Return to “General”

Post

Re: Remaking Limit Theory - From the ground up

#47
I'm glad to see this happening. I don't know what I can offer -- some thoughts on organizing the project and game design, possibly some coding -- especially given I'm still trying to finish the 600-page Audible novel narration project I got "volunteered" to do. But If there's something constructive way I can help, I'm in.

That said, I do have a few quick early thoughts:
Shadowrunner214 wrote:
Sun Mar 29, 2020 11:09 am
Perhaps the question we should be asking is what the exact boundaries of the project are going to be before we discuss technology. If we have to choose between functionality and technology, I'd rather go for functionality if at all possible. There's no point selecting tech before we have a solid idea about the scale of the project, only to have to start over with tech selection once the scope is firmly established.

I think this is very wise. Game vision first -- what's the overall experience supposed to be? Then strategy -- what is the handful of core features needed to deliver that experience? Then do a quick first draft of breaking those down into specific features that can actually be coded. (This doesn't have to be super-detailed or perfect; it's just a general project scoping step.) Get some consensus on those features from project participants.

And THEN make a list of possible technologies, and assess them for 1) speed of building the known required features, 2) cost to license, and 3) how well most project participants know it (or can easily pick it up).

Talvieno wrote:
Sun Mar 29, 2020 11:34 am
In short: Yes, I have it. Am I allowed to spread it around? Is it ethical in any sense of the word? Would Josh be pissed? I have really no idea about any of these three questions.

I feel pretty strongly that there are very clear answers to those questions, and they are "no," "no," and "yes." We don't have to like that Josh so far has not released the code as he said he would, but our likes or dislikes are irrelevant. It's his property, and whether to distribute it or not is no one else's call but his regardless of who has been entrusted with a copy.

I hope he will share it someday, just as pretty much everybody else hopes. But as there is no sign of that happening, a clean-room build of a game with some similar features is the only ethical (and, in my non-lawyer opinion, legal) alternative.



Unless anyone objects, I think I'll see if I can scrounge up some time tomorrow (although I'll be doing my actual day job most of the time) to boil down The Big List of Limit Theory Features into a short list of specific high-level features for consideration.
Post

Re: Remaking Limit Theory - From the ground up

#49
Alright, I've browsed the forums and here's what are the most desired features for a finished version of this game.
---------------------------------
Project must have:

Factions
Asteroid Mining
Dynamic LOD scaling
Emergent production based economy
Ship component configuration
Tactical Dog-fighting
Fleet operations
Colonies
Detailed scanner/sensor operations

General concepts which are desired include:
Combat
Asteroid Mining
Exploring a living universe - exploration vessels
Trading
Planning growth for multi-sector civilization (Large C&C battleships, carriers, science ships)
Large industrial ships (Freighters, mining barges, support fleets, mobile bases C&C)
Management of several fleets, planets and factions.

Modding seems important as well, and will likely extend the life of the project.

---------------------------------

For Alpha build 1, I'd like to be able to fly a ship, and shoot at an object.
For Alpha build 2, add the ability to change weapon types and have a second weapon to equip.
Image
Post

Re: Remaking Limit Theory - From the ground up

#51
I'm hardly a lawyer but I agree with Flat, regardless of the technical aspects of the NDA it's best to start from scratch. Like Tal said, it's not like the demo he has contains a lot of useful code anyway, and it's all stuff that can be redone if necessary.

Sadly I'm in Cha0zz's boat as far as helping, but I'm also happy to test/be a cheerleader :3

More drinks from me if you guys pull this off!
Image
Image
Post

Re: Remaking Limit Theory - From the ground up

#52
zircher wrote:
Mon Mar 30, 2020 9:04 am
I take it that explore a living universe means procedural content such as asteroids, planets, stations, ships, and in-system events (bulletin boards/NPCs) for all that?
Not necesarily. Living just means stuff is happening, and e.g. freelancer had plenty of (premade) universe to explore.
(in fact, looking at freelancer in general for what we want is probably not a bad idea)
Warning: do not ask about physics unless you really want to know about physics.
The LT IRC / Alternate link || The REKT Wiki || PUDDING
Image
Post

Re: Remaking Limit Theory - From the ground up

#53
Good list, BFett. Questions for everyone interested in participating:

1. Should this game be open-ended and ever-expanding as in Josh's original design? Or should every game build a fixed universe, so that every game has an ending?

2. How committed are y'all to having multiple star systems, rather than a single very detailed star system? Are the gameplay features potentially enabled by implementing multiple star systems (e.g., RTS-style fleet management, exploration/discovery of new worlds/resources, strategic sim) worth the significant costs in design and programming (e.g., good procedural generation of multiple connected star systems, multi-system non-inflationary economic simulation, LOD decisions and coding, enjoyable fleet management gameplay and UI, enjoyable 4X gameplay/UI)?

I'm not asking this because I oppose multi-system play. I'm asking because I see it as a key design decision with very large consequences in the overall amount of development effort required.

3. Josh expended an enormous amount of time and mental effort (JOSH did) into making modding work efficiently. You'll remember he called it the FPLT -- the Fundamental Problem of Limit Theory. I don't mean to take anything away from the brainpower assembled here, but... is this a fight y'all want to have? Knowing that it's possible is half the battle, but that still leaves a lot of battle. I agree that enabling modding has a lot of value, but it's also potentially a very high-cost programming challenge, and one that must be decided from the very start; you don't want to try to retrofit this.

Of all the possible design questions, I think these have the most impact on just how hard a project like this may be.

4. Some additional game features for this group to consider:

  • Several ship hull sizes (fighter, destroyer, cruiser, battleship) [you may have assumed this as a feature, but there are important implications]
  • Research/tech tree (only makes sense if this is the kind of game with an ending)

(Note: BFett, your list has Asteroid Mining in both the "must have" and "desirable" sections. I like it. :) )
Post

Re: Remaking Limit Theory - From the ground up

#54
Flatfingers wrote:
Mon Mar 30, 2020 9:12 am
Good list, BFett. Questions for everyone interested in participating:
[snip]
Those are definitely some important questions.

My answers follow:

1. A fixed universe has more opportunities for engagement I think. You (and the AI) can't just run away from a situation. That said, I don't know that one system, large as it is, will be sufficient to support the economies of enough factions that large 1k+ ship battles are possible. I don't know if fleet engagements of that size are necessarily desirable, but it would definitely be cool. I think another question that should be asked as a follow-up is if a story should be written into the game to (hopefully) keep a player engaged for longer.

2. mostly derived from 1. Big enough to support the fleet sizes and frequency of conflict that we want, but probably no bigger.

3. I think that if we design things right, we can turn most of the gameplay elements into mods themselves. This would let us battle-test the modding system, and also give us the flexibility to experiment with new or improved features during development without needing to plunge into the depths of the 'engine' code (whatever form that might take). That would allow us to amortize the cost of developing the modding system across the entire game, with additional benefits that may reduce total development time. At the very least, this would allow the game to exist in many additional variations long after much of the development is completed, extending replay-ability.

I dream of a mod that adds alien invaders that pop up at some point trying to conquer the play area.

4. I don't know how you could play a space game where every ship is the same size... it'd be interesting to see...


Beyond those questions, I have a few more that I'd like to clarify:

5. More than just system size and number, what kinds of actor numbers do we want to be able to support? This is AI for everything from strategic to fighter control and even missile/turret targeting. I'm thinking something on the order of a few million to be able to support a functioning economy of sufficient size, whilst supporting more advanced ship features. This is closely tied with the amount of resources that are available within the play area.

6. How are we going to define success for this project? Is it just for the pleasure of playing a completed game or are we looking for more widespread success from outside of the LT community? A game like this is a complicated beast, and is going to require significant time investment from its contributors. Having some idea of what the light at the end of the tunnel is going to smell like may be useful to constrain or guide our decision making.

7. An extension of 6 is licensing. Will it be open source? Partially open source with 'proprietary' components? Completely 'proprietary'? Essentially, what kind of market model do we want? Please do keep in mind that FOSS and freeware are still market models.

8. How sustainable is this effort going to be? How can we ensure or at least stack the deck in favour of the project's completion?
Post

Re: Remaking Limit Theory - From the ground up

#55
Flatfingers wrote:
Mon Mar 30, 2020 12:26 am
Talvieno wrote:
Sun Mar 29, 2020 11:34 am
In short: Yes, I have it. Am I allowed to spread it around? Is it ethical in any sense of the word? Would Josh be pissed? I have really no idea about any of these three questions.
I feel pretty strongly that there are very clear answers to those questions, and they are "no," "no," and "yes." We don't have to like that Josh so far has not released the code as he said he would, but our likes or dislikes are irrelevant. It's his property, and whether to distribute it or not is no one else's call but his regardless of who has been entrusted with a copy.
Interestingly - and for the sake of discussion - I am not sure I agree.

I assume that we can argue that we - as backers - have a right whatever came out of the endeavor, a right to whatever came from the contract we had with Josh. What exact right is the question.... Sure, We would have no rights to use the code in a new context nor to re-distribute it, of course, so it makes the point moot for this discussion: Any such project must start from scratch.

But looking at the code for solutions that have been developed while Josh was « working for us » should be acceptable as he is not present to explain us the solution he came to using our financing. So having the code available for backers (that sign an agreement not to include it in their projects, nor to distribute it to non-backers) that can re-use (in the sense re-implement but by learning from the example) some concept would make sense.

Of course, we need to have Josh’s opinion. But if we did our best to contact him and he choose not to answer, than we may not need his formal agreement. I, for one, think we should stop protecting him from his own childish behavior.
Image
Post

Re: Remaking Limit Theory - From the ground up

#56
I don't know how much of a must-have this is for others ... but what I miss a lot from when I played Freelancer: To me it felt like it had something that I would call wilderness?
I never again played a space game where I could go on a road trip through vast and dense nebula with sharply edged asteroids and lightning strikes, on the search for a hidden planet, asteroid base or worm hole. It felt uncharted and dangerous without feeling empty.
The one that came closest for me was probably X2 and the fact that parts of the map were separated by systems that were dangerous to traverse early on and through which you should never plan any trade lanes.
2. How committed are y'all to having multiple star systems, rather than a single very detailed star system?
Whatever you guys decide on doing. I, as loyal lurker, word-spreader and potential customer would very much appreciate a focus on structure. I think a smaller, maybe hand crafted universe can feel much bigger than many of the sci-fi games out there.
My vote would be on multiple star systems. Otherwise you can't have stuff like neutron stars or different sky boxes and that might feel claustrophobic?
Other than that, I don't think the size of the universe matters too much, as long as you make sure that going from one territory to another actually means something. (I really dislike the circular highway in X4. You can visit almost all the factions without even noticing. So lame. I love Freelancers trade lanes ... but X4 is an example of how not to implement them :/ )

I know, stuff like this has been discussed before. To me it seemed that this kind of "structure" was also important to Josh. I am bringing this up again because to me it was one of the most interesting aspects of LT.
But I haven't really seen much proof of it in the videos and I imagine something like this being difficult when going full procedural and ever-expanding.
Post

Re: Remaking Limit Theory - From the ground up

#57
Flatfingers wrote:
Mon Mar 30, 2020 9:12 am
Good list, BFett. Questions for everyone interested in participating:

1. Should this game be open-ended and ever-expanding as in Josh's original design? Or should every game build a fixed universe, so that every game has an ending?

2. How committed are y'all to having multiple star systems, rather than a single very detailed star system? Are the gameplay features potentially enabled by implementing multiple star systems (e.g., RTS-style fleet management, exploration/discovery of new worlds/resources, strategic sim) worth the significant costs in design and programming (e.g., good procedural generation of multiple connected star systems, multi-system non-inflationary economic simulation, LOD decisions and coding, enjoyable fleet management gameplay and UI, enjoyable 4X gameplay/UI)?

I'm not asking this because I oppose multi-system play. I'm asking because I see it as a key design decision with very large consequences in the overall amount of development effort required.

3. Josh expended an enormous amount of time and mental effort (JOSH did) into making modding work efficiently. You'll remember he called it the FPLT -- the Fundamental Problem of Limit Theory. I don't mean to take anything away from the brainpower assembled here, but... is this a fight y'all want to have? Knowing that it's possible is half the battle, but that still leaves a lot of battle. I agree that enabling modding has a lot of value, but it's also potentially a very high-cost programming challenge, and one that must be decided from the very start; you don't want to try to retrofit this.

Of all the possible design questions, I think these have the most impact on just how hard a project like this may be.

4. Some additional game features for this group to consider:

  • Several ship hull sizes (fighter, destroyer, cruiser, battleship) [you may have assumed this as a feature, but there are important implications]
  • Research/tech tree (only makes sense if this is the kind of game with an ending)

(Note: BFett, your list has Asteroid Mining in both the "must have" and "desirable" sections. I like it. :) )
Some thoughts on the direction I personally would like to see things go.

1. For this project a universe fixed in size is likely a good way to go. The factions within could still have tensions and rebellions could begin for one reason or another extending the play time.

2. I would be content with a single star system. Seven or eight planets, with several moons, and each having their own factions associated with them may be enough for a project of this size. Each could be connected by jump gates and be unique in how it operates. I believe that a single system could be just as engaging as the procedural systems LT had to offer.

4. Several ship hull sizes are important to game play as they will help deepen space combat strategies. Three to four distinct classes might be good for starters, with the ability to outfit each according to the desired play style.

No comment on number 3 at this time. I'm glad you liked where I placed mining on the lists :P
Image
Post

Re: Remaking Limit Theory - From the ground up

#58
Flatfingers wrote:
Mon Mar 30, 2020 9:12 am
1. Should this game be open-ended and ever-expanding as in Josh's original design? Or should every game build a fixed universe, so that every game has an ending?
I would put the game in a large "star cluster" with fixed boundaries. "No Endgame" sounds neat, but in practice you run into the same problems Josh faced in many of his systems, such as research and progression.

Handcrafted? Don't know. I could lean either way, I think. Star system procgen is incredibly simple compared to, for instance, ship procgen, but handcrafted universes often feel more "alive" - at least for the first playthrough. That said, I'm personally a fan of procgen'd maps and use them whenever they're offered (e.g. 7 days to die).
Flatfingers wrote:
Mon Mar 30, 2020 9:12 am
2. How committed are y'all to having multiple star systems, rather than a single very detailed star system? Are the gameplay features potentially enabled by implementing multiple star systems (e.g., RTS-style fleet management, exploration/discovery of new worlds/resources, strategic sim) worth the significant costs in design and programming (e.g., good procedural generation of multiple connected star systems, multi-system non-inflationary economic simulation, LOD decisions and coding, enjoyable fleet management gameplay and UI, enjoyable 4X gameplay/UI)?
I'm personally highly committed to having more than one star system. The economic simulation isn't that much more complex with a set of star systems rather than one detailed system. Also I don't think Limit Theory (or Freelancer) was ever meant to be 4X in any way. Let's bring it to a grinding halt at RTS gameplay and call it a day.

Flatfingers wrote:
Mon Mar 30, 2020 9:12 am
3. Josh expended an enormous amount of time and mental effort (JOSH did) into making modding work efficiently. You'll remember he called it the FPLT -- the Fundamental Problem of Limit Theory. I don't mean to take anything away from the brainpower assembled here, but... is this a fight y'all want to have? Knowing that it's possible is half the battle, but that still leaves a lot of battle. I agree that enabling modding has a lot of value, but it's also potentially a very high-cost programming challenge, and one that must be decided from the very start; you don't want to try to retrofit this.
Modding is more essential now than it was with Josh's LT, in my opinion. This is a crowdsourced project. The assembled "crowd" needs to be able to have tools to create the game efficiently. That means, more or less, that the guts of the machine be open to easy manipulation.

If we have to make "budget cuts" in other areas to make this happen, such as "fewer ships in one star system" or "less economic simulation", I would say it's 100% worth it. Why? Simple enough.

I want to see the game through to completion. Let's not fall into the same trap that Trillek (the fan project to complete Notch's 0x10c) did - it was so technical that only a tiny core of people could work on it. Eventually the whole programming group just split apart and it more or less died. We need to let as many people as possible work on the thing (within reason of course). As far as I can tell, modding capabilities are non-negotiable here.


RE: Ship hull sizes: random thoughts:
Fighter scale: 1
Cruiser scale: 10
Battleship scale: 100
Dreadnought scale: 1000 (for clarity purposes, I would liken the building of a ship of this scale to building a Wonder in Age of Empires. It's not a thing you do because you want a tactical or strategic advantage. It's a thing you do because you want to show off.)

RE: Research/Tech: YES. THANK YOU, FLATFINGERS. That's something I mentioned to Josh before (only makes sense if the game has an ending) but I think it bounced off. :lol:

Shadowrunner214 wrote:
Mon Mar 30, 2020 10:43 am
5. More than just system size and number, what kinds of actor numbers do we want to be able to support? This is AI for everything from strategic to fighter control and even missile/turret targeting. I'm thinking something on the order of a few million to be able to support a functioning economy of sufficient size, whilst supporting more advanced ship features. This is closely tied with the amount of resources that are available within the play area.
Forgive me, but - holy crap - that's a lot of actors. Missile targeting is incredibly lightweight (at least in my limited experience) and I wouldn't personally count it. In regards to AI entities (ships) perhaps a few hundred at the very most. We might have to do something Mount & Blade style and warp in "reinforcements" regularly for larger battles. It depends largely on whatever we decide to use and how smart we actually want the AI. If Player = AI, the AI will take up a lot of resources.

Shadowrunner214 wrote:
Mon Mar 30, 2020 10:43 am
6. How are we going to define success for this project? Is it just for the pleasure of playing a completed game or are we looking for more widespread success from outside of the LT community? A game like this is a complicated beast, and is going to require significant time investment from its contributors. Having some idea of what the light at the end of the tunnel is going to smell like may be useful to constrain or guide our decision making.
I personally define success in a venture like this as "when we're happy with it". Yes, making money with it sounds cool and all, but I think that would create so many problems it's just not worth it. My personal recommendation is therefore open source - unless someone manages to sway my mind in another direction, of course.
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: Remaking Limit Theory - From the ground up

#59
Talvieno wrote:
Mon Mar 30, 2020 12:17 pm
n regards to AI entities (ships) perhaps a few hundred at the very most.
Could you have a functioning economy with that? I suppose it depends on the granularity of the simulation, but if you have raw materials that you have to refine, then construct components, then build the final product, you're going to need a supply chain, with actors at each level. A couple million is probably too much, honestly, but I don't know that a few hundred will be enough.
Talvieno wrote:
Mon Mar 30, 2020 12:17 pm
I personally define success in a venture like this as "when we're happy with it". Yes, making money with it sounds cool and all, but I think that would create so many problems it's just not worth it. My personal recommendation is therefore open source - unless someone manages to sway my mind in another direction, of course.
I don't disagree, but the success rate of small team projects like this is vanishingly small. Money is not strictly necessary and can absolutely make things very complicated. I just mean that a place to say "Ok, we've achieved what we set out to do" is extremely useful for keeping a group together and focused. I'd hate to see this project make progress, then have people leave because they have different definitions of 'done'.

The idea of marketability (even if it's FOSS) can also be useful as a measuring stick of quality. So many FOSS projects intended for the end user are usable but lack polish. I think that more than anything else, games absolutely require polish.
Post

Re: Remaking Limit Theory - From the ground up

#60
Flatfingers wrote:
Mon Mar 30, 2020 9:12 am
1. Should this game be open-ended and ever-expanding as in Josh's original design? Or should every game build a fixed universe, so that every game has an ending?
For sake of codeability, I'd say we have a finite universe. I don't know how Josh wanted to do the infinite stuff and I'm not sure it's doable for anyone not Josh (and even for Josh, tbh)
Flatfingers wrote:
Mon Mar 30, 2020 9:12 am
2. How committed are y'all to having multiple star systems, rather than a single very detailed star system? Are the gameplay features potentially enabled by implementing multiple star systems (e.g., RTS-style fleet management, exploration/discovery of new worlds/resources, strategic sim) worth the significant costs in design and programming (e.g., good procedural generation of multiple connected star systems, multi-system non-inflationary economic simulation, LOD decisions and coding, enjoyable fleet management gameplay and UI, enjoyable 4X gameplay/UI)?
In the end, I think we want multiple star systems, freelancer style. However, like anything in this development, I would make a game plan that is modular in nature, e.g:
0.1: one system, flying, some enemies that shoot at you and you at them
0.2: add inventory/weapon/shop
0.3: more stuff in the system
0.4 mining and selling of mining (static market, or one fluctuating in a preprogrammed way)
0.5: AI behaviour and some premade factions (miners, pirates, trade caravans)
0.6: wormholes and multiple systems (procgen or handmade, to be seen)
1.0: missions (automatically produced)
At this point we'd be at about freelancer feature parity (except for the campaign), and we can see in how far it's viable to add other things:
-dynamic markets
-fleet management
-research
-dynamic missions based on AI needs
-...

And in many cases I'd start with 'spoofed' things that we can later replace with more complex background, e.g. at first just spawning cargo ships with a premade cargo for the player or AI to steal etc
Flatfingers wrote:
Mon Mar 30, 2020 9:12 am
3. Josh expended an enormous amount of time and mental effort (JOSH did) into making modding work efficiently. You'll remember he called it the FPLT -- the Fundamental Problem of Limit Theory. I don't mean to take anything away from the brainpower assembled here, but... is this a fight y'all want to have? Knowing that it's possible is half the battle, but that still leaves a lot of battle. I agree that enabling modding has a lot of value, but it's also potentially a very high-cost programming challenge, and one that must be decided from the very start; you don't want to try to retrofit this.

Of all the possible design questions, I think these have the most impact on just how hard a project like this may be.
Well, if possibly I think it's a good idea for reasons tal said. I think the trick is mostly in picking the framework - a high level language for scripting on top of a compiled one for heavy lifting as Josh had would be ideal... (I do wish we had the LT engine to use)
Warning: do not ask about physics unless you really want to know about physics.
The LT IRC / Alternate link || The REKT Wiki || PUDDING
Image

Online Now

Users browsing this forum: No registered users and 19 guests

cron