|
I call * the anus operator
|
# ? Dec 12, 2022 21:22 |
|
|
# ? Jun 6, 2024 05:41 |
|
namlosh posted:I call * the anus operator It's an assterisk
|
# ? Dec 12, 2022 21:46 |
|
namlosh posted:I call * the anus operator There is probably an inappropriate joke involving -> there...
|
# ? Dec 12, 2022 21:50 |
|
namlosh posted:I call * the anus operator You're going to get sued for trademark infringement by some proctologists
|
# ? Dec 12, 2022 22:07 |
|
https://twitter.com/AvidHalaby/status/1602127460677844993 Thread
|
# ? Dec 12, 2022 22:52 |
|
wtf they are going to get destroyed by gdpr if half of this is true
|
# ? Dec 12, 2022 23:19 |
|
This does help explain why basic-rear end stylesheet errors would randomly show up in Production
|
# ? Dec 12, 2022 23:47 |
|
OddObserver posted:There is probably an inappropriate joke involving -> there... C++ has a ->* operator…
|
# ? Dec 13, 2022 12:22 |
|
Zopotantor posted:C++ has a ->* operator… The operator is named DEREFERENCE THAT rear end
|
# ? Dec 13, 2022 18:25 |
|
Actually it's a pointer-to-member.
|
# ? Dec 13, 2022 23:14 |
|
Wouldn't it be a pointer from member?
|
# ? Dec 14, 2022 08:43 |
|
Volmarias posted:I cannot emphasize enough how damaging this was for my attempts at learning programming as a child. Oh I assure you I do that often. Especially stuff I wrote in my 20s back in the 90s and early 2000s. That poo poo was just amateurish. Hell sometimes I look at code I've just written and thought "wtf dude, your better than this..."
|
# ? Dec 16, 2022 01:17 |
|
I want github to add a feature where if you git blame a line and it shows you as the author, the price is right losing sound plays
|
# ? Dec 16, 2022 01:19 |
|
tfw you look at code you wrote yesterday and thought "gently caress it, that'll do"
|
# ? Dec 16, 2022 23:41 |
|
Hammerite posted:tfw you look at code you wrote and thought "gently caress it"
|
# ? Dec 16, 2022 23:56 |
|
Hammerite posted:tfw gently caress it
|
# ? Dec 17, 2022 00:15 |
|
when the linter warns on todo comments and you’re like shut the gently caress up why do you think it says todo
|
# ? Dec 17, 2022 03:04 |
|
dougdrums posted:when the linter warns on todo comments and you’re like shut the gently caress up why do you think it says todo code:
|
# ? Dec 17, 2022 03:29 |
|
Anyone ever looked at il2cpp output? I’ve been looking at some and it’s, um, weird. Stuff like null-checking EVERY time a variable is read, even if it’s already been read before in the same routine. Yea, I know, threads, but if you have to be that paranoid then your thread discipline is shot. And in any case the game has no way to recover from one of the pointers being null do all it does is to go haywire by skipping necessary code, instead of doing a controlled crash. Also it looks like the author copy and pasted code to count down every one of ~10 cooldowns manually..
|
# ? Dec 20, 2022 05:09 |
|
It's usually not worth independently doing a whole bunch of optimisations (like eliminating redundant checks) when the compiler that will later turn that into actual machine code has a much better optimizer that will do them for you. Focusing on a correct transformation that easily and obviously maps back to the original source is usually the way to go.
|
# ? Dec 20, 2022 05:55 |
|
Thing is, this is decompiled il2cpp. That should have already happened.
|
# ? Dec 20, 2022 06:15 |
|
In that case, I'm not sure you should be making a big deal out of "copy and pasted code" when it's most likely just an unrolled loop.
|
# ? Dec 20, 2022 06:28 |
|
il2cpp probably does a lot of optimizations in the C++ it emits in order to gently persuade the compiler to do optimizations such as unrolling loops if it makes sense. I wouldn't be surprised if Unity knows what compiler it's going to be ultimately using and can specifically target that. I've only used it with MSVC and clang, but those are both pretty knowable and have directives that can aid optimization.
|
# ? Dec 20, 2022 07:24 |
|
Since IL2CPP has to maintain the same behaviour as you had with the original C# it has to raise a NullReferenceException if you access a null value, as mentioned you absolutely can have a variable become null in-between calls if it's accessed from a different thread so the right behaviour is to null check before every access. Having it generate code that will lead to undefined behaviour in C++ by default would be much worse. If you're absolutely sure a value is not and can not become null you can disable null checks by using the code:
|
# ? Dec 20, 2022 09:12 |
|
I think it's having a lot of faith in Unity to assume it optimizes compiler specifically! I could understand throwing an exception somehow, but instead it seems that it just exits the function if a null is encountered, presumably because there was nothing to handle the NPE in the original program. Which while it avoids low-level undefined behaviour by accessing null memory, does provoke high-level undefined behaviour because no other part of the program is robust to the function having returned without completing its job. Another one I didn't realize wasn't being optimized: it calls the property getter for fixedDeltaTime every time it's referred to. I'm pretty sure that's a performance hit due to crossing the script/engine boundary.
|
# ? Dec 20, 2022 17:49 |
|
dougdrums posted:I want github to add a feature where if you git blame a line and it shows you as the author, the price is right losing sound plays Or bombards them with rubber missiles. https://github.com/codedance/Retaliation
|
# ? Dec 21, 2022 07:34 |
|
https://www.youtube.com/watch?v=7hdJQkn8rtA
|
# ? Dec 25, 2022 21:43 |
|
how timely, a clip of this was in the latest Technology Connections and I was curious to find the whole thing
|
# ? Dec 25, 2022 23:04 |
|
Dijkstracula posted:how timely, a clip of this was in the latest Technology Connections and I was curious to find the whole thing Hmm, yes, I wonder how that happened...
|
# ? Dec 25, 2022 23:34 |
|
EoRaptor posted:Hmm, yes, I wonder how that happened... *checks YT account...*
|
# ? Dec 25, 2022 23:38 |
|
I know it's the times but my biggest takeaway is plainly "Don't write in languages with really lovely syntax". Also they totally did change the program by changing the ≤ to <. If there's 133 orders total, 33 returns now skips the increment. Who knows what kind of other logic, like a strictly Greater Than 0.33 elsewhere, might be paired with or reliant on that ≤ and is now a corner case if not plainly incorrect. Also only now noticing they subtract returns from total orders, so 50 returns on 100 orders is ... 100% return rate. That's probably the real bug they were trying to fix before getting off in the weeds and breaking something else. This is the realest programming video ever! Ranzear fucked around with this message at 14:20 on Dec 27, 2022 |
# ? Dec 27, 2022 14:07 |
|
"Is that really a precise statement?" "Sure it is, just look at the program." "YOU look at it."
|
# ? Dec 27, 2022 20:41 |
|
Ranzear posted:I know it's the times but my biggest takeaway is plainly "Don't write in languages with really lovely syntax".
|
# ? Dec 27, 2022 22:53 |
|
SupSuper posted:You can't change the language, you can only format it. Get a load of this person who doesn't even use Perl
|
# ? Dec 28, 2022 00:00 |
|
Ranzear posted:I know it's the times but my biggest takeaway is plainly "Don't write in languages with really lovely syntax". They talk about how the .33 is an approximation of 1/3rd, so dropping the 'or equal too' isn't going to change the results in meaningful way.
|
# ? Dec 28, 2022 04:24 |
|
EoRaptor posted:They talk about how the .33 is an approximation of 1/3rd, so dropping the 'or equal too' isn't going to change the results in meaningful way. Later they changed the < to > and didn't explain the rationale for that very major change at all tho
|
# ? Dec 28, 2022 04:42 |
|
shame on an IGA posted:Later they changed the < to > and didn't explain the rationale for that very major change at all tho Are you sure you watched the video? Because they clearly spelled that decision out.
|
# ? Dec 28, 2022 04:49 |
|
shame on an IGA posted:Later they changed the < to > and didn't explain the rationale for that very major change at all tho They changed it from "if a < b, skip this next section" to "if a > b, do this section". That's not quite as "major" a change as you seem to be imagining. Consider also that when this video was made, Dijkstra's "GoTo Considered Harmful" letter (and the general concept of structured programming) was still pretty new and not necessarily widely practiced. The motivation for this change is literally to introduce structure to the code to make it easier to follow and understand.
|
# ? Dec 28, 2022 04:49 |
|
Jabor posted:They changed it from "if a < b, skip this next section" to "if a > b, do this section". ohhhhhhhh yeah I definitely missed that
|
# ? Dec 28, 2022 04:53 |
|
|
# ? Jun 6, 2024 05:41 |
|
Very pro click right here, it's great advice and it's way, way ahead of its time.
|
# ? Dec 28, 2022 05:57 |