Return to “General”

Post

Re: Where to be notified about code release, and any fan-based projects to bring LT-like games to life using code?

#33
Lemar wrote:
Tue Nov 27, 2018 5:45 am
CSE wrote:
Mon Nov 26, 2018 12:25 am
I do worry if the silence could be a new dark time... i sure hope he is doing well.
The lights went out, the music stopped to play. New dark time ahead oh my godness :ghost:

Seriously, how much darker can it become lol
Well, you can have a dead project and a healthy project manager planning his future after having learnt from his experience.

Or a dead project, and a sick project manager depressed by project failure. This is darker. And what I fear with a completely silent Josh.
Image
Post

Re: Where to be notified about code release, and any fan-based projects to bring LT-like games to life using code?

#36
BFett wrote:
Mon Nov 05, 2018 11:36 am
In essence technical feature creep was what killed LT. Modding was not supposed to be in LT, and in the end, this is what I think prevented the game from releasing. I'm not a programmer though so I could be wrong there. I just think that if Josh had worked on getting gameplay worked out, just as he had while he was doing work on the prototype, LT would have released at least a year ago.
I'd like to also put this in perspective.

Keep in mind that the idea of 'modding' being in LT came about due to the nature of the project in general. While there was definitely scope creep, this was perhaps one of the few things that really wasn't scope creep and more of a backbone necessity.

Quick background;
When you write a game, program, or really any large application there comes a point when you need to separate out low-level code from high-level code. Low-level code is the fundamental mechanics of how the game works with the game loop, input handling, graphics code, asset code, etc. High-level code is what runs on top of that from game logic to object interaction handling.

In the very early days of LT on the forums, the X series was mentioned a TON. X was using scripts for a majority of the game in general. In the case of X, the high level code was the scripts. This is actually pretty standard for many games. It's why X happened to be very moddable; the choice of high-level code is what made that possible. Even World of Warcraft has this to some degree. In other words, this ends up being WHY there are mods in the first place.

But back to early LT and X. It was (relatively) easy to overload X with too many scripts. The conversations are buried somewhere in the forums (and if you ask nicely, I'm sure our walking LT-O-Pedia could reference it :) ), but it came down to the fact Josh didn't want to use a high level scripting language as it was 'too slow'. That 'too slow' came about because of the very low (in today's standards) system requirements for the game.

However, when you use a high-level scripting while you lose speed, you gain one very crucial piece; development speed. It's what allows you to plug things in on the fly and even outsource if you want to speed up asset creation and items in the game. Before the dark days, Josh ended up on this path with his LTScript and more than likely sent him down his dark spiral of the silent days because he was reinventing the wheel.

It was also seen that once he came back and was back on track, he kept the idea of high-level scripting which demonstrated itself VERY nicely with all the things that were seen from his small team. This was the right path, even if it meant sacrificing some game speed.

So while modding may have been scope creep, it was necessary scope creep -- just the necessity of it was discovered a bit late.
Image
Early Spring - 1055: Well, I made it to Boatmurdered, and my initial impressions can be set forth in three words: What. The. F*ck.
Post

Re: Where to be notified about code release, and any fan-based projects to bring LT-like games to life using code?

#37
Zanteogo wrote:
Sun Nov 11, 2018 9:18 am

LT failed because......

I agree with all of this, Zanteogo. People kissing his ass for YEARS, reassuring him this behavior was exceptable is exactly why it failed the way it did. He should have never taken on a project of this magnitude if mental stability was EVER an issue in the past, and he damn for sure should have thrown in the towel and offer partial refunds when it first start affecting him the way it did. Instead he went silent.... For months? No... Nowhere would that be exceptable! But, his ass-kissing crew, hoping to be buddies with him and get sneak peeks or demo's to play/test, continued to reassure him things were fine. This wasn't a gift to the world he was working on. It was a paid for promise. And forget the basecode release in its current state.... Why hasn't the kickstarter Alpha release been made public to us all? Why are those few lucky people still the only ones allowed to have ANY access to this? However limited it was, it was still in a "playable state". One would think after years of delays and broken deadlines and promises, he could at least happily allow that as a download.
Post

Re: Where to be notified about code release, and any fan-based projects to bring LT-like games to life using code?

#38
Please don't bring a hostile attitude to the table.

ExpressZero wrote:
Thu Nov 29, 2018 12:14 pm
I agree with all of this, Zanteogo. People kissing his ass for YEARS, reassuring him this behavior was exceptable is exactly why it failed the way it did. [...] But, his ass-kissing crew, hoping to be buddies with him and get sneak peeks or demo's to play/test, continued to reassure him things were fine.
Normally I'm content to let people have their own opinions, but the community here is not the reason that Josh failed to make LT. First, it was a project that even AAA studios don't dare to attempt, and the only one that did (Freelancer) almost failed, even with a large team with plenty of experience. It's sad that LT failed, but really not that much of a surprise. It would have gone this way regardless of whether or not the community said helpful or unhelpful things. The people here do not control him. You can say most anything you want about Josh, but insulting other people here is not something we do on this forum.

He should have never taken on a project of this magnitude if mental stability was EVER an issue in the past
It was not.

and he damn for sure should have thrown in the towel and offer partial refunds when it first start affecting him the way it did. Instead he went silent.... For months? No... Nowhere would that be exceptable!
May I remind you that your original post on this forum expressed the opposite of this sentiment. Hindsight is 20/20.
Spoiler:      SHOW
ExpressZero wrote:
Fri Oct 23, 2015 7:32 pm
I really want to play this game. I mean, I REALLY want to. I HATE that I just recently came across it's existence, only to find out that i'm out I'm essentially banned because I wasn't where I needed to be years ago when there was a chance to financially contribute.
I get why he doesn't want to open another round of crowd-sourcing. Potentially opening the door to a whole new group of people demanding to know about updates, delays, the hassle of email after email of "when?? when?? when??" adding pressure, and taking the joy of "creating" out of the whole process, turning it into a job. I get it, I really do.
But there is a whole group of people that just want to be a part of what exists so far. People willing to pay to get their hands on it, enjoy what there is to enjoy however unfinished it may be, sign up for this forum, contribute, and only THEN truly be a part of this community.
I'm left currently just being one of those "when?? when?? when??" people, potentially adding to the stress or "finishing" it, and potentially delaying its arrival.
Anyway, I ask that you please rethink not allowing anyone new to be a full member of this community by denying them access to whats available up until this point. I don't need a crowd-funding page. I'd be fine sending over $20 to Paypal, and another $20 when beta rolls out, or whatever.... All with the understanding that this money is buying me what there is up to date, and has no promise or guarantee of future product releases.
I'm sure now there will be a flood of forum member elitist's all saying blah blah blah, or linking to past posts addressing this, or jr admins letting me know what they think is full in the mind of the person that's running the show.... I'd just like my words to reach the creator at some point, and maybe make have him give it a second thought.

This wasn't a gift to the world he was working on. It was a paid for promise. And forget the basecode release in its current state.... Why hasn't the kickstarter Alpha release been made public to us all? Why are those few lucky people still the only ones allowed to have ANY access to this? However limited it was, it was still in a "playable state". One would think after years of delays and broken deadlines and promises, he could at least happily allow that as a download.
I'm going to take the opportunity to point out (as someone always does) that Kickstarter is not a vending site such as Steam or GOG. It is a system where people donate money to a cause: helping someone reach a goal. If the goal is reached, the people get bonuses as a thank-you. If you use Kickstarter purely for rewards, you are not using it as was intended, and if you're upset by the results, that's on you. One should not sue a company that manufactures kitchen knives when they cut their arm open trying to juggle them.
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: Where to be notified about code release, and any fan-based projects to bring LT-like games to life using code?

#40
ExpressZero wrote:
Thu Nov 29, 2018 12:14 pm
...yadda yadda yadda.....ass-kissing community...Josh is mentally unstable...etc...
Re: People kissing his ass for years
If you're referring to the forums, then yeah. Why? Because the forums here were not created to be about how bad he is or LT is. Much like the walmart.com site doesn't tolerate people who just go on to complain about walmart. That's why the people created walmartsucks.com. it's also why when I was a mod, I redirected people to Kickstarter or to come up with their own community on the same idea -- not just because it isn't what this place is for, but also because those venues are more 'official' than a mostly community-run forums here. Although, I won't deny you your frustration. I had heard much the same via backchannels from MANY more.

Re: Mental Stability
The project is what broke Josh. I'm confident enough to say that there were no issues until this project came about. The exact reasons why? We'll probably never know. I'd only caveat your statement with a small added bit of "paid". If it's a side project, no problems. Some people even felt the same that he should've thrown in the towel sooner, but like Tal said, hindsight is 20/20.

Re: LT Prototype
When the prototype was released, there was an expectation that it would not be shared. Not just at his behest, but I believe there was also some for of contractual agreement. Much like how Tal is under an NDA for a certain amount of time after his 'termination' before he can speak to many more things, it was the same way with the prototype. Just because the project is dead, does not mean the agreement is anulled. So why hasn't it been made public to all? Part in respect to Josh, part in possible contractual agreements. Now, if the question is whether or not anyone is willing to still abide by it now that the project is dead? No idea. Personally if you're that much wanting to see it, I can send you the copy I was given because honestly I just don't care to fight any of these battles now that I'm no longer a mod.

Re: Re: Ass-kissing
I lost count of how many people got pissed off at me for either telling them off, or telling them to take their negativity elsewhere. People even attacked me directly thinking I was one of the elitists that believe Josh could do no wrong.

What many didn't realize is that I was a mod to Enforce that ideal. Feel free to hit up people in the IRC channel. There were many many times where I had to hold myself back because of the mod hat, only because there was no way to definitely separate "Kickstarter backer DWMagus" from "Forum Mod DWMagus". I purposely watched my words, I purposely made sure that anything I said represented LT instead of myself personally just to make sure things weren't taken a wrong way. This is also why I stepped down from being a mod when I did (that and personal reasons beyond LT).

Also, to be fair, there are very few forums run by developers that will allow true 'neutral' speech. If you hop onto Blizzards or EA's forums and talk about how bad their studios are, chances are you'd get banned or at the very least, asked to go elsewhere. This place is no different. Just because there is no 'officially sanctioned gripe forums' provided by the developer, doesn't mean you're not free to create your own.

Seriously. This last bit still pisses me off and it's very evident that everyone here who had gripes felt that if they can't express how upset they are on the official forums they don't seem competent enough to find a different place to do it and that this was an infringement on their freedom of speech.

I think I used this image MULTIPLE times because of it and it still holds true (I swear it's the last time I'll post it).
Image

(jeebus crisps, I can't seem to not write a book anymore :| )
Image
Early Spring - 1055: Well, I made it to Boatmurdered, and my initial impressions can be set forth in three words: What. The. F*ck.
Post

Re: Where to be notified about code release, and any fan-based projects to bring LT-like games to life using code?

#41
About the modding:
There are a hundred ways to separate the basic engine functionality from the game-content.
Using a scripting language is not set in stone.
Especially trying to have this scripting language run pretty low level functionality, and then running into trouble with performance.

The game could have also just defined its content using a custom data-description language, xmls, json or whatever.
Where the content is described by data, but not necessarily by describing processing logic (wich can be kept in C that uses the data).
Developing the logic is not that much less work just because its done in a scripting language. The logic still need to be thought out.

Also the performance optimization discussion was kind of odd: Trying to have 10.000 ships run at the same time? Whats the point in that?
There are no space-games that have that many ai ships run at the same time .. because its adds nothing to the gameplay.
The player can handle maybe a few dozen ships in active combat. The rest would be just backdrop, wich could be a simple animation for all it matters.
Putting so much effort into letting 10.000 ships run at the same time was an optimization exercise with no real feedback to determine the performance in the actual game later.

What would have potentially tanked performance is much more likely the simulation of the world economy and ai decisions, not an unrealistic amount of dynamic 3D objects with a simple combat ai.
And designing that part (economy and ai decisions) was being pushed back for the sake of first optimizing the LUA-C implementation. But even if the scripting engine runs perfect, developing the gameplay logic of the world is not an afterthought that can be quickly raced to the finish line.
- Its likely the biggest hurdle to take -
Not because of performance problems, but because its effect on gameplay was never tested in a running implementation.
Gameplay and balancing are not straight forward procedures that can be optimized like spacial partitioning, physics or pushing drawcalls.
Its takes lots of prototypes, gameplay test sessions, tuning and reiterations to make it a system that is fun and stable in procedural gameworld (without resorting to cheap hacks).
The high level economy and ai-decisions are also something where the prototypes could have been developed in parallel and independent of the game engine, even in another language. Once it looks promising, it can be ported over to the gameengine -> already indicating what parts of this simulation really need to run performant (pushing it to raw C), and what parts dont.
In any case, large parts of the game (economy, ai logic, procedural algorithms for 3D and art content) could have been developed before the actual engine was even hammered out.

If in the end the engine then shows performance problems, the game could be scaled down in scope. But it would have been a running game, with the potential to scale it up later when the engine matures.

Overall there was maybe a well crafted engine in the end, but not much of a game yet.
Post

Re: Where to be notified about code release, and any fan-based projects to bring LT-like games to life using code?

#42
Damocles wrote:
Thu Nov 29, 2018 5:14 pm
Developing the logic is not that much less work just because its done in a scripting language.
One minor point about this: the reason a script language really is much less work for developing the game logic, is because high-level languages provide more facility than low-level ones.

Example:

I want to give a spaceship component a name, like "BFG Blaster".
High-level language can assign a string property to an object.
Low-level language has to reserve some memory somewhere, figure out where the data for the name is currently stored, copy it over to the reserved memory (don't forget a terminating null!), then attach that memory to the memory for the object.

Things that get taken for granted, like the concept of a string data type, might not exist in a low-level language. While C is a high-level language, and could certainly be used if you wanted to, the compiled nature makes it a pain for modding as you're basically recompiling/linking the whole game if you make a tiny change.

So, yeah, developing the logic IS much less work when you can make a change in a moment and try it out, and when you have more facility - literally, at a higher level - to make that change with.
--
Mind The Gap
Post

Re: Where to be notified about code release, and any fan-based projects to bring LT-like games to life using code?

#43
Thanks to Damocles for one of the meatier posts here in a while. (And once again to Magus for his typical clarity.)

Some specific thoughts:

Damocles wrote:
Thu Nov 29, 2018 5:14 pm
The game could have also just defined its content using a custom data-description language, xmls, json or whatever.
Where the content is described by data, but not necessarily by describing processing logic (wich can be kept in C that uses the data).
Developing the logic is not that much less work just because its done in a scripting language. The logic still need to be thought out.

I agree with Ringu's response, but one other note is that if you don't implement a scripting language for most of the gameplay logic, and stick to a compiled language like C, you may require most modders to have to go buy a compiler. That will reduce the available pool of modders, versus the number of people who can use a scripting language that's basically free out of the box.

Of course "more modders" does not magically translate to "better mods." But Sturgeon's Law says it might at least increase the chances of getting more useful mods.

Damocles wrote:
Thu Nov 29, 2018 5:14 pm
Also the performance optimization discussion was kind of odd: Trying to have 10.000 ships run at the same time? Whats the point in that?
There are no space-games that have that many ai ships run at the same time .. because its adds nothing to the gameplay.
The player can handle maybe a few dozen ships in active combat. The rest would be just backdrop, wich could be a simple animation for all it matters.
Putting so much effort into letting 10.000 ships run at the same time was an optimization exercise with no real feedback to determine the performance in the actual game later.

I'm about 99% sure "running 10,000 ships" was never an actual development goal. That was just Josh doing a performance test and discovering the system was capable of handling that many ships at once.

Nothing Josh said that I'm aware of indicated that running that many ships was part of his gameplay planning. It was just a testing artifact.

Damocles wrote:
Thu Nov 29, 2018 5:14 pm
What would have potentially tanked performance is much more likely the simulation of the world economy and ai decisions, not an unrealistic amount of dynamic 3D objects with a simple combat ai.
And designing that part (economy and ai decisions) was being pushed back for the sake of first optimizing the LUA-C implementation. But even if the scripting engine runs perfect, developing the gameplay logic of the world is not an afterthought that can be quickly raced to the finish line.
- Its likely the biggest hurdle to take -
Not because of performance problems, but because its effect on gameplay was never tested in a running implementation.
Gameplay and balancing are not straight forward procedures that can be optimized like spacial partitioning, physics or pushing drawcalls.
Its takes lots of prototypes, gameplay test sessions, tuning and reiterations to make it a system that is fun and stable in procedural gameworld (without resorting to cheap hacks).
The high level economy and ai-decisions are also something where the prototypes could have been developed in parallel and independent of the game engine, even in another language. Once it looks promising, it can be ported over to the gameengine -> already indicating what parts of this simulation really need to run performant (pushing it to raw C), and what parts dont.

This, I agree with. I pretty much said the same thing every time I posted a "Hey, what's the state of progress on the major milestone of having LOD-managed AI, spanning multiple star systems, that is both fun and performant?" comment.

Someone can correct me on this if I'm wrong, but I don't recall ever seeing a Joshpost where he clearly stated that he'd made significant progress on this key part of the game. It's a complex system, and getting it right (fast and fun) is non-trivial for any game, much less one that needs this to be tightly integrated with non-simulated gameplay in a local star system and work in a procedurally expandable universe.

Also, remember that "I've decided the universe should not procedurally expand, and the game should have an ending" (I'm broadly paraphrasing him here) was one of the last big design announcements Josh made. That narrowing of scope is exactly what I would do if I discovered that a multi-star system AI/economic simulation was a bigger task than I originally thought. Making LT more of a conventional game by 1) setting it in a static (but still randomly generated) universe and 2) recognizing when the game is over (set victory/defeat conditions and test for them) and ending play not only simplifies AI/economic processing somewhat, it also allows tech research to work as a gameplay feature that's fun during the entire game.

All that being said, I would not be surprised if making this work -- AI/economic simulation that's fun to interact with directly, that connects enjoyably with player power progression, research, mining, production, and faction operations, that does most of its heavy lifting in the C core engine but implements most of the interactive behaviors in LuaJIT, that simulates and un-simulates properly with increasing and decreasing distance from the player, and that has been tested extensively and proven to be fast and fun for a game universe of any reasonable size and (finite) gameplay length -- was the reef on which Limit Theory ran aground. For one person -- even a Josh -- to get all that implemented and polished could have taken two more years... time that LT just didn't have.

I suppose we'll find out when the code is released whether this theory is correct, or if there was some other fundamental requirement for LT that proved so difficult it scuppered this project.

And then we'll see whether outsiders really are so much brighter than Josh as they think they are. :D
Post

Re: Where to be notified about code release, and any fan-based projects to bring LT-like games to life using code?

#44
Flatfingers wrote:
Fri Nov 30, 2018 11:10 am
I agree with Ringu's response, but one other note is that if you don't implement a scripting language for most of the gameplay logic, and stick to a compiled language like C, you may require most modders to have to go buy a compiler.
Wait, people still have to buy compilers? I thought all the big ones had free ones by now.
Image
Post

Re: Where to be notified about code release, and any fan-based projects to bring LT-like games to life using code?

#45
0111narwhalz wrote:
Fri Nov 30, 2018 6:46 pm
Flatfingers wrote:
Fri Nov 30, 2018 11:10 am
I agree with Ringu's response, but one other note is that if you don't implement a scripting language for most of the gameplay logic, and stick to a compiled language like C, you may require most modders to have to go buy a compiler.
Wait, people still have to buy compilers? I thought all the big ones had free ones by now.
It's been a while since I shopped for a C compiler. If there's a freeware compiler today that's compatible with whatever Josh was using (GCC?), then yep, this concern fades away.

Online Now

Users browsing this forum: No registered users and 18 guests

cron