« first day (1817 days earlier)      last day (3219 days later) » 

@Lokkij I am still waiting for discount on that
the Endless Space was kinda disappointing
 
 
2 hours later…
1:59 AM
what language is this: internal var isoDepth:int;
 
2:17 AM
looks like ActionScript 3
the :(type) is a Flash thing, I think
 
oh ok thank you
i was trying to figure out what internal meant. i think its just an access modifier
 
@Almo did you listen to that MoM album?
 
 
2 hours later…
Joe
3:59 AM
whats up @Chris
 
Does anyone have any experience with ubuntu and iptables?
 
@Joe Reading the starred roguelike article atm! also been playing roguelikes, seeing if I want to take a stab at making a sequel to my game, no prototypes for the past 6 weeks have stuck with me lol
 
I want to run a test server. I managed to get the iptables setup so I now have FTP access, but I don't know how to go about setting up for UDP traffic
 
Joe
@William'MindWorX'Mariager sorry man ive never successfully had the patience to mess with ip tables
 
Hehe
It's quite annoying :P
 
Joe
4:04 AM
but i do know "digital ocean" forms have a TON of good tutorials on linux server setup
 
How fortunate since I'm using DO for this :P
 
Joe
@Chris yea man... im getting sick of this thing... i need to put it to bed
 
I've been working so much and been so tied up I haven't done any development on my game since before the exams.
Quite annoying.
 
@William'MindWorX'Mariager I have linux vps set up on digital ocean, maybe I can help?
 
Well, I basically just need to figure out how I open up port 8110 for UDP traffic. It's a Lidgren UDP server I need to run.
 
4:17 AM
@William'MindWorX'Mariager meh cant help you there
 
 
2 hours later…
6:21 AM
@William'MindWorX'Mariager I always end up manually editing iptables config file
but the guide in the net seems to be pretty good
 
 
3 hours later…
9:08 AM
@Jovito I did, it was interesting thanks!
 
 
1 hour later…
10:15 AM
Yes... a blockchain protocol for games, that sounds like a great idea! reddit.com/r/gamedev/comments/3efcf8/…
 
user92578
10:27 AM
@Hjorthenify Actually I just read a few chapters of Game Engine Architecture :)
 
10:44 AM
Hello
 
user92578
Hi
 
any work with basic 2D with opengl?
 
user92578
I tried it a while back, but ran into performance issues and continued developing with SDL2
 
Ok so you pretty much ditched Opengl and went straight SDL2?
 
user92578
Yeah
 
10:51 AM
I'm thinking of going that route
 
user92578
But one day I'm going to write an 2D OpenGL renderer that I can use
 
user92578
SDL rendering is okay
 
user92578
It doesn't have shaders tough
 
user92578
And I spent a lot of time trying to get tiles render next to each other without artifact lines between them
 
My problem deals with rotating and moving my image
Opengl is doing some weird things in my game
 
user92578
10:54 AM
I've been thinking about starting to use assert in my code
 
I've been thinking about using exceptions in my code..
 
@Tyyppi_77 if you are interested in SDL2 + OpenGL for a 2D game, I suggest checking out MakeGamesWithBen on youtube.
 
user92578
okay
 
@Hjorthenify I've heard from a game conference that there shouldn't be exceptions in your game
 
What?
 
11:06 AM
Well, during testing it's okay
 
True
 
and if you really have to halt the whole program, you need to catch it ideally
the user doesn't want their game to quit with an error
 
11:36 AM
Why wouldn't exceptions be used in games?
 
I'd say it depends on the exception. There's pretty little you can do with OutOfMemory besides trying to close every handle and display a nice message, for example. On the other hand, Java loves to throw a NullPointerException as essentially a warning that it doesn't "like" some input or similarly trivial excuses, and that you better deal with instead of crashing.
 
The idea is that things are bug-free
 
@Hjorthenify causes (caused in some cases I think?) extra runtime overhead even where not use
(Note: this is C++ we're talking about)
I've never heard of that advise in languages other than C++
 
Shameless self promotion, but the company I work for made something called the Seaboard. If you've ever wanted to play an expressive guitar solo with keys, watch this: youtu.be/iSko8vHOf8g
 
I mean, a console game isn't gonna output debug stuff to the end user. Mobile apps just crash. Desktop games, you can at least have a nice output log to send to the dev
 
11:43 AM
The runtime overhead is very small.. Bjarne Stroustrup has even said that exceptions add less of an overhead than checking for null all the time.
 
@Chris Even if my code is bug-free, I still can't avoid things like "This e-mail contains 8-bit characters despite being declared with encoding: 7bit, quoted-printable" or "Connection to server lost mid-stream."
 
Handling exceptions from dictionaries are fine though, right?
 
ah, how hard of an exception are we talking about it?
merely checking to see if something is this or that is okay imo
 
@Hjorthenify that is the cited reason I've heard off
And the overhead isn't for throwing an exception, it's for having exceptions enabled in the build
 
True
But instead of checking for null you would handle exceptions
 
11:45 AM
So no, handling exceptions from dictionaries isn't fine =p
If you're going by that advise
 
user92578
@OMGtechy Sounds really interesting and cool
 
user92578
But IMHO guitar solos should be played on a guitar
 
@Hjorthenify Oh, maybe I'm thinking of the terminology wrong. I think exception as something like a crash. Anything else is just a bool check, null check, maybe a try/catch if you really have to
 
Problem is that Dictionary[] doesn't return null, but throws an exception instead.
I'm talking about try / catch
 
My favourite type of exception: "Method doesn't exist", when you rely on a third-party plugin and that fetches some JavaScript code from the product creator's site, and that changes mid-version. Thanks a lot, Signiant.
 
11:47 AM
well, if it prevents your game from crashing, that's a good thing :p
you'd have to benchmark for performance concerns
if it's an issue
@MartinSojka relying on other interfaces that can change at whim flat out sucks
 
Yeah
 
@Hjorthenify wait, are you talking about C#?
 
Yes
 
If so: Then what are you talking about? You can't not pay the price of having exceptions
So use them
 
^
 
11:49 AM
That's why the advise is against them in games, most AAA games are C++ and you can disable exceptions, not just not use them, completely not have them and as such have no overhead from them
 
hah cool
@OMGtechy I have no sound atm ;\
 
Yeah I know, but I am still suprised that exceptions are frowned upon in C++ gamedev :/
 
@Tyyppi_77 It's great if you're writing something and want some MIDI data for it
 
user92578
Oh true
 
@Chris Check back later :D
 
11:57 AM
@Hjorthenify there's a performance cost, and I'm not 100% sure but I don't think they were supported in, for example, the compilers for the Nintendo DS
Performance cost even for code that doesn't use them
 
user92578
GunHero feels laggy :(
 
Yes I actually read that yesterday that they add to the code size even when not used, but I still don't see why that is a problem on PC since I also read it comes with performance benifits since they are only called when something goes wrong and hence you dont need to check for null all the time
@Tyyppi_77 Disable exceptions
 
@Hjorthenify could also be a case of inertia, exceptions were bad a while ago
 
www.toptal.com/c/after-all-these-years-the-world-is-still-powered-by-c-programmi‌​ng
Look for code size
 
user92578
12:03 PM
404
 
inb4
But I said they were bad, early implementations of exceptions weren't as good as current ones
 
wtf, it's 404 in one tab
 
user92578
GameTime seems to be the problem...
 
and then i googled for it and clicked the result and it loaded fine, url looks the same o-O
 
Yeah thatone
 
user92578
12:05 PM
I wonder why I didn't have any problems before...
 
@KevinvanderVelden Found the qoute from stroustrup:
but exceptions are expensive!: Not really. Modern C++ implementations reduce the overhead of using exceptions to a few percent (say, 3%) and that's compared to no error handling. Writing code with error-return codes and tests is not free either. As a rule of thumb, exception handling is extremely cheap when you don't throw an exception. It costs nothing on some implementations. All the cost is incurred when you throw an exception: that is, "normal code" is faster than code using error-return codes and tests. You incur cost only when you h
 
@Hjorthenify yes, inertia =p
Also I kinda wanna know what test they've done to get that percentage, it's probably not immediately applicable to games
 
Probably not
 
As a test, you could try rebuilding the UDK without exceptions enabled. In fact, I could, if I wasn't at work. :)
 
UDK?
Unreal Engine?
 
12:07 PM
Yes, since it's an AAA engine in C++ which has the source code available.
 
@MartinSojka It uses exceptions? Huh
 
@KevinvanderVelden I don't know, which is why I suggested how to test it.
 
How would we measure the performance though?
 
@MartinSojka aah, the way you said it (without exceptions enabled) suggested it has them enabled
@Hjorthenify make a benchmark with it, run it. Recompile with/without exceptions, run the same benchmark again and see the difference
 
I have literally no idea if they have them enabled or not.
 
12:10 PM
Aight
 
They are disabled by default it seems
... or not
Damn sources can't agree
 
Joe
12:58 PM
@Chris sorry dude i totally passed out haha, thanks for taking a look at the build man, and keep me posted on the plans
be happy to lend a hand on some porting if you need help
 
    CLASS(math){
    int add(int a, int b); //useage int z = add(10, 20);
    int sub(int a, int b); //^
    int mul(int a, int b);//^
    int divide(int a, int b);//^
    int squareroot(int a); //^
    int tan(double param); //^
    int pow(int a, int b);//^
    int roundUp(double a);
    int roundDown();//^
    int randomNumber(int HighestNumberWanted); //useage : int z = randomNumber(200); //random number up to 200
Just encountered this awesome C++ class on a programmer forum
Like.. Wtf?
 
Joe
lol
 
int roundDown(); ... No, really?
 
Probably(hopefully) a typo :P
 
I chose to believe that it's a shortcut for roundDown(randomNumber(0)).
 
1:05 PM
Yeah
 
Joe
either of you do any unity android dev? I have a timer that ticks down seconds since the last time you viewed an advertisement... to rate-limit ads 1/5mins
on editor the timer is flawless ticking every second... on an actual Android phone the timer updates only once every few minutes
 
Not android dev, sorry
@Jon and @Chris might be able to help though
 
Joe
no worries il bug them (@jon @chris) when they turn up with some CODE: github.com/josephjaniga/Mobile-RRS/blob/master/Assets/Scripts/…
 
@Joe you should see a doctor ... :P
 
Joe
haha
apparently 3 am is too late on a work night for this old man
 
1:12 PM
unacceptable
but uh, timer, I dunno
can you just check the time?
and not increment anything
 
Joe
I will for sure at lunch here
but for right now im just curious if anyones used System.DateTime on mono > android i noticed this on one the train
 
@Chris ... and now I have Robert Palmer stuck in my brain. Thanks, brain.
 
1:36 PM
@Hjorthenify "up to 200" so INTMIN to 200? or INTMIN to 199?
 
No clue
 
2:07 PM
Unity answers is basically the worst thing ever
 
Yes
Votes to close: Duplicate of Gamedev.se
 
Votes to close: Irreconcilable quality issues
Any time I feel like there's quality issues basically anywhere, I can just look at Unity answers. Sample a random few questions and have everything put in perspective next to their apparent misinformation rate of something like 3 bad answers per good answer.
 
It's still better than yahoo answers though
 
the poorman's stackexchange
 
@KevinvanderVelden That's very similar to saying saying, "He's a bad guy but at least he didn't skin you and your family alive."
 
2:12 PM
it is sad that UE4 uses the same AnswerHub crap
 
@MickLH I'd have gone with "But he's no hitler"
 
@KevinvanderVelden now now, no need to bring youtube comments section into this
 
Still though, would we want corporations sending people here?
 
would you want your customers to suffer through something like AnswerHub though?
 
As long as they are the scientists doing R&D yes please
A million people saying "Don't modify .velocity! Always use .AddForce!" means nothing to me. The use of the word "always" makes me confident that I have a better understanding of the situation.
But a single developer working on the physics core, could sway my opinion with one simple technical anecdote.
 
2:16 PM
yeah
 
The question I always have when something like this comes up is "So, why isn't velocity read-only?"
 
Thanks for the emotional support guys, I don't mean to complain so much but it's a tough internet out there sometimes.
 
@Chris there's been a few incidents with people demanding support with their bad questions on SO as I recall
 
heh
 
It's people not understanding that stack exchange isn't $COMPANY$ and getting pissy about it
 
2:21 PM
@MartinSojka AHHHHH not to mention, very intentional inclusion of the mode: ForceMode.VelocityChange dkrjghskjrh </rant>
 
I have no idea why the four different ForceMode types anyway. Physically, only one of them is force (ForceMode.Force). The others are acceleration (ForceMode.Acceleration), delta-impulse (ForceMode.Impulse) and delta-velocity (ForceMode.VelocityChange).
 
I'm almost sure it's an attempt at stupid proofing the numerical solver
But if that's easy for slow people to understand then I'm fucked because it completely screws up the issue for me, I always transform to remove .AddForce before thinking about the problem
Ah now to remove all instances of the word "fuck" "stupid" and "bullshit" from my answer and maybe even post it
 
2:37 PM
I dunno, sometimes I modify velocity directly. or use AddForce. or both. whatever to create my desired effect
 
I would have no real complaint if they just shared how it works under the hood.
 
yeah
 
Or at the very least, gave some code "identities" to explain what it even exactly does
I try to read people's Unity code, and they use Unity functions that are grey boxes that are half-ass defined in function from the developer and the unity answers community of users is a flaming pile of dogshit with some whipped cream
 
lol
 
3:06 PM
Can someone explain who would design a class like this?
Those names.. Srly?
 
user92578
that's how union stuff is often named I think
 
user92578
Just did the dishes, so now I can spend the whole evening programming
 
user92578
I just need something to program
 
GunHeroes?
 
user92578
GunHero, yes
 
user92578
3:10 PM
I need something interesting, otherwise I'll just end up playing games
 
I always end up playing games..
 
Joe
3:24 PM
I have on machine that I dont install games on... try to sit at that one when i know i need to do work
 
haha
 
Joe
doesnt really work... if you know theres another machine full of awesome in the same house
 
yeh, it's about self control
I'm always 2 clicks away from launching a game...... or launching Unity
 
user92578
lol yeah
 
Joe
i need an app that will lock my online game accounts durring certain hours
parental controls... for myself
 
user92578
3:26 PM
I opened VS tabs on my both monitors so that web isn't such a distraction
 
no fun allowed
 
user92578
Oh looks like I can finally use this chat in the VS browser!!!
 
user92578
AWESOME
 
lol..
 
user92578
Now I can always dev on two monitors
 
3:33 PM
Does Windows 10 have tabbed explorer?
 
user92578
No idea
 
sounds like nope
 
user92578
When is Windows 10 going to come BTW?
 
user92578
I'm really looking forward to it
 
29th
 
user4704
3:36 PM
29th.
 
user92578
Oh cool
 
user92578
The mobile version of the chat looks better in VS but it doesn't allow me to post with the enter key :(
 
I'm not gonna get win 10 :(
 
user92578
You couldn't get the thing to install?
 
Nope
 
user92578
3:39 PM
:(
 
user92578
Okay, enemies that are spawned with a trigger action now spawn with a parachute
 
Nice
 
user4704
5% of the time they sohuld spawn without one
 
user4704
and fall to their death
 
user4704
like the dwarves in Myth
 
user92578
3:50 PM
They probably should have their own parachute design tough
 
user92578
That sounds good Josh
 
user4704
or for example if they spawn over impassable terrain (which was actualy the trigger for spawning without a 'chute in Myth)
 
user4704
either way
 
user4704
it is hilarious
 
Myth was/is great
fricking dwarves
 
user92578
3:52 PM
Im doing the only right thing for the first time in my life and eating cheetos while programming
 
user4704
Yeah, I loved that game.
 
it's almost 4am, but it's time to make a gif of the dwarves blowing something up
doesn't do justice without sound though
 
user92578
lol
 
for the online Flag Rally game mode, I was ranked #1 for 8 months straight back in the day, those were good times
think I was ranked #7 overall once for a day, crescent moon or whatever
 
user92578
That's really cool
 
4:00 PM
the whole caste system was awesome
 
-3
A: How do I implement a retro-style password-based "savegame" system?

NPSF3000While solutions like Josh Petrie's character-to-state system have benefits (small save size, works on all copies of the game), they couple the games design and state system to be intrinsically to the password system. Any change to the password or state or design and the entire system falls apart...

We've got a classic case of locking onto a wrong idea here.
Requesting downvote storm.
 
Postmortem: Bungie's Myth: The Fallen Lords - 1998 - gamasutra.com/view/feature/131691/…
 
Joe
gotta love google drive app trying to cache all 15 gigs of my drive on my phone
no wonder im out of space
no ty I do not need vm imagines on my android device
 
user4704
@MickLH His approach is not invalid, even if he's somewhat confrontational in his approach to presenting it.
 
user92578
4:17 PM
I think instead of only changing the colors of the parachute I'll try to do a completely different parachute model for the enemy
 
@Tyyppi_77 Make the parachute grayscale and change colors at runtime
 
user92578
I could do that
 
user92578
But SDL's color shading isn't that great
 
How so?
 
user92578
I don't know, it shades quite poorly specific colors
 
4:23 PM
:/
 
user92578
4:34 PM
Too big?
 
user92578
 
user92578
Oh fuck I messed up the shading
 
needs a parachute strap jacket thing
 
user4704
yeah
 
user92578
Oh true
 
user4704
4:35 PM
You could probably get away with slightly smaller if you needed
 
user4704
but it looks reasonable
 
user92578
Okay
 
I'd say smaller.
 
user92578
The player parachute is smaller
 
JESUS CHRIST
oh nvm I thought I had pushed a bug to production
 
4:48 PM
everything I push to production is a bug
 
I am a bug.
 
Joe
@Chris how are you still awake
lol
 
im back-and-forthing with a customer still to get his controller working, my driving force i suppose
bout to head to bed
 
user92578
bout to wish you good night then :)
 
g night!
 
4:50 PM
night!
 
Joe
5:01 PM
nite dude good luck with the controller!
7/27 Stephen Hawking hosts his first Reddit AMA on AI! money.cnn.com/2015/07/27/technology/…
 
@JoshPetrie I disagree. While his approach is a valid method for creating a save game, it absolutely does not implement the "retro style" as found in "older console" that was very explicitly requested in the subject of the question.
This fellow's unrelated answer is no better than recommending a drill press when you need a hole punch.
 
user4704
He talks about putting a password system over the top of another form of saving state, rather than using the password to explicitly encode the state. It has the same outward appearance as the "authentic" approach and deals with some of the problems of that approach rather well.
 
user4704
We don't consider "how did X actually do Y" on-topic, so an answer that provides a solution that looks like Y is fine.
 
user4704
It's certainly no excuse for the anger directed at him by everybody in the commentary.
 
Again, I hold my stance that while it's a valid save-game system, it's doesn't actually look like Y because now you have an encrypted game save file you must lug around.
 
user4704
5:10 PM
If you think the answer isn't useful, vote down and move on.
 
It's absolutely not the same function, you can not tell your friend the password over the phone.
I did.
 
Looks like there's a pretty crazy bug in .NET 4.6: nickcraver.com/blog/2015/07/27/why-you-should-wait-on-dotnet-46
and installing .NET 4.6 introduces the bug to .NET 4.5 and (probably) 4.0: reddit.com/r/programming/comments/3err5p/…
 
user4704
Is this the Roslyn codegen bug?
 
user4704
er RyuJit
 
user4704
but yeah, looks like
 
5:40 PM
Can anyone tell me what is wrong with this piece of code?
public static int sortByCoord(CustomMapObject a, CustomMapObject b) {

if (a.y > b.y) {

return 1;
} else if (a.y == b.y) {
if (a.x > b.x) {
return 1;
} else {
return -1;
}
}
else if (a.y < b.y) {

return -1;
}
return 0;
}
it is a java comparator method
I use it to sort the map objects in my game
If i call it only once it works. but if i call it every frame they objects keep resorting and rendering wrong
btw these are all static objects so their x and y dont change
why does it sort differently though?
 
user4704
The sort you are using may not be a stable sort.
 
user4704
It would be unrelated to the logic of your comparison
 
this is how i sort
comparator = new Comparator<CustomMapObject>() {
@Override
public int compare(CustomMapObject o1, CustomMapObject o2) {
return sortByCoord(o1, o2);
}
};
then i just use java sort method
on my array of objects
 
You're also not ever returning 0 if they're equal
 
what do you mean?
 
user4704
5:48 PM
Which sort method?
 
user4704
Double check the docs for it to make sure it's stable.
 
    Sort.instance().sort(mapObjects, comparator);
it is libgdx...
but i think its the same in java
let me try a different method
when should i return 0?
 
user4704
when two things are equal
 
hmm let me try that
no... it still flickers if i return 0
even if i use the java method (Arrays.sort) then it still doesnt work
 
public static int sortByCoord(CustomMapObject a, CustomMapObject b) {
    if (a.y > b.y) {
        return 1;
    } else if (a.y == b.y) {
        if (a.x > b.x) {
            return 1;
        } else if (a.x == b.x) {
            return 0;
        } else {
            return -1;
        }
    } else {
        return -1;
    }
}
 
user4704
5:54 PM
problem is perhaps elsewhere then
 
Is the new sort something like this?
 
user4704
I think the Java Arrays.sort is is stable.
 
user4704
So, various things to try for diagnostics:
 
user4704
1) simply sort by coord to do something very straightforward
 
user4704
5:55 PM
verify resulting behavior
 
user4704
2) duplicate array, sort both, verify both are property sorted.
 
user4704
3) verify with memory breakpoints or whatnot that you don't change copies of the data before sorting, or simlar
 
ya i think the coordinates might be changing for some reason
thanks though
i will look for a problem like that
Ok so i am completely sure i am not accidentally changing the coordinates
and I made it so if i click it sorts it (so I can see what happens) and the first sort is correct.... but after that it just keeps sorting it wrong
 
LMFAO
I got tetris for the ps4 last week, so happy it isnt like that anymore :D
; - ; wish it had changable music tho like on mobile
 
Ok never mind it does not always sort it correctly the first time
 
6:08 PM
@TheMuffinCoder Fucking amazing.
 
@MickLH the fact that it was so un-optimized before? :P
 
"There's so many voxels in that piece... the PS4 can barely put it down there"
First laugh of the day
 
I fixed it YAY!
 
@MickLH xD
 
@TheMuffinCoder OMFG xD
 
6:13 PM
xD the comments were pretty funny too
 
How can you even make it that bad serious xD
I like how people are blaming the PS4.
 
@OMGtechy "Can you blame em? It's the best they could do with the ps4's weak hardware. I'm surprised it was even able to handle that color palette! You may be right though, this game should have just been cancelled for ps4 and put on true next gen systems only (aka 3ds and xbox one)"
 
Lol I didn't even read comments because I knew if I saw someone blaming the PS4 hardware seriously and not as a joke I would fucking pop off at them
 
They must be trolling xD
 
I'm ready to rant about "Expect more of this" and claim that this proves my theory about software inefficiency exploding faster than moore's law.
 
6:17 PM
@OMGtechy lol i found one guy that wasnt
"Graphics card is superior than the xbox One, but the CPU in the PS4 is inferior. PS4 lacks CPU cache and is the reason why Tetris has a hard time, however it's still Ubisofts fault for not optimizing it for the PS4, they didn't take the PS4s cache in to account.."
 
facedesk
 
I can't wait until you need a new computer every few weeks to achieve the same functions you've always had since DOS.
 
You think there will be a cross over at some point, where the hardware keeps getting better but the software is getting so much "lazier" that PCs just get slower instead?
3 mins ago, by MickLH
I'm ready to rant about "Expect more of this" and claim that this proves my theory about software inefficiency exploding faster than moore's law.
^ is that what you mean?
 
Wait does it seem to you guys that the 360 has better graphics than the wii u?
 
Jon
Greetings fellow nerds
 
6:29 PM
@Jon Greeting fellow Jon
 
Jon
What is up?
I added rotators to my game today, two types
I have come to the conclusion that the bulk of the work now, remains in the creation of levels --- it is quite slow going
LOL
 
@Jon make it so that you send map challenges to your friends which you create yourself that way you don't have to make maps ;)
 
Jon
yeah, ideally i'd have people submit levels
 
Yeah
 
Jon
That would require that I work on a portable level editor that is easy to use
I'm currently using Unity, lol
And then HOPE that people submit levels
imagine spending 3 weeks writing a awesome editor, and then get 4 submissions, 3 with penises and 1 with a nazi sign
Because I am pretty sure that is what would happen, lol
 
6:54 PM
@Jon LMAO
 

« first day (1817 days earlier)      last day (3219 days later) »