ALL BLOG POSTS AND COMMENTS COPYRIGHT (C) 2003-2017 VOX DAY. ALL RIGHTS RESERVED. REPRODUCTION WITHOUT WRITTEN PERMISSION IS EXPRESSLY PROHIBITED.

Monday, May 29, 2017

Star Citizen: two takes, one conclusion

First, Derek Smart's observations of the multiplayer system:
This game was never supposed to be an MMO. And it wasn’t pitched as one. And Chris has gone on the record several times, even after all the stretch goals funding were met back in Nov 2014 at $65 million, saying that it wasn’t. And the stretch goals have no such indication or implication that they were building an MMO. Somewhere along the line, because of scope creep and promises made as they pulled every trick in the book to keep raising money from gullible backers, it morphed into an MMO because that’s the only game model that would support some of the things they were promising. And they’re doing all this despite the fact that they neither have the tech, nor the talent, or the time and money to pull it off.

At this point, as I’ve shown above, if they don’t have the framework for their future networking model already in and working in some fashion, there is absolutely no way they’re going to have time to gut what they have now, and implement a proper solution. Something they should have done from the very start. Now it’s too late. And they are still making promises they can’t keep, even as they continue to defer* promised features into a post-release schedule.

My guess is that the current networking layer is going to remain as-is for quite some time, as they continue to build other features and systems on top of it. Then if by some miracle they survive (they won’t) long enough to actually get around to it, all that stuff they are building on top of a network layer they have to replace, will either have to be ripped out, or modified to support whatever it is they need to do in order to support their long term goals.

All of this means that even if they are around long enough for a 4.0 schedule to go live, and it does include the major networking features they need to make what they plan work, until then, backers are still going to be stuck with 8 player clients in Star Citizen. I can’t wait to see what happens when 3.0 goes live with the two moons. It’s going to be hilarious. Maybe they’ll shock everyone and have 6 clients running in Crusader without problems.

Anyone who still has hopes that this project is ever going to be completed, let alone as promised, is delusional.
DevGame has independently reached the same conclusion:
The problem is that literally all the various engines are incredibly restricted with regards to their multiplayer capabilities. They are all match-based at their foundation. What Star Citizen is pretending to try to accomplish, does not exist and never will exist, due to the restrictions everyone has to live by.
The astute reader will, of course, wonder why DevGame has been researching this issue so deeply.

Labels:

32 Comments:

Anonymous Harambe May 29, 2017 9:05 AM  

I always just assumed you were morbidly curious. But then I've also "known" you for more than a decade so I assume there's more to it than that.

Blogger Bellguard May 29, 2017 9:57 AM  

Your post description reminds me a lot of No Man's Sky.

Such high, high hopes for that game.

Blogger Sam Spade May 29, 2017 10:01 AM  

I backed the game like 3 years ago. I don't regret, but the truth is that I lost the illusion.

Like a lot of people has said, feature creep has destroyed the proyect, instead of focusing, he spreaded his ambition to something almost impossible to manage.

I still respect Chris Roberts, he has made mistakes but I think they have been in good will to make something really memorable, and a landmark in gaming and probably entertaiment in general.

At least I hope other developers learn form his mistakes. The best games usually are focused efforts thats excel in a few aspects like Wizardry, Dark Souls, Doom or Ninja Gaiden.

Blogger Jack Ward May 29, 2017 10:03 AM  

Would it even be worth it to buy out Star Citizen in a fire sale? Maybe if some one[group] had already figured out the problems and were ready to rock and roll. I could see Markku, in all his copious spare time, reworking citizen, just for the fun of it.

Blogger Zak May 29, 2017 10:06 AM  

I would be happy with just Squadron 42 being released with a nice little galaxy to explore in co-op. It seems like the chance of that even happening at all, much less in 2017, is getting worse and worse as they won't even commit to a release date anymore. They would inevitably renege on it, but at least it would mean some knucklehead at RSI thinks it could possibly be released by then.

Guess I should finally try and get that refund I've been putting off. Good thing I only got the smallest package.

Blogger Sam Spade May 29, 2017 10:11 AM  

@2. No man Sky always looked like a poor man's Elite. If you want something alike you have Elite Dangerous now. Nice graphics and GODLIKE combat, but very poor mission design. The procedural generation for missions was bad executed.

Was fun for like 30 hours though.

All I wanted for Star Citizen was a goood campaign, with interesting missions, good combat, entertaining plot and atmosphere. A modern Freespace 2 or Wing commander.

Blogger None Needed May 29, 2017 10:23 AM  

The problem doesn't lie in engines as is.
They are made for instances because the network load on the server scales roughly squarely to the number of entities that have to be synchronized on clients.
It's not only players, it's everything that has any gameplay value and "exists" in the world of the game, like projectiles, debris, all that stuff.
Instancing allows you as a developer to be somewhat sure about the amount of data you'll need to receive, aggregate and send out at least ten times per second.

You can do a lot of clever things, but it'll bleed into the game design inevitably.
For example, if you're doing a target MMO, your task is significantly easier (though still improbably hard).
If you decide to use only hitscan weaponry, that's also good.
If you can afford (from the gameplay perspective) less ticks per second (see time dilation in EVE Online), that's even better.

I think, the only non-target MMOs built from the grounds-up with the vision of unscaled mass PvP were Planetside 1 and 2.
They aren't terrifically popular.

Elite: Dangerous uses heavy instancing and tries to offload some tasks to peer-to-peer connections which is even a worse move from a MMO perspective.
Granted, E:D itself doesn't know what it is and it shows in the lackluster development process, but 2.1 million copies sold in a seemingly niche market makes you think that there is some profit there.
Just not in the MMO way, I personally believe, but in Borderlands coop style.

Or you can try to sell jpegs for eternity and get the same money... with a small caveat that FD can do whatever they want with the money, but CIG kinda has to make the game (or to imitate development until money ends).

Blogger None Needed May 29, 2017 10:32 AM  

Jack Ward wrote:Would it even be worth it to buy out Star Citizen in a fire sale?

They don't yet have a game.
They have a ton of assets that age mercilessly (as assets tend to do) and a ton of liabilities, but not much more.

If you have some four or six million just lying around, I'd suggest to look closely at Independence War 2.
Now that's a game that is still fun to play despite being released in 2001 and in an unfinished state (not EA-like unfinished, but they had to cut out the third act and rush the fourth).

Anonymous SciVo de Plorable May 29, 2017 10:42 AM  

Harambe wrote:I always just assumed you were morbidly curious. But then I've also "known" you for more than a decade so I assume there's more to it than that.

I think it's basically "I wish he could pull this off for real, but no. Here's what not to do:"

Anonymous SciVo de Plorable May 29, 2017 10:45 AM  

Oh wait. DevGame wants to explore how to do it right.

Blogger VD May 29, 2017 10:45 AM  

Would it even be worth it to buy out Star Citizen in a fire sale?

No. Not at all. That's why no one to whom they're shopping it, like Amazon, are willing to get involved. I suspect the switch to Lumberyard was primarily in order to get Amazon to come in as an investor.

Anonymous HWNIU May 29, 2017 10:50 AM  

@Sam: Check into Freespace Open. Modernized graphics, new ship models, and user made campaigns that are better than Freespace 2's. And free if you already have Freespace 2.

Blogger Josh (the gayest thing here) May 29, 2017 10:52 AM  

Would it even be worth it to buy out Star Citizen in a fire sale?

What exactly would you be buying?

Blogger Ceasar May 29, 2017 10:52 AM  

"...there is absolutely no way they’re going to have time to gut what they have now, and implement a proper solution."

I see this all the time in IT. Business lines promise the moon to customers/investors. Their Tech Dev and Support leaders won't say no to the business leaders and they both go down a path where they throw good money after bad. They are consumed by The Sunk Cost fallacy when they would be better off developing in a greenfield.

Blogger Shimshon May 29, 2017 10:56 AM  

"What exactly would you be buying?"

A slightly-used low-mileage spaceship, cheap. Driven only on Sundays by a little old lady. We call it the "Pasadena."

Anonymous aegis-1080 May 29, 2017 11:17 AM  

Scope creep isn't a problem that is fixed with tech since it isn't a problem about tech. Is a human problem caused by the human failure of the management of the project. The whole thing just proves that Chris Roberts is an incompetent that shouldn't be trusted with project leadership again.

Blogger VD May 29, 2017 11:21 AM  

The whole thing just proves that Chris Roberts is an incompetent that shouldn't be trusted with project leadership again.

He's not incompetent. He's an excellent designer. The problem is that he's not a producer. This is the sort of thing that ALWAYS happens when the designer cannot be controlled by the producer.

See Age of Conan for the opposite problem, when the producer tries to fix problems that are fundamentally design-related.

Blogger Josh (the gayest thing here) May 29, 2017 11:25 AM  

See Age of Conan for the opposite problem, when the producer tries to fix problems that are fundamentally design-related.

Can you elaborate on the differences between production related and design related problems?

Anonymous Harambe May 29, 2017 11:25 AM  

RE: Scope creep. I once got a spec that read like this "we need a page on our existing website that lists nutritional information for various types of food". Okay, I said. Let's say two days' work, including me capturing the 1000-odd data items you have. Over the course of the next 3 months, the project turned into a full-featured CMS with workflows, multiple sign-offs and paid ads.

It was a tiny project which managed to balloon to roughly 33 times its size, measured in man-hours. And it was caused by poor communication by the client. That, or deliberately misleading wording to try and get me to work for free. Which, thank goodness, didn't happen. I got paid for every single hour I spent on that thing.

Still, I could've easily given them a fixed cost estimate and have screwed myself royally.

Anonymous MrNiceguy May 29, 2017 12:15 PM  

I'm going to have to look into this. I've been jonesing for a good space combat Sim for a while

Anonymous BBGKB May 29, 2017 12:16 PM  

If you have some four or six million just lying around, I'd suggest to look closely at Independence War 2.

It's sad that I have to ask but is that for buying the game or buying advantages in the game? I could never bring myself to play a game that I knew griefers spent the equivalent of multiple bars of gold for an advantage.

Blogger Matthew May 29, 2017 12:57 PM  

Josh (the gayest thing here) wrote:See Age of Conan for the opposite problem, when the producer tries to fix problems that are fundamentally design-related.

Can you elaborate on the differences between production related and design related problems?


My take:

Imagine a game as a new universe (HELLO MORMONS!).

The designer works wholly within the new universe: he is responsible for making the universe interesting and coherent. The designer has to understand what makes for a compelling and enjoyable experience.

The producer works wholly outside the new universe: he is responsible for its implementation. The producer has to understand what can be done given a specific set of finite resources (men, money, technology, time).

A design problem is when the universe is boring, or full of holes, or full of unexpected punishments for what seems like normal behavior.

A production problem is when a well-designed universe is implemented poorly (runs at 3 fps on a beast PC, or has obviously cheap and repetitive art) or cannot be implemented at all.

The producer must provide constraints on the design at all stages.

Blogger Sun Xhu May 29, 2017 12:57 PM  

Camelot Unchained.

Blogger Markku May 29, 2017 1:42 PM  

full of unexpected punishments for what seems like normal behavior

For a bad design problem see: Universities

Anonymous Severian May 29, 2017 1:46 PM  

If only they focused on a solid single player open world instead of this mmo crap, it could have so much potential.

Anonymous FP May 29, 2017 1:58 PM  

At this stage Roberts would have been better off buying the rights to Freelancer back and starting from there.

Blogger Tatooine Sharpshooters' Club May 29, 2017 2:25 PM  

The whole thing is staring to sound like Star Ponzi: Trapped in the Madoff Nebula.

Blogger en_forcer May 29, 2017 4:34 PM  

There is so much data and so many red flags regarding Star Citizen. No one should be giving a single dollar towards it.

Blogger None Needed May 29, 2017 5:40 PM  

> It's sad that I have to ask but is that for buying the game or buying advantages in the game?

I understand "buy out on a fire sale" as in "achieve full control over the IP, codebase and assets when CIG will go belly up and make it right".
I'm not a native speaker, though.

4 to 6 million are my assumptions on what'll it cost to make a game like I-War 2 in modern times.
Single/coop, of course, not a MMO.

I don't think that Independence War as an IP may add something to sales, but finding out who holds it now is tricky. It may not be worth it.

The problem is that the budget places the game into an uncanny valley of sorts.
You need about 200k of sales to break even and it's kinda hard for a new game in a niche market.
Everspace, for example, is slowly inching to 80k and is unlikely to break 120k this year.

Blogger wreckage May 29, 2017 10:26 PM  

@29 The problem with the IP is it's worthless, the assets depreciate off the books virtually overnight, leaving you with an engine and netcode; the latter being probably worthless and the former being better off re-coded from scratch.

See indie game Natural Selection 2. They had a narrow focus, and as a result decided partway in that it would be faster and better to code their own engine from scratch than to work with a pre-existing engine. They did it, it worked, lesson: the games' code isn't worth jack.

For an odd hybrid of MMO and 4-player co-op, see Warframe. It essentially started with the netcode and some novel client-server jiggery pokery, then added systems on top of that; but the game developed in the exact reverse process of what we're looking at in Star Citizen.

Blogger beerme May 29, 2017 11:41 PM  

@30 From an outsider's perspective, the way they've gone about creating Star Citizen is hilariously backwards and the equivalent of starting out production of a car body when you don't even have a tested drivetrain to put it on.

Or in a more relevant case for Chris Roberts, trying to make Wing Commander 3 and 4 without creating the engine earlier via Strike Commander. The development path of Strike Commander sounds eerily similar to what is happening on Star Citizen.

Blogger WK May 31, 2017 7:30 AM  

Put aside what was promised, maybe SC will take the direction SWTOR took. That game is a MMO, but the newer expansions were aimed at single player or co-op with a few friends.

Instead of 900 people grouping and heading off together, it will be a small party doing Mass Effect like missions. Parties limited to 4 or 8, large ships crewed by NPCs, the universe a 'sandbox' because these small parties can leave a mark.

Think the last Riddick movie. Solo exploring planet, finds outpost made by another player. Small parties arrive and interact.

I am not a computer guy, but it sounds to me that Smart's problem (one of his problems) with SC is the number of players who can play together at the same time will be low. Space battles with thousands of ships looks to be impossible for this engine, Smart says clearly so to a competent designer. Maybe it is an honesty problem. The game might be headed to solo/small co-op and the fear of backlash/lawyers is keeping the designers from admitting this until there is a game to point at.

That there is no game to point at after all this time...

Post a Comment

Rules of the blog
Please do not comment as "Anonymous". Comments by "Anonymous" will be spammed.

<< Home

Newer Posts Older Posts