2015 22:10
|
|
|
dyzzy posted:Normally I'd expect a gimmick souvenir like this to come as part of the package when you give 1,000 or more to crowdfund a product and the access to social events to be an implicit understanding, but I guess I'm just an enemy of PC gaming This is the case. Concierge and Subscribers are emailed if the venue is limited so they have first digs on getting tickets. I believe Subscribers get first digs on tickets, but I don't pay much attention to the convention stuff.
|
|
#
¿
Nov 9, 2015 22:16
|
|
|
Chalks posted:Is there a figure on how many people are at this level? Not that I've seen. I don't really keep track on the social side. I kinda go through phases of following for awhile, then wait for something new to come out, then follow again for a bit. Right now I'm interested in AC 2.0. Once that's delivered and I've played around a bit, I'll go back to other things (Fallout 4 and MGS V, which I haven't played yet).
|
|
#
¿
Nov 9, 2015 22:24
|
|
|
dyzzy posted:
Otherwise they couldn't have this glorious thing: https://www.mainperformancepc.com/product/obutto-gaming-cockpits/obutto-r3volution-gaming-cockpit/ I am totally trying to save for one
|
|
#
¿
Nov 9, 2015 22:38
|
|
|
Madcosby posted:lol does that only support one monitor do they even know their audience Following the Star Citizen model, it has Upgrade Packages: https://www.mainperformancepc.com/p...ount-accessory/ Were I a virgin, I would happily lose it to this thing. Trying to build my own was an ego-deflating experience. And then: http://gamechair.net/
|
|
#
¿
Nov 9, 2015 22:44
|
|
|
theultimo posted:Ever had chip and pin cards, p much the same Almost all of the merchants out here don't take chip cards despite them being 'required' as of October 1st. So far I've found one place that had a reader for Chip and Sign that worked.
|
|
#
¿
Nov 9, 2015 23:06
|
|
|
|
| # ¿ Mar 23, 2025 06:39 |
|
|
OhDearGodNo posted:Great, now I'm going to have to put effort into a post. Just to save some time, my source for the GTA V information was here: http://www.theguardian.com/technology/2013/sep/07/grand-theft-auto-dan-houser and http://www.develop-online.net/studio-profile/inside-rockstar-north-part-1-the-vision/0183989 The latter: "For GTA V, that discussion started as GTA IV was wrapping up – almost five years ago – although the latest game has been in full production for just three years." Similar, though longer for Fallout 4: "The development spanned seven years, delivering successes, failures, and unexpected detours like porting Skyrim to the Xbox One (keep reading). Just 25 days before Fallout 4 releases, the creators sound off on the lengthy development process" source: http://www.gameinformer.com/b/features/archive/2015/11/06/the-making-of-fallout-4.aspx?PostPageIndex=1 The first two years for FO4 were mostly preproduction (artwork) and setting their target, location research and technology research, like deciding to switch to a deferred renderer. Then full production slowly ramped up. One thing to note is that Bethesda's RPG team is significantly smaller than CIG. I recall the article mentioning they've stuck with a team size of around 100, compared to Rockstar's size of over 1,000. That's what I based my statement on, I just didn't link the GTAV blogs to keep the size down.
|
|
#
¿
Nov 10, 2015 01:09
|
|