|
Hey cool, thanks for the new thread. You can add me to this thing if you like, my twitter is @grandseiken but I mostly just follow people and don't use it very much. Riding the exhilaration of having released an actual complete game my next project is obviously a roguelike that will optimistically take 90 years to make. In celebration of the new thread I wanted to show it off but since roguelikes aren't suited well to videos or screenshots I'll just throw up what I have right now. (If you start in the dark which you should probably restart for until you get an indoor unlit map since it's way cooler, press . to pick up the matches and torch which will be at your feet, W to wield the torch and then l to light it. Pretty much all you can do right now is set things on fire and get attacked by bats which also set you on fire. The rest of the (implemented) commands are listed if you press ?) (If you have any comments about it please say so, I don't know much about UI design) seiken fucked around with this message at 00:38 on Sep 15, 2012 |
#
¿
Sep 15, 2012 00:22
|
|
|
|
| # ¿ Apr 26, 2024 13:52 |
|
|
Mug posted:I picked up a matches and a torch, pressing W was to Wear stuff, though, and it kept saying I can't wear that. I was to Wield, but the wield menu didn't seem to respond to any key presses. After a while i keyboard mashed and my torch caught fire but I don't know how I did it. Oh, capital W is different from small w; w is wear, W is wield Edit: the random maps are the only other feature that aren't completely half-assed right now. seiken fucked around with this message at 00:47 on Sep 15, 2012 |
|
#
¿
Sep 15, 2012 00:39
|
|
|
Ularg posted:Worked my rear end off for a month, failed with an 82% I don't understand, are exams in the states way easier or something? Here in the UK it's usually something like 70% for an A and to fail you gotta get under 50% or even 40%.
|
|
#
¿
Sep 17, 2012 02:15
|
|
|
If the biggest performance issue is the 2X SDL graphics scaler that would explain why better video cards don't seem to improve performance: as far as I know SDL isn't hardware-accelerated at all. If you could just switch that out to run in OpenGL or something I'm sure you'd be golden cause your game really doesn't look like it should be straining the hardware. I'm not sure if you have some limitations that are forcing you to use SDL though
|
|
#
¿
Sep 18, 2012 12:23
|
|
|
ChesireCat you went to a lot of effort on that great post so I don't mean to annoy or anything, but I saw you (not exclusively, a couple people did it) use "dereference" to mean "set variables referencing [the object] to null". I think this isn't the correct term and in a programming context it already has a meaning (getting the value a pointer points to) so calling it that could be a little confusing, even though it's a reasonable name for it otherwise. I don't know what the correct term is since I rarely use garbage-collected languages, sorry! 
seiken fucked around with this message at 22:45 on Sep 24, 2012 |
|
#
¿
Sep 24, 2012 22:41
|
|
|
ichibanMuffin posted:I haven't yet added things like scripted events (either talking, or a scripted animation/cutscene) but it seems to me that it would also be a hassle that should be planned in advance, along with multiplayer: if a game is planned to have any kind of multiplayer it should be set up from the very beginning with that in mind (with the quake style local server/client setup) to avoid the problem minecraft fell into with having two separate code bases for singleplayer and multiplayer. I imagine some books or articles would help but I'm not sure where to start looking. For instance, most of the stuff I read about for XNA tutorials were just that: Tutorials with no forward planning or only hinting at the logic of a layout without explaining WHY the layout of things were chosen. (And I'm not referring to why data structure x or algorithm y is chosen over alternative z; I'm a CS major, I'm talking about bigger picture choices) You should never look to tutorials for advice about general architecture design because even if the tutorial isn't poo poo (most are) it's still going to be a minimal example to demonstrate whatever end result rather than having a good design. I would advocate it's not the kind of thing you can really learn from a book even because there's no "one approach fits all" kind of design and it depends heavily on what you want to do - especially in a medium as diverse as games. Even if a tutorial or book is particularly focused on how to design a large system it's not going to be much use because it's very difficult to understand the whys of a good design decision without actually doing it yourself a terrible way and experiencing first-hand the limitations of a bad design. This is why the advice is always to start with small games and work your way up because that's the best way to learn stuff like this: by starting with something manageable and seeing what works and gradually progressing to more complicated systems. Once you finish a game of size N you've demonstrated that you can design a system of complexity N that isn't necessarily perfect but is just about good enough to support a game of size N (probably with some hacking and cludging towards the end). You can then use the experience and knowledge you've gained to either make a new game again of size N with a cleaner, more well-thought-out design; or you could hope to complete a game of size N+1 with another good-enough-but-not-perfect design. If you jump too many levels of complexity at once (YMMV on what constitutes a level of complexity) you'll inevitably get too much poo poo wrong, your code will become a confusing mess and you'll give up on the project. If you take smaller steps you have much more leeway to make some mistakes and still manage to finish something. I mean it sucks because who doesn't want to make something huge and awesome right now? But making big games is really loving hard! There are few other coding challenges (compilers are one example) that require you to flex as many different fundamental programming muscles to get properly going. seiken fucked around with this message at 21:12 on Oct 4, 2012 |
|
#
¿
Oct 4, 2012 20:31
|
|
|
Those whole discussion is kind of daft. Don't make engines, make games. Summary: your indie game does not need an "engine". If you want to make your game in unreal or unity or whatever that's another story but if you're doing things yourself: forget about the whole concept of a having an "engine" because it will only hinder you. If you have not completed a whole bunch of games any "engine" you write will be completely useless for using to make any nontrivial game. If you are sensible and write a game rather than an engine, whatever "engine" parts of it you might be able to extract will be (mostly) useless for making any nontrivially different game. Also, "engine" doesn't really mean anything. Making a game mostly from scratch is not at all a waste of time as opposed to using unity or unreal or whatever. vvv good attitude! seiken fucked around with this message at 16:22 on Oct 6, 2012 |
|
#
¿
Oct 6, 2012 12:11
|
|
|
SSH IT ZOMBIE posted:https://www.youtube.com/watch?v=CTXo2PHcsM8 Hmm, opinions on this mechanic? This is basically Yoshi's Island so I can't see how it could be a bad mechanic.
|
|
#
¿
Oct 14, 2012 11:16
|
|
|
Gordon Cole posted:I think a major problem with the icons is the lack of contrast in the color palette. The musical note is pretty obvious, and the lighter faces aren't too difficult to make out, but most of the colors kind of just blend together. I had to bump up my brightness to be able to see the contents of the first icon without moving my face up next to the screen. Yeah I definitely appreciate what you're trying to do here and it's a good effort but I still agree with this post 100%. Could you perhaps double the size of the icons so that you have a bit more room to work with?
|
|
#
¿
Oct 15, 2012 20:10
|
|
|
Lord Humongus posted:Does this look better? It's now "I can hardly see anything" as opposed to before's "literally a 90% black square". Like, almost playable but it'd probably give me a headache... I'd still make at least some bits brighter if I were you. As above posters mentioned contrast can be good for tension. For instance the earlier SH shot, even if your immediate surroundings are bright and mostly completely visible it's still pretty spooky to have no idea what's in the darkness a few feet away.
|
|
#
¿
Oct 18, 2012 19:19
|
|
|
Yeah that's great and I'm pretty sure would work way better against your backgrounds than the current sprite.
|
|
#
¿
Oct 21, 2012 23:22
|
|
|
Actually I already said this was awesome but it struck me right now that it'd be even more awesome if you could draw the static main body part another couple of times or so and loop it (like just draw exactly the same thing so that the slight inconsistencies cause it to flicker in a nice way. That'd be great and also help it pop out from the background more which you were worried about)
|
|
#
¿
Oct 22, 2012 11:52
|
|
|
the chaos engine posted:Ambushsabre and I did a lil' crunching these last couple of days and now we have this: This is cool, but with such awesome & atmospheric art I kind of wish you were making a bit more of an interesting game to go along with it...
|
|
#
¿
Oct 31, 2012 15:38
|
|
|
Yes, the level will be predictable based on the random seed unless you're doing something over the top and weird like reading from /dev/random.quote:but there has to be more to it than just "if random() is between 0.0 and 0.25, place this obstacle; 0.25-0.50, place this; etc There isn't really more than this, just adding more layers of random choice essentially (or other abstractions around generating random numbers). Like, if between 0 and .25 do {thing involving more random decisions}, and so on. A good thing to do is use weights: have a system where you can say, "do this thing with weight n, this thing with weight m", such that something with weight n is five times more likely to happen than something with weight 1 (it's easy to generate results with the correct probability given the weights). Then you can implement things like difficulty by ramping up the weights on the harder game elements. Edit: if you're familiar with C++ you may be interested in this weighted generator thing I wrote (a long time ago so I can't guarantee it's well-coded) which forms the basis for all the randomness of a roguelike I hack on every now and then. Basically Generator<Foo> generates objects of type Foo and you can add any subtype of Foo to it with a given weight, or you can add other weighted Generator<Foo>'s to it for a hierarchical system. Then you generate stuff to your heart's content. Here is an example of how I use it at the top of this file. I'm probably doing stupid things everywhere. seiken fucked around with this message at 17:46 on Oct 31, 2012 |
|
#
¿
Oct 31, 2012 17:13
|
|
|
the chaos engine posted:It'll be more interesting! Tip of the iceberg and all that. For sure but I still hope by "more interesting" you mean "having absolutely no similarly to an endless runner/helicopter game"
|
|
#
¿
Oct 31, 2012 17:44
|
|
|
Polo-Rican posted:I'll probably compile a second version that's double-wide, for people that have two monitors. I'd love to make a version that runs in two totally separate windows, but at the moment I have no idea how to do that. Haven't looked into it though, so maybe it's possible. It's not the kind of game where you need to keep an eye on space at all times anyways, though — There's going to be a lot more scanning, planning, and puzzling than there will be ship-to-ship combat. And not to say that the game will be easy, but your ship will be able to take a few big hits before it blows up. I also want to say that looks totally great with the characters in motion and seems like a really unique gameplay set-up! As for the view-switching, the first idea that popped into my head is maybe you could have a big screen on the back wall behind the characters with some overview of the space layout? Perhaps keep the in-ship view you have now, but when the player moves the mouse up near the top of the screen then the camera pans upwards to show the space monitor, or something like that? Just an idea.
|
|
#
¿
Dec 3, 2012 22:26
|
|
|
Sorry for bringing this up from the last page but I feel like I missed my opportunity to rant a little about pixel graphics and out-of-place stuff: it irks me when a game is all like indie pixel retro YEAH but it's only the assets that are done in a pixel style and the output isn't actually constrained to pixels. The most obvious example is linear interpolated (rather than nearest) sprite scaling, which thankfully looks so horrible that almost nobody makes that mistake. But subtler examples show up even in otherwise very professional games and it looks loving amateur: pixels that aren't actually aligned to any grid, so the environment is fine but the characters on top of it are drawn a fraction of a pixel out of place, or when something starts rotating and the pixels turn into diamonds jesus gently caress (you see this everywhere) All you have to do is render at the actual pixel resolution and then upscale to the monitor, it's not hard. Related, but probably nobody other than me cares about something so anal: when the sprites have nicely chosen limited colour palettes but then graphical effects or post-processing sporadically adds like a thousand new colours and ruins everything. vvv yeah I didn't mean to direct this at anyone in this thread, it was just a general rant. Your game looks baller. seiken fucked around with this message at 14:12 on Mar 21, 2013 |
|
#
¿
Mar 21, 2013 13:00
|
|
|
Mug posted:The camera is given a list of things which are "Worth being visible", and 90% of the time the list is just the player's mouse cursor, and maybe one playable creature. The camera looks through the list of "Worth being visible" objects and records the North-most Pixel, East-most Pixel, South-most Pixel, West-most Pixel out of all the X and Y axis of everything on that list. So if the mouse is at 10,10 and a player agent is at 50,40. The Northmost is 10 Eastmost is 50, Southmost is 40, Westmost is 10. The camera works out the Pixel that's "Halfway between East and West" and makes that it's own X location, and which pixel is "Halways between North and South" and makes that it's own Y location. What prevents the averaged camera position from actually showing nothing that's worth being visible, for example if the two things on the list are farther apart than the camera dimensions?
|
|
#
¿
Mar 24, 2013 14:26
|
|
|
Mug you are the best
|
|
#
¿
Apr 2, 2013 10:00
|
|
|
thedaian posted:Added a construction period for rooms. Horribly mangled animated gif found here I'm not sure if it was you who saved it as a JPEG in an intermediate step or if the site you used was so terrible it converted to a JPEG in-between, but if it was you don't do that and you'll get much better results.
|
|
#
¿
Apr 9, 2013 21:24
|
|
|
Mug posted:I haven't used Blitz or even heard of it until like a fortnight ago when some guy who works for them tried to convince me to change Black Annex to it. I just ignored him. This is hilarious and my new favourite gamedev story.
|
|
#
¿
Apr 15, 2013 11:24
|
|
|
octoroon posted:so you don't have to go down the "how do I learn modern OpenGL when the entire internet is filled with nothing but outdated tutorials for deprecated versions" rabbithole. Not invalidating your advice here, but just in case anyone really does want to learn proper modern OpenGL, I recently found this so recent it's not even finished yet book/tutorial thing which is really good (and has a lot of sections that are pretty interesting even if you're not strictly learning OpenGL). In particular, it explicitly doesn't use the fixed-function pipeline ever and starts off teaching shaders from the very beginning.
|
|
#
¿
Apr 16, 2013 15:25
|
|
|
Aaaa White Dragon that scene's the coolest thing but I really want to see it in motion 
|
|
#
¿
Apr 22, 2013 01:45
|
|
|
Working on a map editor for this game, fairly close to being usable
seiken fucked around with this message at 00:00 on Apr 30, 2013 |
|
#
¿
Apr 29, 2013 23:57
|
|
|
Mug posted:https://www.manfightdragon.com go crazy Mug, it looks loving awesome, but maybe you could tone it down a bit? Make it more subtle most of the time and only crank it up for effect sometimes (e.g. cutscenes, player characters taking damage)? It's great to look at but my first impression is it might be a little distracting when you're actually playing and everything is in motion.
|
|
#
¿
May 4, 2013 15:15
|
|
|
That is supposed to look like an Escher drawing, right? It's crazy! I don't know how moving around in it would work, but make it happen I guess
|
|
#
¿
May 4, 2013 18:44
|
|
|
Polo-Rican posted:Seeing image makes me wonder if you should make it a conventional zelda-like, or if you should totally go 100% the Escher angle and have the rooms actually look like this. I completely agree. That tileset is wonderful to look at just on its own.
|
|
#
¿
May 8, 2013 16:28
|
|
|
Gah! That looks lovely but don't rotate your pixels argrgrh
|
|
#
¿
May 10, 2013 11:09
|
|
|
Implemented dithering. It's a shader in GLSL and it dithers each channel independently for a more even effect and then animates them all in different directions because gently caress why not Something along the lines of what I might use (9 bits per pixel): link because it gets oddly resized inline. I totally dig the static-y, grainy effect it gives (look close, it's animated). Just for fun, it's kind of surprising how much detail you can preserve with 8 colours total: 8 colours (i.e., red green and blue are each either fully on or fully off) seiken fucked around with this message at 00:49 on May 15, 2013 |
|
#
¿
May 15, 2013 00:07
|
|
|
field balm posted:Mainly, I was wondering if there is an easier way to blit parts of a spritesheet without hand-writing loads of coordinates for each frame? It's not hard, just really time consuming. Correct me if I'm misunderstanding what you mean, but if you make all the frames exactly the same size in the spritesheet then you don't have to manually figure out any co-ordinates, and if you're worried about the filesizes I'm pretty sure a png or whatever will compress away all the empty space anyway.
|
|
#
¿
May 15, 2013 00:47
|
|
|
Mug posted:That logo says Moly Wow. Badass. Make sure you wave to the OUYA guys in the parking lot on your way past.
|
|
#
¿
May 23, 2013 15:20
|
|
|
FuzzySlippers posted:This is my vote for thread title. This is incredible, please forget everything else and make Attacked By Bees: The Game.
|
|
#
¿
May 24, 2013 00:01
|
|
|
I'm rather convinced everyone who reads this thread also reads the Cavern of Cobol gamedev thread, but I guess I'll do the crossposting thing like everyone else just in case: So, recently I've got a lot done to the point where this huge pile of C++ is almost starting to resemble a game. I implemented sweep collision detection, a basic camera, tracing of the world geometry (you can see it in the gif when I hit the debug button to show it). Most importantly, Super Metroid-style wall-jumping, the best mechanic ever  (click to open in new tab if it's not animating, seems to happen sometimes) Working title Super Fungus Adventure II. It would definitely look more impressive with better placeholder art than that stupid loving mushroom I found somewhere. Also I should work out how to put gifs on youtube (my game just takes each frame it renders and saves it as an image, and then I make a gif from all the frames)
|
|
#
¿
May 26, 2013 21:35
|
|
|
I have a first draft of a lighting system for my thing. Since it's a huge gif, here's a demo linked on another page. Next I gotta do normal mapping, so the shadows and highlights on the actual tiles are consistent with the lighting. seiken fucked around with this message at 22:12 on Jun 3, 2013 |
|
#
¿
Jun 3, 2013 22:05
|
|
|
Zizi posted:Very nice! I've experimented with doing shadows like that through pixel shaders, but the result wasn't nearly as crisp as yours are. Thanks! I posted some more technical details in the cavern of cobol thread recently. The heavy lifting is done by pixel shaders, yeah. DeathBySpoon posted:This looks great- it's something I'm going to try implementing myself, soon. Yours looks fantastic, awesome work! Cheers. Let me know if you run into any issues. I think we're working on projects that are spiritually very similar (Super Metroid with tech that makes use of the compute-power we have today is my philosophy, ignore the daft placeholder art). Your climbing animation is super smooth, I'm impressed you got it so quick, you said you didn't have much animation experience right? seiken fucked around with this message at 00:31 on Jun 4, 2013 |
|
#
¿
Jun 4, 2013 00:27
|
|
|
DeathBySpoon posted:I'll definitely check out your posts about the lighting system, the more direction I can get with that, the better. I don't know how useful it'll be, it's fairly specific to my set up but here's my implementation, the comments might be something at least.
|
|
#
¿
Jun 4, 2013 00:52
|
|
|
First try at water effects. Looks alright so far, but once I add specular highlights it'll look a lot better, I think.
|
|
#
¿
Jun 24, 2013 00:20
|
|
|
Orzo posted:gifcam test: What I do is capture the gif directly in my game code. That is, I have some keybindings to control capturing a gif in the game and when it's activated I turn off all the framerate/frameskip logic, so that every frame is guaranteed to be rendered, and use glReadPixels to grab each final assembled frame just before it's rendered to the screen. Save all the frames out to images when recording is over, and manually combine them into an animated gif. It's hardly any code at all, and gives the best possible result. Any external program isn't aware of or controlling your framerate and is likely going to drop frames because of it. Edit: also be aware a delay of 2 hundredths of a second is the shortest that works on most browsers, slightly above the 16.6~ms you want for 60fps. seiken fucked around with this message at 19:05 on Jul 7, 2013 |
|
#
¿
Jul 7, 2013 18:44
|
|
|
I wanted to add some cool specular highlights to my water effect. The end result isn't that much different from the last thing I posted probably, but doing this (particularly that the highlights only appear when the light is outside the water) forced me to implement a whole bunch of systems I was going to need eventually anyway, so it's been a while. GIF: GIF
|
|
#
¿
Jul 15, 2013 19:32
|
|
|
|
| # ¿ Apr 26, 2024 13:52 |
|
|
I adapted my lighting system to handle lights that emit light in a particular direction from all points along a line, which makes stuff like this (not quite perfected yet, but mostly working): GIF Edit: at some point I need to stop working on fancy graphical effects and make a game. seiken fucked around with this message at 02:17 on Jul 24, 2013 |
|
#
¿
Jul 23, 2013 23:19
|
|