tiny RPG, 3D open world: Page 3 - Godot Community Forums

tiny RPG, 3D open world

justinbarrett

it's more compounded with emissive materials...you can really see it when you increase the max health at the "vendor" person......but I am aware..

justinbarrett

Sorry it has been a while since I posted an update...

In general I have been unhappy with loading times for the game and it has caused me to rethink how I am setting up my world and level(s) in general... The original plan was to just have a large open world, and have dynamic loading and unloading of "actors"(npc's and enemies) with a simple spawner and distance check...a spawner would have a list of actors to spawn and keep track of some of their key stats like where they should be depending on time of day or how much health they had.

but since it takes over a minute to load up I took a few days to consider what I wanted to do about it...did some research on threading and thought about the basic logistics in general...

in the end I decided that there would be an overworld that would consist of all the exteriors of homes, enemies that help determine player path(like ones that are too tough to get by) and entrances to caves etc...

this leads to having all interiors loaded as separate levels...much like the elder scrolls games. I feel like this would be a better path to take...feel free to speak out if you have any advice. I will take it into consideration as, I myself, am not master of Godot...that's the whole reason I started this "small" project...to get to know the engine in all it's aspects...

with that being said...it will change the world in enough respects that I have decided I need to just scrap the world that is there and make another one, more in tune with my plans.

aside from that I have been working on the story boarding, which may just end up turning into the cinematics portion...I think some still images with dialogue is enough for a simple game..I don't really want to force the player down any particular path and I think lightly hitting on some key points should do the trick...and push the notion of exploration...

I just wanted to make this post to assure whoever is following this, that it is still under development...it's not gathering dust or anything..and I have not just magically "dropped" it in favor of a new project :).

EDIT: I've managed to get the load times down to about 20 seconds...I still have to rework everything, as the original plan and this plan function quite different.

justinbarrett

finally something to show...I put up another video...took a while, I had to refactor a lot of code to make sub levels work..this is not a very exciting video, but I wanted to get something out there.

Saitodepaula

Hi @justinbarrett , I've been reading your posts here in the forum and viewing your videos for a while now, and the progress has been very nice!

If I can give you a humble suggestion, it would be to jump to the "final boss" of the game. By that I mean, what do you have in your mind, as a complete vision of the game? Will the player be fighting giant dragons or commanding large armies or casting powerful spells against dozens of enemies? Whatever it is, just try to implement it, and see if that is fun.

It may sound counter intuitive to begin by the end, by I've read something like that a while ago from an experienced developer. I've tried myself to make an open world RPG some years ago, and now I know one of the mistakes was to spend too much time setting up the wolrd and the systems, and not checking if my final vision was fun or even possible (no need to say I've never finished it).

I'm not saying you're making this mistake now, your progress seens very consistent, but I think this approach can help a lot a game like yours. By triyng to make the "final boss", you will have an idea of how the player is supposed to be equiped in the end of the game, what skills it will have, and how easy or difficult the enemies are right now. Even placing this "final boss" in your game visible from the beginning maybe fun, like Zelda Breath of the Wild: the player understands from the first moments where he is not supposed to be yet, and I think that is a motivating challenge!

I'm not an experienced nor a professional game developer or game designer, so I might be wrong in many ways, but I think it is interesting to think about how the finished game will be.

justinbarrett

No, your suggestion makes total sense..it's a great way to develop a game. I'm not doing like this for this title for two reasons. 1) this is my learning project, my first ever title using Godot...thus the simple graphics as well...but I am still developing it as an actual title..maybe free, maybe donation based...depends...but leads to my next reason.

2) I also do not want to release any story or game information, aside from basic functionality...

my plan is not to garner a lot or attention or build a large following, but to do something I enjoy while learning the engine strengths, weaknesses and best practical methods for creating a game. I do have a design document(loose) and story(also loose). To work off of.

I hope you keep checking this out, if it interests you :)...I say I'm not "trying" to garner a following, but it's still nice when people are interested in what you are doing.

justinbarrett

I am having an issue with SSAO(built in) I've tried everything..it seems it may could be alleviated with a depth test...but it seems to do ssao for the entire view...does anyone have any tips to clean it up...I've even turned up the bias to 0.025 and turned down the edge sharpness to 0.. I've also just tried the default settings etc etc...nothing seems to help. looks the same when I disable shadows and have distance fog on/off...view distance is set to 1K...clip near is default.

TwistedTwigleg

For me, SSAO has been hit and miss. Sometimes it works great, other times not so much. I have also noticed that it can tank performance, with many of my projects halved simply be enabling it.

It looks like what is happening is you are getting a form of shadow acne and banding. Have you tried changing the radius? Maybe the radius is too large (or small) and that is causing the issue because of the long draw distance. That's the only thing I can think of, outside of bias and edge sharpness, that may be able to be adjusted to fix the issue.

justinbarrett

I have tried adjusting the radius...I get the best results though when I change the bias, but that also removes all the detail...I wish it had a depth value(view z-depth) I think that would be more performant, as well as, grant better results.....but I don't know...maybe there is a way to write my own??? I have not looked into it. Anyone have any experience with screen space shaders in Godot?

Saitodepaula

I don't think it is the cause, but, try to increase camera near clip by an order or two of magnitude (to 0.5 or 1). I've had problems with this before.

And as a more generic comment regarding graphics, are you using 3.2? What if you try 4.0? I'm not using 4.0, but I have forced myself to not spend much time on the graphics side, hoping that when 4.0 comes out, there are going to be many improvements we will be getting "for free".

justinbarrett

I think it is...but....I'm not touching 4.0, even for testing until it get's into a beta stage. I can try increasing the near clip range, but that might lead to camera clipping when it collides with geometry.. I can test it...and will ofc...come to think of it...it looks the same in the viewport, not just in game...I've even looked at my gpu settings.

Reneator

Looking really good! Looks like it has a lot of potential!

justinbarrett

Thank you...I wil probably be quiet here for a while...I started to actually build the game itself..as in set up the actual quests etc but I was running into so many issues with b=structure..the gliding etc(contemplating removing gliding..tell me what you think)...I also decided I needed to scope it down a little...in order to max out all your abilities you would need to level up 2400 times...uh..no. It would extend the game, but it would be a crappy way to do it.

I also have a lot of issues with the ssao...I really don't have the will to write my own, whith Godot 4.x around the corner...it's fine with a limited number of objects...but the more I add the worse it gets...tested it and it litterally gets far worse with more objects...si I also need to scale down the world size..

Basically I am just going to get everything more focused...or I will never get this done.

Megalomaniak

For AO you might be able to get pretty decent results from baking AO into vertex colors in blender. Might have to make sure you have some support loops in place to control the vertex AO better though.

justinbarrett

It's fine. My original shader was set up to only use vertex colors..that is why I was able to just get away with a single shader..I'm still doing that, but with a palette texture...the issue with vertex color is it increases the vertex count...and using the palette is more performant with a large number of models..besides..I've already started and commited to it by completely deleting the old world, the models and the import data so I'm just in the blocking phase. :)

Megalomaniak

You could always bake into a light-map in blender, too.

justinbarrett

ok, so it's been a while since I updated..I was doing serious contemplation if I would continue to develop this.. The whole point of doing it was to learn godot..and I recently ran into some issues when I realized I was going to need to re-do/refactor either just about 80% of the code or go in and start doing things totally different..since I had coded almost everything in this...not even using raytracing nodes..so I have thousands of lines of code...pointless :) but I took some more time to consider if I should go back and rework it(as I said) or just go on to the real project..so yes, I decided I was far too unmotivated to do all that work for something that was intended as a learning exercise and I will instead start working on the game....

it's in the same genre, open world adventure rpg, but with less cute graphics..so I just wanted to clear the air. I'm not going to do anything rash and delete this project folder from my cloud or anything..but, as of now, it is on hold. so final image I guess...for the foreseeable future.

justinbarrett

don't know why there are two images?

TwistedTwigleg

@justinbarrett said: don't know why there are two images?

Probably the drag-and-drop functionality. Both the Imgur plugin and the post editor support drag-and-drop image uploads, so the image gets uploaded and added twice: once to Imgur and once to the forums. I'll see if I can disable one of them so it's not an issue.

Edit: Fixed! Now drag-and-drop should only upload to the forums, so there will only be a single image now when drag-and-drop is used.

Also, I totally get having to drop/sunset projects. It's never easy though. :smile:

Hippo

This has been a really cool project to follow along with, thanks for posting! Looking forward to seeing what you have in store for the game.

justinbarrett

@Hippo said: This has been a really cool project to follow along with, thanks for posting! Looking forward to seeing what you have in store for the game.

Thank you... I just wrote the inventory system,not using any type of database...fully dynamic. I'll probably do dialogue next, then save load...and finally settings and menu. I'm writing all the systems like that before I start actual world development..I have 3 base story ideas...still deciding. I probably wont post anything until I am a bit further in development of the story and mechanics.

« Previous Page