41–50 of 149 posts
This is on the front page again. Please take some time to read this wonderful function: https://github.com/CRYTEK/CRYENGINE/blob/release/Code/CryEng... EDIT: That whole function is a minefield. Just taking a quick look: * 814 lines of code * goto inside 3 nested for-loops * macros * commented out code * new/delete, with no RAII * thread specific variables and locks (?)
Earlier quoted context omitted.
I disagree. Splitting a function up can help with readability and testability. The parent function becomes shorter and the child function can have a descriptive name. The parameters to a function are the fields that are needed for the function to perform its function.
One of my professors used to encourage the heavy use of helper functions that just... well, like in A or B in carmacks article, break your code into chunks with clear names that can be unit tested. How common is automated testing in AAA games?
Games that use automated tests often drive high level systems and test high level output. See for instance Riot Games's automated League of Legends test suite.
This is on the front page again. Please take some time to read this wonderful function: https://github.com/CRYTEK/CRYENGINE/blob/release/Code/CryEng... EDIT: That whole function is a minefield. Just taking a quick look: * 814 lines of code * goto inside 3 nested for-loops * macros * commented out code * new/delete, with no RAII * thread specific variables and locks (?)
This is on the front page again. Please take some time to read this wonderful function: https://github.com/CRYTEK/CRYENGINE/blob/release/Code/CryEng... EDIT: That whole function is a minefield. Just taking a quick look: * 814 lines of code * goto inside 3 nested for-loops * macros * commented out code * new/delete, with no RAII * thread specific variables and locks (?)
I came across the same kind of thing when I was kicking the tires on the Unreal Engine, and I wanted to attempt to add a double jump. I thought surely this should be an easy task, I would just need to find where the jump occurs, add a counter, and remove the restriction which only lets a character jump when touching the ground. What I found was a monstrous tangle of indirection similar to this one.
Now that's not to say that these engines are "bad code" - when you look at all the things a modern game engine does, including supporting interactive editing for non-coders, I'm sure there is some explanation for the level of complexity seen in code like this just because of how many systems must be layered on top of each-other. But that is the thing which makes me question whether general-purpose game engines are really a good idea at all.
In most other domains of software we've long ago eschewed this type of do-everything monolithic software design in favor of more loosely coupled composible toolsets. I'm not entirely sure why it seems that game development has yet to escape this paradigm.
This is on the front page again. Please take some time to read this wonderful function: https://github.com/CRYTEK/CRYENGINE/blob/release/Code/CryEng... EDIT: That whole function is a minefield. Just taking a quick look: * 814 lines of code * goto inside 3 nested for-loops * macros * commented out code * new/delete, with no RAII * thread specific variables and locks (?)
> * goto inside 3 nested for-loops.
C has no pattern for breaking out of multiple for loops at the same time. Other languages like Java and JavaScript introduced "break label;" to handle this edge case. goto is perfectly acceptable to break out of multiple loops.
> * new/delete, with no RAII
They use RAII, but it's not applicable here. In this case, the developers only want to allocate a temporary array when it needs to be resized. Since we don't want a large stack allocation (stack sizes are super small on consoles), using a delete/free to resize an array seems fine to me. Game developers have a long-seated distrust of std::vector, and for good reasons.
RAII is used for the WriteCondLock which you criticize in the next bullet point, so it's not like they were unaware of it. Just not the right tool for the job.
> * thread specific variables and locks (?)
You seem to be upset that they have code that uses locks at all? I don't really know what this bullet-point is saying, other than "I looked for 5 minutes and didn't understand the threading structure".
If I had to take an issue with this code, it's the lack of enums for e.g. iSimClass, despite it having an enum with definitions. That's the sort of stuff that's difficult to reason about and follow along with without having a mapping in my head. And it has no overhead, so why not do it?
https://github.com/CRYTEK/CRYENGINE/blob/6c4f4df4a7a092300d6...
This is on the front page again. Please take some time to read this wonderful function: https://github.com/CRYTEK/CRYENGINE/blob/release/Code/CryEng... EDIT: That whole function is a minefield. Just taking a quick look: * 814 lines of code * goto inside 3 nested for-loops * macros * commented out code * new/delete, with no RAII * thread specific variables and locks (?)
> new/delete, with no RAII There is nothing wrong with this. Plenty of people have no issues keeping track of memory in their head.
This is on the front page again. Please take some time to read this wonderful function: https://github.com/CRYTEK/CRYENGINE/blob/release/Code/CryEng... EDIT: That whole function is a minefield. Just taking a quick look: * 814 lines of code * goto inside 3 nested for-loops * macros * commented out code * new/delete, with no RAII * thread specific variables and locks (?)
As a hobbyist who likes to tinker with game and interactive media development, a sentiment I often come across is that in 2020 it makes no sense to implement a game engine, and that I should just use something which already exists to avoid re-inventing the wheel. Code like this is one thing which helps me to calmly ignore than sentiment. I came across the same kind of thing when I was kicking the tires on the Unreal…
Earlier quoted context omitted.
If it works, it's an example of ugly code. Plenty of game code is ugly as sin, though.
The code for VVVVVV infamously has switch/case for every possible screen/level in the game[0], which I think is hilarious. Best example of "it works!" [0] https://www.polygon.com/2020/1/13/21064100/vvvvvv-source-cod...
It's a bit of a strange thing; in my personal experience, after spending some real time with this type of constraints, it can be a bit painful to come back to "general software best practices". You become so aware of the performance implications of everything you do that all those things we do in the name of software quality can feel incredibly wasteful in terms of CPU and memory resources. One has to remind themselves that in 90% of cases that level of optimization is not warranted.
Earlier quoted context omitted.
Here's a good article from John Carmack on why that can be a good approach in game development: http://number-none.com/blow/john_carmack_on_inlined_code.htm... I feel like this Cryengine example may be a bad example of that, though.
Not just in game development. If you have a function that is only called once, it shouldn't be a function yet. Make it a function when you have a second or third use for it. Then, and only then, you will know what the parameters should be.
Earlier quoted context omitted.
It falls more in the category of having a lot bugs which could've been caught if they used static code analysis, code review, etc... I understand that you might think that messy could mean it's fine tuned for performance. In this case, I highly doubt it and think it's more reasonable to think it's messy because they had deadlines. The messy part isn't about performance optimizations. It's more about things that got c…
What sorts of static code analysis tools do people here use in their game projects? I know carmack is a big fan of them
https://www.viva64.com/en/b/0417/
https://www.viva64.com/en/b/0495/
https://www.viva64.com/en/b/0574/
I'm not endorsing pvs studio nor am I saying it's bad. Try out some tools and see what works best for you.