• Hey, guest user. Hope you're enjoying NeoGAF! Have you considered registering for an account? Come join us and add your take to the daily discourse.

Indie Game Development Discussion Thread | Of Being Professionally Poor

Status
Not open for further replies.

Popstar

Member
I vote #1.

There's a lot of evidence that Greenlight is not long for the world. #2 may not have any meaning soon. (Although I'd vote #1 even if I thought it was around for the long haul.)
 

bkw

Member
Between those two, I pick 1.
Not all indie devs are shooting for Steam. (Though I suppose we aren't all going high def either...)
 
Yeah, I'd got with option 1 as well. It's a more general thing that's easier to "get", I think. I don't think we want to come off as whiny Greenlight devs, either.
 

friken

Member
This reminds me - what was the approval process like? I'm in the program but going through those hoops right now. Any advice on the pitch?

I signed up early last year, early in our dev process. Microsoft was behind with all the initial interest and I didn't hear back for a long time (could have also been because of how early we were in dev- - priority to projects further along). mid November they they sent me a a link to submit the game concept for review (they call it a Game Information Form). The GIF wasn't much of a pitch. It was more or less just info on the project... at least I didn't' fill it out like a pitch. I'm really not sure what criteria they use to determine what to approve or if they approve everything one all the forms are filled.


another vote for #2
 

Five

Banned
I'm not convinced that running with the "bah, we're so poor" byline again is the best idea, but of the options I like the wordplay in the first better.
 
icLogo.png


FINALLY - I present my humble trailer for Infinite Cosmos
Protip: A really sweet set of cans or something with a subwoofer is preferred

Completely crushed YouTube version (1080p 60fps):
https://www.youtube.com/watch?v=NPzIF7wM43U

Uber quality download (1080p 60fps):
http://downloads.absinthegames.com/videos/InfiniteCosmosTrailer60fps.zip

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

8-5 now... DAAAAAAAAAAAAAAAWWWWWWOOOOOOOOOOOO!

I'll give the votes until I get up on Sunday.
 
primitivesplash50ufy.png


Another quick update, since that seems to be the done thing. Almost finished with level 4's environment design. I have one last puzzle that introduces moving objects, something which has only been involved in a minor way so far, which I'll probably put up a video of tomorrow once I have it working.

Also on this weekends to do list is to fix my save game function (which broke when Unreal Engine 4.6 came out), to rework the pause menu, and to add comfort mode for Oculus Rift play. For anyone that hasn't heard of comfort mode, some people find that having a control method that instantly snaps the player 45 degrees clockwise or anticlockwise with a button press or flick of the stick to be much less motion sickness inducing than analogue turning when playing something in VR.

I don't have any problem analogue turning, but it's a very simple thing to implement and I think it'll work just fine given that my puzzles are all built orthogonally so far (and given that I intend them to stay that way).

I've plotted out a puzzle by puzzle breakdown of level 5 since I'm almost done with level 4. I've also started doing some work converting BSPs over to static meshes. I'm not going to do this for every situation, since my BSP counts are far from crazy and no one has complained about performance yet, but where it's not too much of a headache to create the collision meshes, I'm doing it.

The puzzle I built out tonight I built entirely with static meshes.

Since it's pretty similar to the walkway area I posted about before, here's an editor shot.


A couple of things about this. You'll notice the default texture all over the place. The way I'm achieving the look of the game is to not include any lights in the game world at all, and to use glowing materials to provide the highlights where I want them (the green and the blue glowing surfaces in the preview viewport). I have a worklight blueprint I can drop into any level, that has a simple piece of blueprint code that turns the light off the instant gameplay begins.

I attach the preview viewport to this worklight so I can see what I'm doing while designing, and it instantly disappears when I preview.

You can see how much of the map is made of BSPs from the blueish purple colour. The greenish blue colour stuff are the static meshes. You can see the puzzle I built tonight as it's the only area entirely made from static meshes.

I really hope I'm not stepping on Michael's toes. Obviously he has been working on his game longer than me. I've been working on this since summer 2014. I got the idea for my game reading a news story about Vantablack and had no idea someone was doing something similar. I'm glad in a way, because I'd never have started it if I thought someone else was exploring the idea.

I don't expect Primitive to accomplish much. I'd love for people that buy a consumer Oculus Rift to spend some time with it. It's not going to be especially long, or deep. I hope to do something engaging enough from a writing standpoint, and I think I have more than enough puzzle concepts for the 13 levels I have planned. Hopefully that's enough to warrant a download for Rift owners.

It's been a while since I finished a major project (last one was a novel a few years back) and this is my first (and quite probably only) attempt at making a proper game. I doubt I could come up with another idea that I could work on all by myself... but it continues to be a rewarding project to spend time on.
 

scaffa

Member
It is not suitable for high def graphics, as you will run out of memory quite quickly. I was using large resolution images (1080) and the game would not run with some people's computers doing a limited release of the game.

I can't divulge the Wii U specs, but suffice to say the Wii U also has problems with memory and resolution (concerning NWF) so graphics have to be on the low rez side there as well.

Yeah I can understand the choice now. I actually encountered the same problem with my game and had to do some redesigns. Big images files are a no no with C2 :) You can immediatly see the, estimated, memory usage skyrocket when you use them.
 
I give a vote for #1.

Though I am not in greenlight process yet, but I think greenlight needs to change somehow to something better, I think it is not working as it is now.

Year end was quite slow for me, lots of things to do at home so little time for development especially when stuck in some shader coding but finally was able to make this to work... now I have procedural earth type planets and started working on different planet types.

ProceduralPlanets.png


After this, it will be the planetary landing with parties, that will be fun to do! especially on planets with low or high gravity and many other things...looks like alpha is getting closer and closer but I can't reach it yet :-/
 

Skinpop

Member
How do you do proper screen space based movement?

Say we have a non-first person 3d game, for example, 3rd person or isometric. Now when the user pushes forward on the stick I want the character to move up in relation to the screen, when he presses right go to right in relation to the screen and so on.
To my understanding you simply multiply the movement vector with the inverse of the view matrix. Then you translate the character by that new projected vector.

The direction mapping works as intended for me, problem is the magnitude of the movement is off. The character moves faster(30%) in vertical(screen space) than horizontal. Renormalizing the vector doesn't help.

Am I missing a step? If not guess I have a bug somewhere.
 
Another vote here for option #2

My "trying-to-make-a-game-so-I-can-learn-to-program" game is starting to come along. It has now been 3 years (!!!) since I started, almost to the day.

Very long term:
- Save/load system.

Nice! That's similar to how I'm doing my current game, also just shy of 3 years working on it. You're smart for taking programming classes, it's something i'd like to do if I had time.

I'm about to release my game this month, and I also decided to put the save/load system to the backburner. Man am I ever regretting that. I just can't seem to get it working, tried mulitple different assets, watched so many videos. I've basically given up on it. Luckily my game is only about half hour long. I might just do a simple "load checkpoint" right from the main menu.

For my next game, first thing I do is a save/load system. Lesson learned.


FINALLY - I present my humble trailer for Infinite Cosmos

That is a really awesome trailer man! Good production value. Are there any on foot playable sections in the game?

Here's a .gif of the main menu:
shutter_menu.gif.gif
 
icLogo.png


FINALLY - I present my humble trailer for Infinite Cosmos
Protip: A really sweet set of cans or something with a subwoofer is preferred

Completely crushed YouTube version (1080p 60fps):
https://www.youtube.com/watch?v=NPzIF7wM43U

Uber quality download (1080p 60fps):
http://downloads.absinthegames.com/videos/InfiniteCosmosTrailer60fps.zip

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

8-5 now... DAAAAAAAAAAAAAAAWWWWWWOOOOOOOOOOOO!

I'll give the votes until I get up on Sunday.

This looks really great, are you doing a PC version?
 

kiguel182

Member
icLogo.png


FINALLY - I present my humble trailer for Infinite Cosmos
Protip: A really sweet set of cans or something with a subwoofer is preferred

Completely crushed YouTube version (1080p 60fps):
https://www.youtube.com/watch?v=NPzIF7wM43U

Uber quality download (1080p 60fps):
http://downloads.absinthegames.com/videos/InfiniteCosmosTrailer60fps.zip

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

8-5 now... DAAAAAAAAAAAAAAAWWWWWWOOOOOOOOOOOO!

I'll give the votes until I get up on Sunday.

Really cool trailer!

And the game looks awesome too, puts my humble shmup to shame. Good stuff.
 
That is a really awesome trailer man! Good production value. Are there any on foot playable sections in the game?

Thanks! No on-foot sections. Just single relative touch control. Keeping the core gameplay simple and easy to pick up and play. The trailer was made using a single 12 minute playthrough video I capped ha! Not that I was pressed for time but a lot of things happen in the newest build thanks to finally getting all pieces of the spawn system in place.

This looks really great, are you doing a PC version?

Thanks bro and not at the moment. Considering it and if so, would require a rewrite of enemy AI, spawning, etc to fit with controller/kb-m control scheme to keep the difficulty curve similar. Plus if we were to do a PC sidescrolling SHMUP i'd pull out more stops with supers, weapon upgrades, enemies, etc.

Really cool trailer!

And the game looks awesome too, puts my humble shmup to shame. Good stuff.

Thank you, too! I highly doubt this puts any game to shame. It's just a simple 1-finger drag-n-shoot pew pew space shmup designed around being pick up and play. No real steep learning curve.

-

Aaaaaaaaand now the score is:
9-6!
High def work keeps Steam says I'm green behind in the poll! WooOOOOo!

On a side note I need to finish writing 1 more section then the new thread is done, outside of new graphics for headers and editing for grammatical and spelling errors. Just a few more days. Will be up before 20k :D
 

friken

Member
I give a vote for #1.

Though I am not in greenlight process yet, but I think greenlight needs to change somehow to something better, I think it is not working as it is now.

Year end was quite slow for me, lots of things to do at home so little time for development ...

I agree greenlight seems a bit broken. I think steam has known that for a while given their talk about changing it. I just wish they were more public about how/when it is going to change. The holiday really kicked my rear here... I planned on getting a lot done but instead I'm getting a divorce. Life isn't always what you plan on.

On the bright side though, I'm looking forward to getting my head back into the game and hoping my focus can be better in 2015. I feel like I was only able to get 3 months of work done in all of 2014.
 
I agree greenlight seems a bit broken. I think steam has known that for a while given their talk about changing it. I just wish they were more public about how/when it is going to change. The holiday really kicked my rear here... I planned on getting a lot done but instead I'm getting a divorce. Life isn't always what you plan on.

On the bright side though, I'm looking forward to getting my head back into the game and hoping my focus can be better in 2015. I feel like I was only able to get 3 months of work done in all of 2014.
Sorry to hear :(
 

Popstar

Member
How do you do proper screen space based movement?

Say we have a non-first person 3d game, for example, 3rd person or isometric. Now when the user pushes forward on the stick I want the character to move up in relation to the screen, when he presses right go to right in relation to the screen and so on.
To my understanding you simply multiply the movement vector with the inverse of the view matrix. Then you translate the character by that new projected vector.

The direction mapping works as intended for me, problem is the magnitude of the movement is off. The character moves faster(30%) in vertical(screen space) than horizontal. Renormalizing the vector doesn't help.

Am I missing a step? If not guess I have a bug somewhere.
Are you also re-projecting the vector to the 2D plane of movement? Maybe you should post a code fragment.
 
Drawing persons. For the new thread:

If any of you want to contribute to header graphics with roughly 100x100px sized sprites, drawings, etc, post them here. I'll be going over section headers this weekend. I want to pick drawings that some of you do for the headers.

Nothing fancy. Don't have to relate to any specific header.
 

Five

Banned
Drawing persons. For the new thread:

If any of you want to contribute to header graphics with roughly 100x100px sized sprites, drawings, etc, post them here. I'll be going over section headers this weekend. I want to pick drawings that some of you do for the headers.

Nothing fancy. Don't have to relate to any specific header.

Anything you want in particular? Characters from our games? Something generic?
 
https://www.youtube.com/watch?v=bVk3K-jUB5U

Music is just Youtube free stuff so that the video isn't deathly silent.

This is a portion of level 4 of my game to better illustrate exactly what I'm up to. At the end of the video you can see the unfolding room I added today. It took quite a bit of work to get it working just the way I wanted, but I'm very happy with how it looks.

If you want to just skip ahead to just see that, use this link.

https://www.youtube.com/watch?v=bVk3K-jUB5U#t=152
 

Ashodin

Member
I'm starting on the game idea I have today, just to do preliminary research and work on it. It'll be an exclusive FPS for Wii U, so that's something.
 

Five

Banned
Could be anything :D

Like on the first page, instead of 3 colored boxes I am creating small sprites to use for section header icons. Anything you wish!

Here are a few different things from my game. Not sure which one will work best for you, so I have a lot of options. (crosses fingers that others join in so it's not just me!)

gi01.png
gi02.png
gi03.png
gi04.png
gi05.png
gi06.png
gi07.png
gi08.png
gi09.png
gi10.png


https://www.youtube.com/watch?v=bVk3K-jUB5U

Music is just Youtube free stuff so that the video isn't deathly silent.

This is a portion of level 4 of my game to better illustrate exactly what I'm up to. At the end of the video you can see the unfolding room I added today. It took quite a bit of work to get it working just the way I wanted, but I'm very happy with how it looks.

You broke my brain a little bit, but I think I managed to eventually catch up with what was going on near the two minute mark. This looks great!

I'm starting on the game idea I have today, just to do preliminary research and work on it. It'll be an exclusive FPS for Wii U, so that's something.

Is there a reason you're so fond of the Wii U right now? I got one for Christmas, so it's something I'm potentially looking at now instead of ignoring entirely. Is it just that your experience has been quite good with it?
 
I really hope I'm not stepping on Michael's toes. Obviously he has been working on his game longer than me. I've been working on this since summer 2014. I got the idea for my game reading a news story about Vantablack and had no idea someone was doing something similar. I'm glad in a way, because I'd never have started it if I thought someone else was exploring the idea.

Oh no don't worry at all! :)

On the surface the two games look similar, but we've really taken the same core idea in completely different directions. VizionEck has nothing even remotely like this below:


All my edges have outlines, without exception. Nothing is hidden like with your Vantablack concept.
 

Ashodin

Member
Yeah I'm in the midst of grabbing all the free unity packages and stuff I can use to learn the ropes. It'll be fun for sure!
 
Oh no don't worry at all! :)

On the surface the two games look similar, but we've really taken the same core idea in completely different directions. VizionEck has nothing even remotely like this below:



All my edges have outlines, without exception. Nothing is hidden like with your Vantablack concept.
Fantastic news. I breathe a big sigh of relief. Thanks for watching everyone.
 

Skinpop

Member
Are you also re-projecting the vector to the 2D plane of movement? Maybe you should post a code fragment.

My code looks sorta like this at the moment:


Code:
// Deadzone handling
...
// then ->

vec4 Input = (x, 0, z, 0)  // Y is my up vector so z is the "y" coordinate for the plane
mat4 Inverse = inverse(ViewMatrix);
vec4 ViewSpaceInput = Inverse * Input;

ViewSpaceInput.y = 0;

normalize(ViewSpaceInput);

//  use ViewSpaceInput to update position

In my mind this should work but I'm probably missing something or have a bug in the deadzone handling.

I'm aware another approach is to get the rotation then simply use the forward vector but I want to get this to work before I start writing any rotation code.
 
Here are a few different things from my game. Not sure which one will work best for you, so I have a lot of options. (crosses fingers that others join in so it's not just me!)

gi01.png
gi02.png
gi03.png
gi04.png
gi05.png
gi06.png
gi07.png
gi08.png
gi09.png
gi10.png

OOOOOH I will be using one or two of these. Right-click saving mmmmmmmmm.

Absinthe, #1 sounds good to me.

always poTAYto

10-6!

Low def pay leaving the bank legitimately in the red!

14 more hours for others to cast their vote before the thread title is secure. That is my Sunday late morning.

@Michael @plagiarize
I hereby commission you both to remaking my game STRAFE in the 1st person.

Now back to installing things on my new drive :D
 

Popstar

Member
My code looks sorta like this at the moment:

In my mind this should work but I'm probably missing something or have a bug in the deadzone handling.

I'm aware another approach is to get the rotation then simply use the forward vector but I want to get this to work before I start writing any rotation code.
If Y is your up vector perhaps your input should be { x, y, 0, 0 } ?

I usually don't use a full matrix invert for this. Just the components of the camera matrix I need directly:
Code:
// Transform a vec2 from viewspace to world (XZ) space.  Useful for Mario64 style movement
vec2 Camera::ViewToWorld( vec2 & coords )
{
	vec3 camforward(-m_right.z, 0, m_right.x);
	vec2 ret;

	ret.x = -coords.x * m_right.x + coords.y * camforward.x;
	ret.y = -coords.x * m_right.z + coords.y * camforward.z;

	return ret;
}
 

Skinpop

Member
edit: another question..
I'm wondering if anyone here has experience with implementing and rendering a 3d-tiles system. Essentially the same as 2d tiles, but with tiles being represented by actual geometry. I suspect games like dota 2 and starcraft uses this approach but I'm wondering how they handle the actual geometry.

A few approaches I've thought of:

"Bake" tiles by welding them all together to a single level mesh and upload it all to a single vbo. Lots of geometry data both on file and on video memory. Kinda defeats the purpose of tiles as a data structure for graphics. Texture handling is more complex.. not sure how to cull efficiently.

Bake chunks, have each chunk be sized some fraction of the screen(I'm using a top down fixed perspective). More efficient, easy to cull. seam issues?

Both of the above but no welding just "loose" tiles. Seam issues? Light baking?

Bake the visible tiles runtime, update and upload them all together according to player movement.

same as above but no baking. seam issues?

some kind of instancing?

I guess this isn't the right forum for the question but it can't hurt asking I guess..
If Y is your up vector perhaps your input should be { x, y, 0, 0 } ?

I usually don't use a full matrix invert for this. Just the components of the camera matrix I need directly:
Code:
// Transform a vec2 from viewspace to world (XZ) space.  Useful for Mario64 style movement
vec2 Camera::ViewToWorld( vec2 & coords )
{
	vec3 camforward(-m_right.z, 0, m_right.x);
	vec2 ret;

	ret.x = -coords.x * m_right.x + coords.y * camforward.x;
	ret.y = -coords.x * m_right.z + coords.y * camforward.z;

	return ret;
}

Thanks a lot. I'll see if that fixes my problem tomorrow(been programming 10 hours straight.. need sleep).
 

Popstar

Member
edit: another question..
I'm wondering if anyone here has experience with implementing and rendering a 3d-tiles system. Essentially the same as 2d tiles, but with tiles being represented by actual geometry. I suspect games like dota 2 and starcraft uses this approach but I'm wondering how they handle the actual geometry.
Modern graphics cards have enough precision that you shouldn't need to bake stuff together. Make sure you move your near clip plane as far forward as you can so you're not wasting Z precision.

Do whatever is simplest to start. If you have problems with seams you may need to add small 'skirts' to your tiles around the edges. Once you've got it working you can look into instancing if you need more performance. You can even do simple pseudo-instancing in OpenGL by passing each tile's offset through a vertex attribute.

(We used 3D tiles on Hoard)
 

As a very "light poster" but heavy lurker I'll throw myself in with #2. Colours are always fun to read.

Ended up sticking with Unity since I first asked about it all those months ago, it's been a great way to keep plugging away at C# since working with PL/SQL as my day job. Got a little tired making odd test projects with shape "art" so I partnered up with some folk to help out on a little game they are putting together. Been a couple months since I started, actually having assets to work with is a nice motivator!
 

Blizzard

Banned
If #2 ends up winning, I hope Greenlight goes away like a month after the thread goes up, resulting in very confused people for a year or so. :p
 
As a very "light poster" but heavy lurker I'll throw myself in with #2. Colours are always fun to read.
10-8!!!!!!!!!!!!!!!!

Ended up sticking with Unity since I first asked about it all those months ago, it's been a great way to keep plugging away at C# since working with PL/SQL as my day job. Got a little tired making odd test projects with shape "art" so I partnered up with some folk to help out on a little game they are putting together. Been a couple months since I started, actually having assets to work with is a nice motivator!

Grats, man! Glad things are rolling along for you. What type of game are you guys making?

If #2 ends up winning, I hope Greenlight goes away like a month after the thread goes up, resulting in very confused people for a year or so. :p
Damn your logic. I change my vote from 2 to 1.

11-7 :O

Edit: is that allowed?
 

Kritz

Banned
http://i.imgur.com/HzRJOC5.gifv

*bangs head on keyboard*

network view ids are the worst sometimes

edit:

http://i.imgur.com/pBTVI4b.gifv
get out of the way rats, I'm trying to test new features. Stop making bugs by running into stuff that doesn't work yet! >:eek:

edit2:
http://i.imgur.com/t2zxhvB.gifv
Drawers! There's a weird mishmash of code going on with these to stop unity's physics from physicsing all over the place and making a mess. But it's pretty seamless so far for dropping stuff into drawers and taking them out again.
Players will be able to store stuff in drawers, but more often than not I think the drawers will be pre-filled with goodies and they'll essentially just act as the restaurant's inventory for the day. Because after experimenting with delivery vans for a while, I've discovered that carrying boxes back and forth as a form of resource collection is really boring. So I'm going to expand the scope of the game (just a little!) to make grocery shopping a little more entertaining and involved. Drawers are a small step towards that goal, and to making the restaurant more interactive in general.
 
Status
Not open for further replies.
Top Bottom