Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Lap]

Lack of direction

True enough. Since most of my time was spent doing a lot of basic stuff: gathering people, creating forums, familiarizing myself with the engine, etc. I was content to just let people talk and see where that led us.

Unfortunately, more focus and direction means that some executive decisions are going to have to be made. There are a lot of topics that have some great ideas, but no clear consensus and seemingly no more actual discussion. To move things along an option has to be picked even if that means cutting cool features or even hurting some feelings. Whether an actual idea was bad, infeasible to code, or simply lost to a slightly different idea, it's inevitable that people will get offended and/or disappointed when their idea isn't chosen. This is the part I hate the most.

I've been free to bounce around between different parts of the code, experimenting, and implementing stuff on whim, but that has to stop. I think the "code what you want" directive for new coders clearly didn't work so well. I thought it would attract more help since people wouldn't feel like they were just being assigned boring bitch work, and maybe it did, but without direction most of the people who contacted me never followed through with anything. The one thing that did work well was assigning the inventory system out to a single coder. That worked very well and I'll definitely keep an eye out for more systems like that that can be easily assigned.

This has convinced me that yes, the next step is a focused design doc/roadmap. I'll throw out a rough draft soon so we can look it over and see what needs to be reprioritized, added, or removed.

[nenjin]

Cool. I've always proposed mechanics waiting for you to come in and go "can't make it work" "can make this work" "like it" "hate it." Executive decision in my mind has always rested with you about exactly what kind of mechanics or philosophy we go with.

It was nice, we gave people a several month grace period to get their opinions out, but the time to get firm and get some real code started is nigh.

[Lap]

I try to immediately axe stuff I know can't be coded if it's discussed in more than passing. I have fairly eclectic gaming tastes so I don't even quickly gravitate to one type of playstyle either.

The hard part is that there isn't a right answer to some of these debates. A lot are just matters of preference. They're relatively similar in coding difficulty and have some nice pros/cons on each side. I hold off as long as I possibly can with these hoping that someone brings up something that pushes one idea towards being the winner.

On that note, one of the most well debated topics (skill advancement) looks like might finally be solved. An idea was just posted about skill advancement should satisfy everyone. That one is looking like a winner.

[Meta]

Just a small post to let you know that there's a friend of the programmer of Rogue Survivor who started to program a rogue-like with a Grand Theft Auto theme: http://grandrogueauto.blogspot.com/

[Lap]

I really like the editors that guy made for his game. I might end up making something like that for the tiles and items further down the line. Luckily for us, there's really no competition here as the game seems a lot less ambitious and "serious" than this project. Not in terms of quality, but in terms of realism and depth.

I threw up a basic guideline/design doc for the next stage.

https://docs.google.com/document/pub?id=1BCgE3cb3dXwe_dAe3mpHUY9cu0G9cZRyH3G3dET9f-s

It looks like we might have a new coder too. Kronos seems qualified and interested so I'll be spending some time trying to get him up to speed.

[Serial Kicked]

Hi there, it's hard not to notice the massive visitor income from this board, thanks for that. Even if it was way too early. So, yeah, i did read about Serial Killer RL but instead of doing a poor copy of something that's impossible to make in the first place i decided to build a city simulator with a relatively depth GTA/crime related gameplay.

(btw, yes I play DF and read this forum regularly)

If you have questions about GRA, you're welcome to ask, preferably on the development blog. Post a comment on whatever article, and you'll get an answer. However I am not interested in feature requests, yet.

Luckily for us, there's really no competition here as the game seems a lot less ambitious and "serious" than this project. Not in terms of quality, but in terms of realism and depth.

lol.

You shouldn't do that when you don't have any kind of release to back anything up and especially in this particular case (you know the fake serial killer RL and all the related drama). But as I am not a bad guy I won't hold a grudge and give you a few free tips. The more games the better, IMHO, anyway. First, I have written a roadmap for you so you can avoid to read the boring design documents all over my website and stop saying this kind of stuff

Two,

You have a few interesting ideas there and there, but it's lost amongst waves after waves of impossible things to implement correctly. I don't recall the programming language, but if it's LUA/script based, it's slow as hell and not open/dynamic-world friendly. So put your game in a "feature locked" mode, and work toward building a half decent city. Nothing fancy, just NPC following their schedules (whatever AI system you want to use). Then, and only then, you'll be able to figure what kind of feature can be included without killing the CPU. Believe me on that, anything said before having a relatively good working environment, should be discarded for this kind of project.

Cheers,
SK.

[Kusgnos]

Serial Kicked, I don't think Lap intended that comment as an insult of any kind.

Anyway, it cheers me up to see that this is still active. College and its associated work has kept me away from Bay12 for the most part, so I look forward to any new coding additions you add, Lap.

[Lap]

I went out of my way to clarify that there was no competition because there is a difference in play style. And yes, after looking at your road map along with the rest of the site it still looks that way. If you want to be offended by the difference in direction, go ahead.

[Serial Kicked]

Err, did i really sound offended ? Because i wasn't. Amused would be a better word here.
Oh, well, sarcastic humor and non native english don't mix well, i should know that by now.

Anyway, the more the better like i said. The advices I gave you were sincere ('been there done that' kind, feel free to ignore), with a little bit of trolling inside because it seems the usual method in this board. Anyway, my apologies if I have hurt your feelings, it wasn't intended at all. Still, my main point stands, gameplay and rule related discussions with the community are irrelevant until you know what your engine can or cannot do CPU wise (especially when you don't know the agent related AI that will be used in your engine yet). 

Cheers,
SK.

[Lap]

The discussion is not irrelevant because most of the mechanics that get suggested and discussed are not CPU intensive. The only things that are, are the requests for fully scheduled AI. A Dwarf Fortress like game where every single citizen is fully simulated with desires and all sorts of detailed routines is not really feasible or worth it. I'd rather just do more elaborate out of sector calculations.

Last year, before this project was started, I did a stress test with a similar Lua engine for a space game. It couldn't really handle the amount of sprites, physics, and AI. Unfortunately, I already wasted a couple days coding in a ship editor. Lesson learned.

Before this project was started I took the already large Tome module and did a stress test with intentionally extra laggy AI that spammed AStar pathfinding and sight checks for no reason. Aside from the engine taking a couple more seconds to load up maps than engines in other languages, the actual turn to turn time was acceptable. The engine has since had 12 major revisions since then so I'm even less worried.

[rawr359]

Err, did i really sound offended ? Because i wasn't. Amused would be a better word here.
Oh, well, sarcastic humor and non native english don't mix well, i should know that by now.

Anyway, the more the better like i said. The advices I gave you were sincere ('been there done that' kind, feel free to ignore), with a little bit of trolling inside because it seems the usual method in this board. Anyway, my apologies if I have hurt your feelings, it wasn't intended at all. Still, my main point stands, gameplay and rule related discussions with the community are irrelevant until you know what your engine can or cannot do CPU wise (especially when you don't know the agent related AI that will be used in your engine yet). 

Cheers,
SK.

No, you don't sound offended. You sound like a cocky bastard who has no real reason to post.

[Soadreqm]

Yesss! We shall become archnemeses! Exchanging rude notes, posting vitriolic rants, and if any actual games ever get released, inserting some insults into them too. It will be so much fun!

[mendonca]

Hey SK, just so you know, you do sound a little cocky - but that's okay by me, you obviously know what you are doing so good luck to you with GRA.

but Rawr, was that really necessary? The opening line of SK's post was thanking the forum for the extra traffic, I think he deserves the benefit of the doubt?

Ninjaedit:

Yes, every crime focused roguelike clearly needs an archenemy, it couldn't have worked out better!

[Serial Kicked]

@rawr:

I did not want to cause any kind of drama. Like I said, I tried to put some humor in my post in a non native language, it failed and like I said (again) I should know better. Anyway, I apologized, case closed. Flame all you want, I couldn't care less (and if i sound 'cocky' here, it's intended).


@Lap:

I have been following your project for quite a while with interest, otherwise i wouldn't be posting here no matter what other people may think. TBH the "we got tiles" post was what motivated my post in the first place (+ the income of visitor i wanted to thank), it got me a bit worried. I've seen so many projects gathering a hell lot of good ideas dying because of engine related issues that i don't want something i follow to do the exact same mistake. And I am glad to know that you've extensively tested the basics beforehand.

Out of interested, what space game are you talking about ?
(I have probably played with it like most if not all space games)

Cheers,
SK.

[Ioric Kittencuddler]

Space game?  Hm... single player Shores of Hazeron with AI empires would be so amazing.