« first day (3662 days earlier)      last day (46 days later) » 

1:18 AM
I mean like, there are lots of factors to take account when calculating target velocity to reach a position
we might calculate some velocity on which will lead us to a terrain that has a high friction value or the wind might just change or something else that will screw up our calculated velocity
 
2:07 AM
hello
 
2:38 AM
Hello
 
Trying to type with a cat in the way....
You?
 
i am alright, trying to add a playtime counter to my game while also get my xbox to update some games properly
 
2:56 AM
i could actually use a bit of help, if no one would mind
nvm, i think i got it
 
Well done. :) Ping here if you need anything after all.
 
Ok, thanks!
 
3:30 AM
@Henri This is not velocity calculation, it's path finding
 
 
7 hours later…
10:02 AM
Seriously do users not see their questions after posting?
How do people not go "Oh, whoops, my post is a total mess because I can't manage to format my code, better fix that so people can focus on answering me instead of fixing my formatting"???
 
nwp
People have better things to do than format their code on some website.
You can argue that then other people should have better things to do than answering their question, but that's not how it goes.
And arguably it's SO's fault for not integrating auto-formatters.
 
Yeah it screams "I can't put minimal effort into getting help, but please help me answer soon plz"
 
nwp
It's efficient.
Cross-post on Reddit and Quora and you probably get your answer quickly.
Generate a few accounts in case you get banned.
 
IMO we need a mechanism to prevent users who put no effort into their questions from posting before completing some kind of gamified formatting tutorial
 
nwp
Why though? Making people put effort into it is bad. Ideally we'd perfectly answer zero effort questions.
 
10:14 AM
I think ideally we wouldn't see them at all
 
nwp
And if we can't because critical information is missing we should make it easy to detect and fix that. Ideally with minimal effort.
 
 
4 hours later…
2:34 PM
@Tyyppi_77 I ask myself this daily. :/
I feel like there's not a lot of appreciation that this site represents a community of collaborating human beings. Folks often treat it like a code vending machine, without considering the perspective of the users who need to read their question to help them, or learn from the exchange in the future.
 
yeah
 
3:09 PM
Also agree. For me it's part of the burn out I feel here.
 
3:25 PM
So i need a way to listen for value changes of components inside my ecs... is double buffering the only way to realize this ? If so i have a little problem, in c/c++ i could simply copy the struct into new memory... thats a bit harder in java :/
 
I'd love to chat through ways we might be able to turn that around and encourage more community-mindedness/community-building around our site. ☺
@genaray Setters are a common way to catch changes - is that applicable in your situation? It's generally cheaper to mark the change when the change is made, than to try to compare all your data after the fact to find the bits that changed (which are often the minority)
 
^
 
Observer design pattern (or any pattern for that matter) doesn't care if you use Java, C++ or a Turing machine made out of rocks (the API for the latter suck though).
 
hahahahah
are you referring to this? xkcd.com/505
 
@DMGregory Well that could work too ^^ and a lot easier than tracking the whole history of nearly each component.
 
3:32 PM
Haha, I was just looking for that comic! The API most definitely sucks. 🤣
 
Ha! No, I wasn't intentionally referencing that, but it's on point, funny, and very similar to thoughts I had when I took CS theory I.
 
:D
 
To follow up on patterns, though, I recommend starting with game design patterns. It's more approachable than the hallowed Gang of Four classics (which aren't bad, but aren't tuned for game dev & a bit more dry).
 
+1, Game Programming Patterns is a brilliant resource.
 
Seriously though, if "copy the struct into new memory" is good enough, you can do that in Java.
Java doesn't prevent you from making copies into new memory. Why / how do you see the language as being a barrier here?
 
3:40 PM
@Pikalek Best site ever... i love the prototype-pattern :) Type-Object is also very usefull... not sure if you can implement that one into a ecs nicely. Service-Locator is my main source for decoupling :p
 
Glad it's useful. Author also has a neat write up on line-of-sight code as I recall. Probably some other stuff too that I just never even got to.
 
ECS kind of is a form of the Type Object pattern. Instead of defining different classes for all your entity types, you form them out of data as different combinations of components and component parameters. ;)
But on a more micro level, there's nothing to stop, say, a Tile component in an ECS tile map from referencing a TileType object that stores shared terrain information used by all Tiles of one kind.
Similarly, my Weapon component could hold a reference to a WeaponParams object that defines the rate of fire, bullet spread, etc. attributes shared by all similar weapons, rather than duplicating that data in every instance of the component (getting into the flyweight pattern here)
 
Yes, used carefully. Applied indiscriminately, eventually you can effectively remove the advantages of your ECS. Which underscores: patterns shouldn't be used as straight jackets or silver bullets.
As you say though, there's nothing stopping one from applying a change to a pattern when it serves the greater good of the project.
 
3:57 PM
That makes sense ^^ thats something i could do once i have a usecase for this... currently theres only the "basic" stuff like... position, movement e.g... :D
 
 
2 hours later…
5:46 PM
hello
 
Heyhey
 
How are ya?
 
6:08 PM
Pretty well. About to clear out two tasks that have been hanging over my head since July. :)
 
noice, noice
 
Yourself?
 
I am doing well, still doing a lot of polishing to my game, currently adding an Achievements system
gonna write down all the ones i wanna have the player be able to earn, so i have a good idea on how to write all of into code
 
So far I've only implemented achievements once, for a jam game. Other times I've been lucky enough to only have to design them and set up their data, not do all the hard plumbing. ;)
 
LOL
i'm just trying to get a broad cover of milestones the player can get
both small and great
 
6:18 PM
I like assigning my students this talk as a reading on achievement design: gdcvault.com/play/1018089/Designing-Compelling-Achievements
 
The Stanley Parabel has one of the best achievements: "Don't play the game for 5 years". I'll be earning that one in another year I think
 
interesting
 
Acheivements aren't hard on steam or iOS or android using unity
 
Hahaha
Must've been a pain to test that one...
 
thought about adding one for people who play the game so long it breaks the playtime counter lol
 
6:19 PM
I wrote a system to cross-pollinate achievements across services since our game allowed playing on any device.
 
I get grief from our QC teams when it takes more than 7 days of play to validate a feature...
 
so you could get an achievement on iOS, then when you logged in to android, you'd get it there
 
oh dang, i am just having them as in-game, per save file
 
it worked by coordinating with our own DB
 
I think consoles sometimes have rules against that. ;)
 
6:21 PM
yeah I believe that
not sure how it would work though, you need 10 hits to get achievement, if you have cross-progression...
 
ah well, ain't like Microsoft or Sony want my silly game on their stores anyways :p
 
7:10 PM
@Almo nice!
 
7:21 PM
I think on the Nintendo DS, you could have game features that would depend on something like "at most" one year from now. It had to be "hackable" by changing the console clock.
 
They don't let you do viral achievements anymore either. ;)
 
What's that? Share a link to the game with 28 of your friends?
 
Play a match with a member of the dev team / with someone who has this achievement.
 
Ah, I suppose it's irritating for players who play in a cave..!
 
7:36 PM
Or who pick up the game after the virus has run its course.
 

« first day (3662 days earlier)      last day (46 days later) »