Sabreseven posted:

It's from last year, just before they split the game packages to squeeze more money from potential pledgers

Chalks posted:

Yeah, when you see backers saying the game is guaranteed to come out this year don't mistakenly assume that the quote is less than 12 months old.

code:

if (month < 9) {
	return "It's coming out this year!";
} else {
	return "Everyone knew it wouldn't be out this year.";
}

Since 2014, this has accurately simulated Star Citizen backers.

AlmightyPants posted:

Man, remember the MoMA from like Spring-Fall that was actually starting to acclimate to SomethingAwful? Like he rolled with punches and was occasionally entertaining when he wasn't shilling or outright lying. There was a certain playfulness to him too. I wonder what happened to that guy? Ever since the CitizenCon disaster MoMA has been rude and humorless with a propensity for insults and attacks. That was a pretty drastic switch too, I'd suggest account sharing if the voice behind the account had changed as dramatically as the tone.

Sillybones posted:

Rewind 5 years and borrow some of the people who made Planetside 2 netcode. Roll your own engine instead of using Cryengine. Work on the base systems of an MMO instead of making three singleplayer games and trying to hack them together along with multiplayer support. Or, burn the loving project to the ground, sell the assets and leave the country for some unknown island forever.

Rudager posted:

I totally forgot about Planetside 2, which brings me onto my second point, Planetside 1 was what 2003'ish and was the 1 planet version of the FPS MMO Star Citizen wished it could be.

I remember trying to play that through a 56k modem connection and having a blast.

Crazy_BlackParrot posted:

Y U GUYS NO POSTING?!

AP posted:

https://www.youtube.com/watch?v=vI-ZivgsOXc&t=25s

Galarox posted:

Just been messing about browsing rubbish on youtube and I noticed a thing. A LOT of the "Games coming in 2017" listicle things seem to have Star Citizen on their list. Anyone have any idea why that might be?

Here are a couple of examples, there are more if you can be arsed to look for them: https://www.youtube.com/watch?v=8EmaSZljqNE https://www.youtube.com/watch?v=ov3KnO3YfvE

I wonder if this will help prolong funding and keep the shitshow rolling for longer...

Pope Corky the IX posted:

Do you think Ben's penis looks like a pig tail?

Scruffpuff posted:

The reason is the saturation of wanna-be game media "journalists" who don't have the established reputation, or balls, to wind up on the wrong side of this. Predicting the failure of Star Citizen, despite its near guaranteed eventuality, is still a risky endeavor. Much safer to include it. If Star Citizen succeeds, they can validate themselves with the new cool kid on the block by claiming they were rooting for them all along. If it fails, they get fodder for free follow-up articles along the lines of "What happened with Star Citizen?"

Reporting on its failures is less rewarding, regardless of it being more accurate. You're pissing off CIG in the here and now, getting toothless threats from their attorney, and attacks from a cult. Even when they're proven right, it's not like they'll rise to prominence as the "one company who had to balls to tell it like it is." They'll just get more death threats from the backers for being part of the reason SC failed.

It could also be that CIG is paying for this coverage, which would explain where some of the money is going, since they're certainly not making a game.

Galarox posted:

Interesting, I agree particularly that CIG might be paying for the coverage, as that would be a cheap way for maximising publicity and ties in with your suggestion that the listicles are created by wannabe-hacks who would, I suspect, do anything for cash & be very cheap too. Some of those listicles have a LOT of hits. I hate to admit it but IF it is an active marketing ploy by CIG its actually quite clever - very cynical, but clever.

VealCutlet posted:

https://www.youtube.com/watch?v=vZ-Md56jK9Y

Scruffpuff posted:

The reason is the saturation of wanna-be game media "journalists" who don't have the established reputation, or balls, to wind up on the wrong side of this. Predicting the failure of Star Citizen, despite its near guaranteed eventuality, is still a risky endeavor. Much safer to include it. If Star Citizen succeeds, they can validate themselves with the new cool kid on the block by claiming they were rooting for them all along. If it fails, they get fodder for free follow-up articles along the lines of "What happened with Star Citizen?"

Reporting on its failures is less rewarding, regardless of it being more accurate. You're pissing off CIG in the here and now, getting toothless threats from their attorney, and attacks from a cult. Even when they're proven right, it's not like they'll rise to prominence as the "one company who had to balls to tell it like it is." They'll just get more death threats from the backers for being part of the reason SC failed.

It could also be that CIG is paying for this coverage, which would explain where some of the money is going, since they're certainly not making a game.

Galarox posted:

Interesting, I agree particularly that CIG might be paying for the coverage, as that would be a cheap way for maximising publicity and ties in with your suggestion that the listicles are created by wannabe-hacks who would, I suspect, do anything for cash & be very cheap too. Some of those listicles have a LOT of hits. I hate to admit it but IF it is an active marketing ploy by CIG its actually quite clever - very cynical, but clever.

Pope Corky the IX posted:

Do you think Ben's penis looks like a pig tail?

spacetoaster posted:

EminusSleepus posted:

Lol! Still better than the wing commander movie

Nicholas posted:

A fact I haven't seen brought up with regards to CIGs plan for infinite scaling using AWS, is that while new servers can be spun up to increase player connections, the database does not scale at all. As soon as you clone a database for a new instance it gets out of sync, and you can't just merge all the new information together in later on. All clients must be communicating to a single database. This is probably the reason why Lumberhard specifically states that it's not suitable for anything persistant across instances (such as an MMO)

I'm not sure how most MMO's handle it; I'm guessing it's probably a combination of sharding servers, having a hard cap on the number of simultanious connections (so definitely not "infinite scaling"), and being really careful about what each client is writing back to the database (so no millions of NPCs doing things that update tables constantly for no reason)

Scruffpuff posted:

The cascade effect of even the tiniest change on the entire system is astounding.

Facebook Aunt posted:

In EVE Online they added Time Dilation (TiDi).

Sabreseven posted:

It's from last year, just before they split the game packages to squeeze more money from potential pledgers

Hav posted:

Firstly, deferred updates on some things. These are writes that can be queued. Also use a table driver that handles row level locking rather than table level.

Sharding in database terms usually means splitting actual tables, plus you can use temporary table schemas to handle adhoc transaction states that are 'bundled' back into the main table.

After that, you're into the heady world of binary replication and master/slave configurations. Binary replication is effectively a tight communication line to the slave that just replicates any changes on the master.

But no, you're entirely right, database scaling is a absolute pig. Row locking can be an arsehole, Sharding increases complexity and replication - I had a database fall out of sync once with a slave, and it took 45 minutes to rebuild from the binary logs; a straight dump and rebuild of the master would have taken fifteen. Admittedly that was late 90s, but the second you step out of the technological comfort zone, costs skyrocket.

Eve online had this problem with their MSSQL DB, so the first thing they did was slap it on the biggest SAN array available. They already sharded the systems themselves, but it still gave them an upper limit on number of things they could track that had nothing to do with Moore's law or elastic compute, and more to do with the honking bottleneck around table writes.