Return to “General”

Post

Re: Don't release the code yet

#76
Lemar wrote:
Wed Oct 10, 2018 3:27 am
[...]

Also tremendous amount of time was wasted on thinking / building towards making it being highly modable? Why would one who never delivered a game waste time on such a thing. Focus focus on the right thing.

Just funny reading the last comments how we all agree that we had some smaller version in mind (even so with some differences on what to leave out) we left this poor guy trying to climb the mount everest (his too big impossible version) instead of telling him more of what we wanted (the more possible version)....one could think we let him fail on purpose :twisted: so that after he handed over the engine we can make it how we want it :lol:
For the first part, Josh was making the game moddable because it would speed up his personal development time by giving him the ability to rapidly add content without having to touch or think about the "core code". I think in the end it ended up being more trouble than it was worth, but as they say, hindsight is 20/20. The majority of people were in agreement when he laid out his reasons for LTSL at the time, so it wasn't like he had a forum full of people saying, "No, that doesn't make sense! Keep doing it the same way you have been."

As to the second part (we let him fail and none of us tried to convince him to do something smaller), I don't believe it's true at all. In fact, Flatfingers made a poll titled "What features would you cut?", which ended up with 130 votes. That's a fair quantity of people - with the option to vote thrice, I'd like to say around 50, which is a lot for one of our forum polls. The things people supported the most for dropping (by a large percentage) were "Moddability", "Slick, highly configurable user interface (UI)", and "Information as a commodity".
Have a question? Send me a PM! || I have a Patreon page up for REKT now! || People talking in IRC over the past two hours: Image
Image
Image
Post

Re: Don't release the code yet

#77
IronDuke wrote:
Tue Oct 09, 2018 7:10 pm
RedDwarfMining wrote:
Tue Oct 09, 2018 6:02 pm
I'd be happy with ONE planetary system!
I've been considering making my own space game for the longest time, and while it would be very cool to have dozens or hundreds of stars, or use PCG and have billions, I have been thinking more and more lately that just one system is more than enough if the gameplay is clever about it. c:
Yeah. One system is challenging enough! Grand Theft Auto series is about one city not hundreds...They get it.

I'd get one planetary system nailed...Then maybe make another.
Post

Re: Don't release the code yet

#79
If the engine is capable to generate and connect hundreds of star system, (its just triggering procedural creation after all), they could stay in the game.

But the area that the AI populates and interacts in should be limited to some core system. For performance reason (too keep the calculations in check), but also for gameplay reasons (then the player is capable to experience the AIs world -and impact it - as a whole)

Outside the core system, the player is free to fly around. maybe find some spawned enemies or loot. And have a nice exploration. But it would not be required.

Unlimited space does not need to mean unlimited populated space...
Post

Re: Don't release the code yet

#80
With respect to star systems, I believe the magic word wasn't "procedural" -- as Damocles points out, that already has to be in the game.

The magic word was "infinite." That had two major implications, one positive, one negative.

The positive effect of spawning new star systems forever was that the game never had to end. The universe had no "limit." For someone like me who wants some games to make exploration their primary play experience, I really wanted this feature. Not just in terms of the simple "explore new worlds" mechanic, where I can take a ship and visit new places constantly for some in-game mechanical reward. That's nice, but the real fun is simply in knowing that there is always someplace new to see just beyond the horizon -- it's knowing that the game never has to end. What an incredible promise!

But a design intended to insure there's no ending does some interesting things to other conventional features of "large world" games. And one of those features is research. As I pointed out several times over the years, how do you design research -- which increases technological power -- so that it never stops being fun no matter how long you play? Either you have to simply stop allowing improvements, in which case all the work you (the developer) put into the research feature stops having any return on that time investment; or you must at some point gradually start reducing the numeric value of each new research result so that it can go on forever... but at some point you're talking fractions of a percentage improvements, which is basically equivalent to no improvement at all. I know people tried to find ways around this -- usually some form of combinatorial sleight of hand -- but I never saw a solution that clearly solved the "eventually useless" problem that, I believe, is 100% inherent in putting a power-increase feature in an infinite game. Those two features do not work together.

So those are the two big effects of dropping the "infinite procedural generation of star systems" feature. Giving that up pretty much means designing Limit Theory so that it has an ending. No more infinite exploration, no more knowing that however big you get, there's always be an NPC empire that can challenge you. But in return for giving up infinity, research remains useful and fun for the entire duration of the game. And not having to expand the universe every time a player-managed asset comes near the edge almost certainly allows for AI LOD performance optimizations (which is probably important).

I think to make LT work, there's one more key design idea that ought to go, and that's the will-o-the-wisp of a closed economy, where every object comes from somewhere and is always affordable because both inflation and deflation are generally under control. That is a Hard Problem in a game. Too many good developers have crashed on that rock. It's a fun simulationist challenge but it's just too hard to be worth trying to implementing it in a game when faucet/drain -- as boring and "unrealistic" as that may be -- gets the job done and can be completed in less than geologic time (so that the rest of the game can be finished).

But yeah. Infinity probably has to go, too.

And then there's the conceptual design goal for LT that anything the player character can do, NPCs should be able to do. Does that design goal also need to be dropped? Or is it something that could mostly be achieved without taking too much development time?
Post

Re: Don't release the code yet

#81
Flatfingers wrote:
Thu Oct 11, 2018 12:24 am
And then there's the conceptual design goal for LT that anything the player character can do, NPCs should be able to do. Does that design goal also need to be dropped? Or is it something that could mostly be achieved without taking too much development time?
I generally like multiplayer games because other players can do exactly what I can, and in that sense I would like to see the AI in LT be able to do the same. Not every AI needs to have the ability to do what I the player can, but to maintain the illusion I think at least some should be running businesses, making fleets and building things.
Image
Post

Re: Don't release the code yet

#82
Flatfingers wrote:
Thu Oct 11, 2018 12:24 am
I think to make LT work, there's one more key design idea that ought to go, and that's the will-o-the-wisp of a closed economy, where every object comes from somewhere and is always affordable because both inflation and deflation are generally under control. That is a Hard Problem in a game. Too many good developers have crashed on that rock. It's a fun simulationist challenge but it's just too hard to be worth trying to implementing it in a game when faucet/drain -- as boring and "unrealistic" as that may be -- gets the job done and can be completed in less than geologic time (so that the rest of the game can be finished).
Apart from the problem of how a closed economy should work in an infinite world (its kind of a contradiction):

Its certainly possible to make a closed production economy, where every good and asset that is used, is being produced somewhere inside the enclosed economy - without spawning it in or making it vanish.

-> But the major problem is combining that with the monetary side (pricing the goods and the businesses cash reserves). The economy can easily crash by running into price deflation (growing economy without new money being created and circling around) or inflation (to much cash created by some system, like the player getting monetary quest rewards), or cease production since the cash reserves of some of the businesses (production and trading) ran out, which then trickles down the production chain halting all other production.

It would be way to hard to get that all into a stable economy while also being an economy that is fun and enjoyable for the player. (and not be obscure, so the player cannot understand any pattern on what is happening).

So to keep that game economy running, the artificial (game controlled) faucet or drain should then not be regarding the goods, but regarding the financial liquidity. Like a very generous AI-Bank, handing out cash to any AI that runs low in funds.
Or maybe even internally ignoring the cash-flow for ai actors, and only have the player have financial limitations.

The game then can stage interventions to keep the economy in check:
-if a certain good is underproduced, the game orders the AIs to create new production facilities / mining operations. (ignoring the financial side)
-if there are too many assets overall (ships, stations) produced, the game orders the AIs to get mental, and start staging attacks against each other, resulting in a higher destruction of ships and stations.

But most important: all in an economy with a limited size.
Post

Re: Don't release the code yet

#83
I don't quite understand why the economy has to be stable? I mean what does that mean? Is It stable in the sense that it will never crash, because that's not how real economies work. If it's stable in the sense that it will eventually return to some equilibrium after a Crash I agree, but I do think that the idea of market bubbles and crashes should not be feared in a sim, if done right it should actually be a sign that you're hitting a viable solution.
A life well lived only happens once.
Seems deep until you think about it.
Post

Re: Don't release the code yet

#84
Maybe a good way to say it is that a reasonable goal is "global price stability."

That is, it's OK if there's some local variation in price stability, as long as for (roughly) every inflation there's a deflation, and for every rare hyperinflation there's an equally rare place where overproduction isn't matched by money increase (presumably because the amount of money available is pegged to a gold equivalent).

It won't be much fun for the player who's just starting a new game if the star system in which the player's character begins immediately lurches into some financial crisis. So there might need to be some artificial local price stability.

But once the player starts expanding factional operations to multiple star systems, I'd think it's probably OK if a few of those neighbor systems have some financial problems, as long as there's a balance overall so that the whole universe doesn't suffer an economic collapse. That might be too hard to recover from!
Post

Re: Don't release the code yet

#85
Hi.

So ive been lurking around this place for a few years now, and given whats happened i felt the need to finally create an account.

Without going into why/how/what could have been, given the current situation its in my view a case of what to do from here rather than dwell on the past.
However just for the record im sad to see that josh ran out of steam, but its not too surprising, and i can accept that. I was looking forward to this game like many others, and its always a shame to see nice things run into problems.
I can certainly understand that Josh would not be interested in working on this anymore, and thats absolutely fine in my eyes, but if i would be so bold theres just a little more that i would ask of him (if youre josh, please do read the "decisions for josh" bit! thanks!)

Regardless, that leaves this community (dare i say us, given i just joined) with a few options, one of which is to attempt to finish the game when josh releases the source.
This one is the most interesting imo, so im going to talk about that.

So a little about the success of open source projects:
Spoiler:      SHOW
Ive been involved with other open source projects for a few years now, and ive seen some things. Not too much but enough to get a feel how these things would go.
First, while it depends on the code state, its certainly possible to finish this game given a team and proper organization. They dont need to be paid, volunteer efforts can work just as well, but there are nuances we will get into later.
The point is, that other community projects of similar scale have been successful in the past (see: Krita, Atom, the SpringRTS community for an RTS engine)
Some have community funding, some have barely any funding, but its certainly possible. There are very large mods around which represent huge amounts of work that dont even have access to the engine source code, and create things which are pretty huge, all of the people working on them in their spare time. It turns out that motivation is a hell of a thing. Seeing Josh go at it for 6 years is a testament to that if anything.

The next thing is that whether this is successful or not depends on the management of the project the most in my view. How the source is released, under what conditions is very important here. To ensure best success i would create a github repository and give someone trustworthy complete access. Licence the thing under GPL, and then get a team together from the current members of the community. As long as expectations are managed, there are enough people here who dont need payment or gratification, just the knowledge that they are working on something.

Judging by the size of this forum, i think there are enough potential developers to scrape together 10-20 contributors. After that it depends on ensuring that everyone knows what the goals are, and what is expected to happen, so there arent any misunderstandings.

After that it depends on what state the code is in, but really as long as the project of finishing LT is launched in the right way, it certainly has a good chance of success.

There are other tidbits with developing for free, such as having the project go into new and exciting directions, and usually a given contributor lasts for about 3 years. Likewise, working in your free time takes a lot longer, so just bear that in mind, its important to keep the community happy, and then everything else will sort itself out. This is something that will take years, but if kept on track it will certainly go very far indeed. no really, it works!

Its important to note that developing LT is NOT making LT from scratch, the impression i get is that an awful lot of work has already been done, and if we consider that for a moment, and say that 1 Josh is equal to 20 mere mortals, then getting 20 community members to work on this for another 6 years will get LT to an equal amount of progress as has happened so far.

Another thing, is that when things are bad, people step up. A bit like me. I know about Lua coding and ive worked on unpaid projects for a number of years, I also do art and design, and magic like that. Not the most skilled but certainly willing to help if there is something i can help. There are people on this forum. There are more people lurking this forum, and they all just need an opportunity. The most important part is to provide it.

The bottom line is, even if its not successful, i would suggest that the process of working on an open source project is a good one to be involved in, and that has value on its own.

Decisions for josh
This is the part where i overstep any reasonable bounds and make requests of you, josh, after you have already spent all the motivation you have on this project. sorry.
Spoiler:      SHOW
1.Please do make a decision on what you would like to see happen with LT in the ideal world. and then let us know, if you would like to see the game finished by someone else, or by this community, or would prefer to see it kept in its current state, or something else entirely, please do let everyone know of your intentions. Writing just a few lines will greatly shape the future of this place. Depending on your answer, you can stop reading here :)
2.If you would like to see this game developed by the LT community: When releasing the source, please do release it in a github repo, and give a clear line of succession, rather than just throwing it into the air. Pick someone you trust and hand over the rights to them. That way, they would be able to organize this community for years to come, and your project will live on.
3.You may or may not want to have anything to do with LT anymore, but please do let the community know, even if its "i will now move house and never code again" it will really define the direction that this community will go in, and certainty is certainly(!) better than uncertainty. I have a little hope that you may pop in once a year or so but its really not up to anyone but you.
4.Judging by the nature of the announcement and previous actions, the decision to stop looks like a decision that was not taken lightly. I would commend you for that, and urge that you continue making decisions with time to think about them, usually at least 3 days is good for important things. I dont want to sound like someones mother but even after the announcement of this projects end, there is still a lot of things that could change, possibly even a new beginning.
5.Lastly i can imagine that your motivation is gone, but thanks for the stuff that you did so far. As someone who does modding, to me even what you did show was pretty spectacular, its a one of a kind engine, and thats something to be proud of for sure, especially given the absolutely tiny budget for something of this scale.

TLDR: Nothing is certain; Any project can be salvaged; Any effort can be made, even if it does transcend any one person.
Post

Re: Don't release the code yet

#86
vector67 wrote:
Fri Oct 12, 2018 3:51 pm
I don't quite understand why the economy has to be stable? I mean what does that mean? Is It stable in the sense that it will never crash, because that's not how real economies work. If it's stable in the sense that it will eventually return to some equilibrium after a Crash I agree, but I do think that the idea of market bubbles and crashes should not be feared in a sim, if done right it should actually be a sign that you're hitting a viable solution.
Its very easy in a poorly balanced game economy to have either noone producing anything (production disrupted), or going into hyperdrive (way to fast expansion), or have way too high prices, as there is too much money floating around.
Either situations are not much fun for the player. That why most games rely on artificial faucets (getting money and items into the game) and drains (removing money and items) to keep the economy in a balanced state (similar prices and goods availability over time).

The ideal game-economy is somewhat stable, but with some visible cycles (and not completely static) -> While not falling back on the game using artificial faucets and drains.

In this case, the player can actually notice his impact on the economy by pursuing planned actions. For example by reviving an area by providing enough resources. Or disrupt the economy by killing traders.
Or dominate and expand the economy with the players own production-chain.
Thats the ultimate reason why the economical side is fun in the long run. Not just as a provider of ships and weapons.
If the economy collapses by itself, without the players inventions, its taking away the players influence.
If the game has artificial faucets and drains, it also takes away power from the player to influence those fixed mechanics.

It should not be too easy to dominate the economy. But also not take hundrets of gameplay hours to have an impact (unless its an MMO like EvE where people plan and plot over years)
Post

Re: Don't release the code yet

#87
pwaffle2, thank you for taking the time to add your comments. I want to say some things about how Josh inspired people, but maybe it's still a little too soon for that.

On a more practical note, I can endorse nearly every single one of your suggestions. (Other than GPL. :lol: )

In particular, the single biggest question in an open project is how it's managed -- both the assets and the people. Basically, if there's no one with power running the thing, who gets firm commitments from participants and actively pushes toward delivery on those commitments, the odds of an actual product ever being released drop to near zero. That's maybe not a happy thought for some, but it's the only conclusion an honest person who's seen a lot of open projects can hold.

And specifically:

pwaffle2 wrote:
Sat Oct 13, 2018 4:06 am
The next thing is that whether this is successful or not depends on the management of the project the most in my view. How the source is released, under what conditions is very important here. To ensure best success i would create a github repository and give someone trustworthy complete access.

...

usually a given contributor lasts for about 3 years. Likewise, working in your free time takes a lot longer, so just bear that in mind

...

1.Please do make a decision on what you would like to see happen with LT in the ideal world. and then let us know, if you would like to see the game finished by someone else, or by this community, or would prefer to see it kept in its current state, or something else entirely, please do let everyone know of your intentions. Writing just a few lines will greatly shape the future of this place. Depending on your answer, you can stop reading here :)
2.If you would like to see this game developed by the LT community: When releasing the source, please do release it in a github repo, and give a clear line of succession, rather than just throwing it into the air. Pick someone you trust and hand over the rights to them. That way, they would be able to organize this community for years to come, and your project will live on.
3.You may or may not want to have anything to do with LT anymore, but please do let the community know, even if its "i will now move house and never code again" it will really define the direction that this community will go in, and certainty is certainly(!) better than uncertainty. I have a little hope that you may pop in once a year or so but its really not up to anyone but you.

Every bit of that is very well said. In particular, I can't emphasize enough the importance of there being one person with control over and interest in the entire project. Without that, without someone who actively drives progress, progress doesn't happen. This is true in many aspects of human life, and it's true in software development generally and open game development specifically.

So much depends on Josh. If for whatever reason he makes the grand, passionate gesture of simply flinging the existing LT code into the winds, then LT will IMO never happen. If instead he wants some version of LT to exist, and hands over unitary control of the code to one person, then whether a version of LT happens depends on how far Josh got on making the LuaJIT engine work for a restricted subset of the promised gameplay features of Limit Theory. We need to find out whether his terminating the project was due to him personally running out of steam (for which no decent person can blame him), or if there were insuperable technical difficulties that apply even to the most basic subset of desired LT features.

If the former, a new team might succeed. If the latter, then some individual person still ought to be in control of how the assets -- which seem amazing -- are parceled out to the public.

And I say all this because I believe delivering some version of Limit Theory for people to enjoy would be the most wonderful testament to Josh, and to his positive, energetic, smart, creative, and friendly outlook on life in general.
Post

Re: Don't release the code yet

#88
I have to be honest here, I don't believe handing control over the codebase to any one individual is a good idea; given this is the only concrete output of the six years' effort KS backers paid for, giving it to one person seems wrong to me, it should have collective ownership simply due to the collective backing it received.

Further, I think that in similar way to City of Titans, any interested group could get together to produce SLT (Slightly Limited Theory) at any point, and still be organised the sensible way Flat and pwaffle2 have mentioned - one product/project manager, and a series of devs.
It is even possible that multiple groups with different views on the final result could take up arms and start an alternative project; there's room here for all, I believe.

The only reason I can think of for having one person in charge of the codebase, is to deny access to it for some reason. But I can't think of a reason to do that, if someone wants to look at it, why not? If someone wants to make an alternative game, why not?
Most of all, if someone was a KS backer, and wants to learn from the codebase, who is going to say that they cannot, that they do not 'deserve' it, or that we think it's a bad idea?

No, releasing the codebase, ideally as a git repo because the history will help enormously, to all and sundry, seems to me to be the only viable approach.
If someone wants to start a project, then clearly, they own that project, but it must not be used to bar anyone else's access.
--
Mind The Gap
Post

Re: Don't release the code yet

#89
I want the code if it is released to anyone, personally I see it as the best refund Josh can do. I don't care if it's released to the world at large as well. If it's released to one individual, I'd better be getting a 2nd copy.

That's my personal opinion on this entire discussion.
This is my signature, there are many like it but this one is mine.
Post

Re: Don't release the code yet

#90
How has no one brought up Blender? Blender wasn't even a community to start... it basically got open-sourced off of a massive crowdfunding effort, and has only gotten better since! Open source works, this has been empirically proven! I'm a bit sad about how poorly this community seems to think of it... especially considering how many people here program and seem generally knowledgeable and thoughtful about these sorts of things. I guess there's not a whole lot of precedent in the gaming industry, but to my eyes, this is one of the best opportunities I've seen to date to prove that it is viable.

Also, and this one started FOSS, but Linux is open source... And that has incredibly high market share in like... every computing market besides the desktop one. Obviously GIMP/Krita and stuff but those have been brought up so I won't mention them.

Sorry about the fact that this has been the only thing I've commented about recently; I'm a long time lurker who programs who is... err... passionate about FOSS.
"The chances of getting picked up by another ship within those thirty seconds are approximately 3,720 to 1."
"Never tell me the odds!"
Intentional misquotes are very exciting.

Online Now

Users browsing this forum: No registered users and 5 guests

cron