(tweet: Elite: Dangerous Twitter - People playing the game and doing things together. Star Citizen Twitter - Twitch streamers giving away ships, ship raffle entries.)
I mean c’mon charlie, even chat doesn’t fucking work in star citizen
what else do you expect them to do
The only bug-free program is one with zero lines of code. Star Citizen already has more bugs than any game I’ve ever played so how could it be a scam lol wiggle your way out of that one, Goon bitches.
Another leaderboard update, and the numbers are dropping faster than Bitcoin. I almost don’t trust these numbers and feel like something’s wrong, but I haven’t changed my methodology and CIG’s leaderboard code also hasn’t changed as far as I can tell. We report, you decide.
Total concurrent players for Jan 29 through February 1:
Hello again commandos. Performance Theory back again, drawn by the siren song of this lovely screenshot.
Full disclosure: I have not seen all of the source material. I’m basing this off of what I know from this screenshot and from what I have witnessed in past displays of CIG’s code. I could go into even more detail about minutia, but will endeavor to keep this brief and focus on the major pain points.
if (abs(speedA) < 1)
speedA = 0;
if (abs(speedB) < 1)
speedB = 0;
**This is what is called a kludge. Some bit of functionality which is added in order to fix a separate issue. A band aid, if you will. It has the short-term benefit of solving whatever the problem is. Presumably, the root cause is that their physics engine runs into issues with near-zero velocities. The correct and proper solution would be to resolve whatever issue is causing that unwanted behavior.
Instead, they kludge it.**
This may solve the issue in that specific instance. But, odds are, someone else will eventually try to do something very similar and will run into the same issue. Hopefully they communicate well enough that the known fix is shared and simply copy and pasted in that new area. This is a poor design practice as it leads to code bloat and makes the codebase less maintainable, but at least not too much time will be lost. Unless it’s someone else who runs into it who has not been over this particular piece of code where the kludge is stored, and/or the right person is not asked, and/or any of a thousand other scenarios in which the developer is forced to spend a bunch of time debugging until they discover the same kludge.
I could go on, but you get the idea. This type of practice is not good for Star Citizen.
if (!IsBlame(speedA, speedB))
{
return;
}
I saw some discussion from people who seemed to think that this was a really bad thing. They were right, because it is a really bad thing. But perhaps for slightly different reasons.
There are actually two issues hidden in here. The first is that it relies upon the function being called twice. This is actually okay in the very specific context of collision response code. A properly designed physics engine would create an event for each collision, which collision objects could then handle. Every collision event would logically have at least two participants. There isn’t actually anything wrong with this approach.
The second issue is function control flow which begins here and continues in the next conditional statements. In other words, we jump out of the function and skip the rest in some cases. This design makes a lot of sense.
Except it leads to the inevitable creation of enormously bloated superfunctions which contain arcane sets of complex business logic.
In enterprise software, business logic frequently changes. Perhaps the way an interest rate is calculated, for example. A naive approach would be to have a function called GetInterestRate() which returns a hardcoded mathematical function. This makes sense and works well. Until it changes, later on.
Then, you have to decide what happens when you change GetInterestRate(). What if you need to interact with older data sets which rely upon the original function? Are you going to just create GetInterestRate2() and remember to call that if older data is detected? What happens when it changes a second or third time? Etc. Probably best to just put some conditional logic inside of GetInterestRate() which checks what formula we need to use.
But what happens when you now start doing business with Elbonia, the land of waist-deep mud? Their interest rate is calculated based upon the square root of the width of the shadow cast from the nearest village’s oldest pig at noon. Sounds like we need to add yet more logic to GetInterestRate() in order to handle that edge case.
A few mergers and acquisitions later, and GetInterestRate() is one thousand lines of code long and you can never fire Bob, because he’s the only one who remembers how it all works.
Now, games are different than enterprise applications. Game developers cheat a lot, especially when it comes to performance optimization. But poor design practices can be universal, and bloated superfunctions like that are incredibly common.
A better design would involve a proper refactoring of the codebase so that you don’t need a mountain of brittle conditional checks. Most physics engines utilize an event-driven paradigm, wherein objects simply subscribe to certain types of events that they want to listen for, such as OnIsCollisionBlame, OnVehicleHostile, etc. Encapsulating each of those conditions into a collinion information tag, which the physics engine checks once and assigns to the raised collision events, is a fantastic way to support that very same logic in a much more robust, easier to use, and less error-prone manner.
–-
But, game developers are only human. Humans make mistakes. Humans get tired and sometimes even want to go home after working 12 hour days trying to do the impossible. We love it, because doing the impossible is exactly what we signed up for, and we wouldn’t have it any other way.
But sometimes that means that we end up writing superfunctions instead of painstakingly redesigning the way in which the physics engine communicates.
Sometimes it works out okay. I wrote my fair share of code that makes the above look like a paradigm of amazing design. Sure, I was working 16 hour days at a time. Sure, the deadline I was under was impossible. But it worked out okay in the end.
I genuinely hope it works out okay in the end for Star Citizen, because I have enormous respect for those who are in the trenches and trying their best. Maybe they simply don’t have the experience to recognize these design pitfalls. Because I am quite certain that the average Joe Programmerdude is just trying to do his job and collect his paycheck, and maybe do something cool in the process, rather than sitting back, rubbing his hands together, cackling and muttering evilly about how he’s going to do the worst hack job he can get away with.
That said, we should ask why after $175,000,000 and 7 years, core engine features fall victim to this.
MoMA
LOL you fucking pansies are so scared of risk. I had money when I bought into SC. I have money now. Little dip in crypto is not affecting me - trust me.
When a kickstarted belt is the least of your stupid financial decisions
It’s been one month after the premature disaster launch of 3.0. Ignoring that most of what 3.0 was supposed to be was pushed to 3.1, it is simply unacceptable on all levels that nothing has been done about it. That there is no visible effort to address it. You may say; 3.1 will fix it. And I say; when? This should have been fixed yesterday. And why should I believe that 3.1 will help? They have never delivered in the past. Not one single deadline has ever been met. And most releases have been lacking compared to what they said they would have. So why is 3.1 any different?
I was patient when they missed the original November 2014 release date. I thought it was too optimistic from the start. I was patient when Squadron 42 Episode 1 failed to materialize. I was patient when 2.0 released full of bugs and with nearly no content, thinking they would deliver on their promises for monthly updates. I was patient when Star Marine failed to materialize, thinking it was just a testing module. I was patient when 3.0 was delayed for weeks, then months, then more months, and then a whole year. They said it was almost nearly done in December of 2016. Yet they talked about just having started developing basic features for it in October of 2017. How can both be true? They were blowing smoke up my ass for an entire year. So why should I believe anything they say now? I don’t want to hear any more lies. My patience has run out. I want action now.
You give people shit for calling it a scam but that’s exactly what it looks like. Give me one good reason not to pursue a refund.
1:08:20 - ‘Even before [Derek] latched onto this thing. We were here calling out CIG. Mr Smart dialed that up to 15 where we were more around a 6. We were saying a lot of the same things, he was just saying them in a lot less nice way. A lot of what he was saying was hyperbole and some of it didn’t check out; as time has gone on a lot of what he said have come more or less true, and so as time has gone on it’s harder to dismiss him as being uncredible. He has gained credibility over time in my view.’
The one thing that really makes me laugh is the response “no one is making you spend the money they aren’t forcing you” thus that means it can not be a scam. Scams are predicated on fooling or otherwise taking advantage of a person to gain something of value from them willingly. Taking something from a person against their will is theft\robbery not a scam.
TheAgent
crytek and cig decide to settle this by a one on one jacking session where the first person to shoot a burly load loses the case
cig picks ortwin, whose dick hasn’t worked for anything but barely legal vietnamese sex workers
skadden choses derek smart, who has no penis or a huge penis or something with his penis because every backer can’t stop fantasizing about his dick size
“I’m almost there,” Ortwin pants, clutching his hard cock and the right side of his legal pad. “I can’t help it.”
“Think of your blonde, big-titted ex-wife, Orttie,” Jared delicately whispers into his ear. “Wasn’t she successful? Didn’t she challenge you? Wasn’t she one of the only female directors in Hollywood?”
Ortwin relaxes. The sweat on his brow fades, his face calmer, more serene. “Thank you, Jared. Thank you.”
“Don’t worry, buddy. We’ll win this.”
Jared glances over at the plaintiff’s table.
“Just let me finish!” Derek Smart screams. “Just let me fucking finish!”
Jobbo_Fett
MoMA
Ok, honestly? I thoroughly enjoy this genre, and there isn’t quite anything out there like what SC aims to accomplish. Yes - we have Eve, Stellaris, Homeworld, Kerbal, E:D, etc, but none of them give you the sandbox, open-world, mmo feel that SC will. It’s niche, but something I (and apparently, over a million others) wanted.
mfw I’m fucking retarded and have to have blind faith in a living thumb accomplishing one of the most complex sets of requirements to fulfill a literal dream game that I write about cause I’m some sort of turbonerd, except even dumber. In 7 years of work CIG has come nowhere near their end goal and even conservative estimates would place their gold release being closer to, if not after 2020. Meanwhile, other companies have or are building games that do some, most, or all that Star Citizen is claiming it does.
Keep in mind the “does” portion, because my high-functioning 200IQ brain that subsists on dreams, shitty salads, and endangered animals keeps forcing me to say shit like “CIG is building the game of my dreams and has all these things included!” except it fucking doesn’t and most of the features that would make it “complete” are months, if not years, off of their initial implementation. I have to force myself to forget that games like Starpoint Gemini, X 4 Foundations, Kerbal Space Program, EVE, Rebel Galaxy, either don’t exist or can’t fulfill every last one of my impossible demands, and which Star Citizen is no closer to it either.
But I’m sure Star Citizen will get there, eventually. There’s no reason to believe that the constant incompetence shown through their forums, twitter, and various video release platforms would ever shake my faith in a videogame that has received over 180 million in funding. Anyways, guess I’ll give them another thousand dollars or whatever because I want a better ship that the company I’m buying it from is telling me will be the best of them all to support development of the most ambitious thing ever even though nobody working on the game has had any amazing breakthrough within the gaming world for over a god damn decade.
MassivelyOP’s latest podcast discusses the Star Citizen Alpha subscription controversy starting about 40 minutes in.
Good lady Bree was bemused but on point as usual:
“I feel like it was probably a bad move given all the folderol from the last the couple of months with Star Citizen but… it’s been years of folderol with Star Citizen so if you’re in it at this you’re gonna dismiss anything.”
Methinks the lady doth protest just fine.
Beet Wagon
I wonder if he realizes the full implications (and how accurate he is) in painting CIG LA as a mismanaged hellhole like the company from The Office
