Derek Smart posted:

RIGHT. SO NOW WHAT?

CIG/F42, if you are reading this, as I’m sure that most of you are, as someone who has a lot of experience with this, and who has made a decent living out of complex and ambitious games, spanning over thirty years, my recommendation for 4.0 (3.0.x is hosed anyway, and there’s no saving that branch):

Forget about making this into an MMO. It’s just not going to happen. You all know this.

I know that you have a headless server. I also know that the code that runs on that server is already in the client in some form or another. Merge that back into the client. It’s easier to go from server -> client than from client -> server in such a merge. The latter is bad. Don’t do it because everything will break.

After the above, create a module running on it’s own thread. All it should do is handle network traffic like a server. It’s basically the code you salvaged from the deprecated headless server. You now have a client and server within the same executable. You could, if you wanted to, split that into it’s own executable and spawn it on-demand if you want. More on this later.

The backend database that handles the client sign-ons, cash shop etc can remain as-is. However, you should plan to be able to disable all of this because this project is going to crash; and when it does, you won’t want to leave backers with an unplayable game. NOTE: Please start using packet encryption for all such traffic. Not going to say more than that; but don’t be surprised to see hacks down the road with players having all kinds of poo poo they never paid for, nor should have in the first place. Whoever wrote this, did a sloppy job btw. Also, you could adopt something like PlayFab, or even move to SteamWorks. Like Amazon with Lumberyard, let a third-party handle all that crap because they will be around longer than this project will.

The current matchmaking is sub-par. Get rid of it and implement Lumberyard matchmaking via GameLift. In fact, as luck would have it, Bruce Brown recently wrote a dev article about it. The goal is for Star Citizen Spergs to be able to invite, meet, and play with their friends in expensive chariots, all in the same session, without having to play guessing games. Yes, there’s a party system but like everything else in Star Citizen, it’s half-assed, restrictive, and needs work.
Then, focus on fine tuning the 16 player experience in this new environment. It’s better to have a LOT of 16 player game sessions in which everyone is having fun, than to have none at all – which is precisely where this project is headed. As you are probably aware, 3.0 is as stable as 2.6.3; and the later is only stable because nobody is playing it.

Having done all of the above, you can go back and release a headless console server to backers, allowing them to setup their own dedicated servers either on their own, or via third-party services like G-Portal, Ping Perfect etc. Note that this satisfies the private servers promise which, as you know, you’re never going to fulfill if you’re making an MMO. This also cuts down drastically on AWS costs which I know has to be killing you guys.
With all of the above as the 4.0.x branch, assuming you survive (you won’t) the 6-8 months it will take to implement, go back and draw up a list of everything Chris promised (use ours). Then pick out the ones that at doable within the new non-MMO framework and present it to backers. Scrap everything else. Yes, backers are going to be mad as hell, but it’s better to deliver something, than to deliver nothing.

Furthermore, Chris’s upcoming plan to start pushing SQ42 as the way to bring in new money, aside from the fact that there is no game there either, is going to fail. It’s a non-starter, and it’s not going to bring in enough money to fund a project that’s burning over $3 million a month. So it’s better to spend the next few months building a stable 3.0.x branch so that backers will have something to play and own when the final lights go out.



Yes, my guess is that some of you may have already considered some or all of the above, but was met with resistance because without Chris and Erin selling MMO, the money train will come to an end. But here’s the thing, I have to believe that there are some of you who are “leads” and “execs” who have some humanity and decency left in you, and which should guide you into seeing this whole thing for what it is now: a complete scam that you’re ALL a part of. The road to redemption and forgiveness, always starts with coming clean and admitting that you’ve done the wrong thing. Yes, it’s shameful, it’s distressing, and it’s career breaking, – but you ALL are going to suffer all of those anyway, so why not do the right thing sooner, rather than later?

You’re all so hosed, that I don’t even know how else to express it. And you have Chris Roberts, Erin Roberts, Tony Zurovek, and Sean Tracy to thank for that. When you have an entire company with four studios around the world, and there is nobody to stand up to the creators, you don’t have a leader, you have a dictator. At some point, it has to stop being about a pay check. I know that work in the industry is hard to find, especially in the UK, but at some point, you have to do what is right and just because there is no excuse for complicity.

Star Citizen is yet another example of why games – regardless of cost – get canceled. If this game were at any developer or publisher, it would have been canceled by now because it will cost more to fix it because the alternative is that you’re in the territory of diminishing returns.

Ursine Catastrophe posted:

Thread recap. Bulletpoints on what CIG should do to salvage what they have. CIG's hosed and I called it. Parp.

Spiderdrake posted:

a stinky piss-baby

SomethingJones posted:

http://dereksmart.com/forums/reply/6078/
I think this is worth posting:

Kramjacks posted:

A couple days ago there was a tweet from a citizen about their new PC they were putting together for SC, including an Intel optane drive, obviously. Someone said the the guy was streaming building it and did poo poo like not use standoffs when putting the motherboard in the case, and not putting any thermal paste on the cpu before attaching the cooler.

Did anything come of that? Did the smoke get out?

SomethingJones posted:

http://dereksmart.com/forums/reply/6078/
I think this is worth posting:

Flared Basic Bitch posted:

The entire history of the internet has led up to this, the moment when I see an actual Geos reference.

Ghostlight posted:

lmao that this thin skinned fruitcake is reading this thread enough to understand beltdowns.

Toops posted:

AP posted:

https://twitter.com/SandiGardiner/status/940366062956060672

AP posted:

https://twitter.com/SandiGardiner/status/940366062956060672

Cao Ni Ma posted:

"Failure is not an option" sounds like a threat

AP posted:

https://twitter.com/SandiGardiner/status/940366062956060672

SomethingJones posted:

http://dereksmart.com/forums/reply/6078/
I think this is worth posting:

SomethingJones posted:

http://dereksmart.com/forums/reply/6078/
I think this is worth posting:

Beer4TheBeerGod posted:

Failure is not an option.

D_Smart posted:

Let me copy and paste a properly formatted version.

RIGHT. SO NOW WHAT?


CIG/F42, if you are reading this, as I’m sure that most of you are, as someone who has a lot of experience with this, and who has made a decent living out of complex and ambitious games, spanning over thirty years, my recommendation for 4.0 (3.0.x is hosed anyway, and there’s no saving that branch):


1) Forget about making this into an MMO. It’s just not going to happen. You all know this.

2) I know that you have a headless server. I also know that the code that runs on that server is already in the client in some form or another. Merge that back into the client. It’s easier to go from server -> client than from client -> server in such a merge. The latter is bad. Don’t do it because everything will break.

3) After the above, create a module running on it’s own thread. All it should do is handle network traffic like a server. It’s basically the code you salvaged from the deprecated headless server. You now have a client and server within the same executable. You could, if you wanted to, split that into it’s own executable and spawn it on-demand if you want. More on this later.

4) The backend database that handles the client sign-ons, cash shop etc can remain as-is. However, you should plan to be able to disable all of this because this project is going to crash; and when it does, you won’t want to leave backers with an unplayable game. NOTE: Please start using packet encryption for all such traffic. Not going to say more than that; but don’t be surprised to see hacks down the road with players having all kinds of poo poo they never paid for, nor should have in the first place. Whoever wrote this, did a sloppy job btw. Also, you could adopt something like PlayFab, or even move to SteamWorks. Like Amazon with Lumberyard, let a third-party handle all that crap because they will be around longer than this project will.

5) The current matchmaking is sub-par. Get rid of it and implement Lumberyard matchmaking via GameLift. In fact, as luck would have it, Bruce Brown recently wrote a dev article about it. The goal is for Star Citizen Spergs to be able to invite, meet, and play with their friends in expensive chariots, all in the same session, without having to play guessing games. Yes, there’s a party system but like everything else in Star Citizen, it’s half-assed, restrictive, and needs work.

6) Then, focus on fine tuning the 16 player experience in this new environment. It’s better to have a LOT of 16 player game sessions in which everyone is having fun, than to have none at all – which is precisely where this project is headed. As you are probably aware, 3.0 is as stable as 2.6.3; and the later is only stable because nobody is playing it.

7) Having done all of the above, you can go back and release a headless console server to backers, allowing them to setup their own dedicated servers either on their own, or via third-party services like G-Portal, Ping Perfect etc. Note that this satisfies the private servers promise which, as you know, you’re never going to fulfill if you’re making an MMO. This also cuts down drastically on AWS costs which I know has to be killing you guys.

8) With all of the above as the 4.0.x branch, assuming you survive (you won’t) the 6-8 months it will take to implement, go back and draw up a list of everything Chris promised (use ours). Then pick out the ones that at doable within the new non-MMO framework and present it to backers. Scrap everything else. Yes, backers are going to be mad as hell, but it’s better to deliver something, than to deliver nothing.

Furthermore, Chris’s upcoming plan to start pushing SQ42 as the way to bring in new money, aside from the fact that there is no game there either, is going to fail. It’s a non-starter, and it’s not going to bring in enough money to fund a project that’s burning over $3 million a month. So it’s better to spend the next few months building a stable 3.0.x branch so that backers will have something to play and own when the final lights go out.

Yes, my guess is that some of you may have already considered some or all of the above, but was met with resistance because without Chris and Erin selling MMO, the money train will come to an end. But here’s the thing, I have to believe that there are some of you who are “leads” and “execs” who have some humanity and decency left in you, and which should guide you into seeing this whole thing for what it is now: a complete scam that you’re ALL a part of. The road to redemption and forgiveness, always starts with coming clean and admitting that you’ve done the wrong thing. Yes, it’s shameful, it’s distressing, and it’s career breaking, – but you ALL are going to suffer all of those anyway, so why not do the right thing sooner, rather than later?

You’re all so hosed, that I don’t even know how else to express it. And you have Chris Roberts, Erin Roberts, Tony Zurovek, and Sean Tracy to thank for that. When you have an entire company with four studios around the world, and there is nobody to stand up to the creators, you don’t have a leader, you have a dictator. At some point, it has to stop being about a pay check. I know that work in the industry is hard to find, especially in the UK, but at some point, you have to do what is right and just because there is no excuse for complicity.

Star Citizen is yet another example of why games – regardless of cost – get canceled. If this game were at any developer or publisher, it would have been canceled by now because it will cost more to fix it because the alternative is that you’re in the territory of diminishing returns.

Hav posted:

Then one glorious, golden shining day there will be something and there will be walling and gnashing of teeth as the unbeliever will smote their breast and rent asunder their T-shirts of the Lord and Saviour God Emperor. As it shall ever be, forever and amen.

...or something that that _effect_.


They see me rolling, riding dirty.

Sarsapariller posted:

Derek with the surprisingly helpful post.

SomethingJones posted:

I don't know

They've even set up a licensing company to collect the royalties on the IP from public performance, you can see the whole thing in plain sight

Normally a publisher collects those for 50%, so they've licensed everything to themselves

This is why they've been scrambling for artists and lighting techs and modellers

SomethingJones posted:

Here's something else everyone missed in their thread recaps:

Last year in one of these blog posts Derek said that a decision had been taken by CIG that had essentially sealed its fate. In this latest one we find out what that was: the decision was taken to not pursue a custom engine and instead move to Lumberyard. It was known last year, at least in private, that the 'engine question' was being looked at seriously and an engine change was in fact on the table as an option. The reason that everyone thought sticking with Cry/Lumberyard was a Very Bad Idea can be summarised thusly; the networking loving sucks and will never ever cope with being an MMO.

Like, everyone knows this. I don't have a loving clue about game engines so I have to defer to people who work with this stuff and they all say the exact same thing - don't try and make an MMO with CryEngine are you loving insane.


Why the gently caress is this important

Because all those guys sat down around a table with Erin and Chris on the old Skype and told them that the only way this was going to be an MMO was to scrap the engine and build one from scratch (other engine options were discussed). The decision to stick with Cry/LY was taken against the advice of their engineers and that led directly to 3.0 being what it is now.