This is indeed really impressive!

(direct link: https://boricj.net/atari-jaguar-sdk/2024/01/02/part-5.html )

Wouldn't this defeat the purpose of having an immediate-mode API ?

- "Null pointer dereference"

- "Out of memory"

- "Disk is full"

- "File does not exist"

- "File does not exist in cache"

- "File exists but is corrupt"

- "Access denied"

- "Connection reset by peer"

It's pretty obvious that all of the above is generally unwanted most of the time.

However, putting them all in the same bag labeled "error", and forcing them to be treated the same way might be counterproductive. Sometimes you might want to panic. Sometimes you might want to retry. Sometimes you might want to ignore!

Now, if your program isn't interactive (such as a compiler), halting on any error might be a choice. But you still have to provide contextualized and accurate error messages, which is easy for the case "File does not exist", and a lot less easy for the case "Out of range index".

What about Prolog ? How is it not a counter-example of this ?

The self-correcting mechanisms of science can only correct knowledge. Those mechanism work mainly by requiring the research works to be checkable by others. Self-correctness emerges by the accumulation of checks on the same topic, all leading to the same conclusion, and by the progressive retractation of bad research ... not by the elimination of "bad researchers".

Efficiently "correcting" people, whatever that means, is a different beast. Such a mechanism belongs to an administrative entity who can emit decisions - and, by construction, who can make errors.

Which makes sense, at least firstly, considering that understanding a proof is a lot easier when its conclusion is already familiar.

Of course, every physics engine behaves a little differently from the others, so, if you game is physics centered (pinball, racing), switching implementation might result in a slightly different game - but having the core of your game depends on the implementation details of some 3rd-party library might not be a good idea anyway.

In general, the more you sprinkle your code with dependencies to 3rd party libraries, the less control you have over the resulting product.

Your point still stands, though, for some very low-level "utility" libraries like boost or the STL, or any standard library replacement , and more generally, libraries holding "vocabulary" types.

In the good old times most video games weren't like that.

It's a pity that those times are gone, to the point video games are now used as such a comparison point!

[...]

A final precision: Operations that are designed to block do not disqualify the algorithm. For example, a queue’s pop operation may intentionally block when the queue is empty. The remaining codepaths can still be considered lock-free. ```

I don't get it ; how could "locking a mutex" not be considered as an "operation that are designed to block" ?

Would somebody be kind enough to tell me what am I missing here? Thanks !

I've been running a 3.5GHz CPU (with 4 HT cores) for nearly 10 years.

That's a lot of weeks. Surely, today, I should be able to buy a 10GHz CPU, right? :D

And yet, GNU/Linux is supported by many commercial games: Amnesia, SOMA, SpaceChems, Towerfall, Braid, HyperLight Drifter, Shovel Knight, Undertale, VVVVVV, Hollow Knight, Overload, WarGame Red Dragon ...

So there must be something else there than extra revenue.

The game uses Unity3D, it might only be a matter of ... recompilation :D

For the same reasons you might engage in any entertainment activity.

Real-life still-unsolved puzzles are hard ; whereas puzzle games keep the difficulty under some control (except for the infamous boat puzzle in The Witness).

Real-life problems are hard because, often, proper domain modeling/formalization is part of the problem. Puzzle games generally don't have this 'domain modeling' part, they provide you with an extremely consistent set of rules, and you have to build a solution inside it. Sometimes this set of rules is partly implicit, and discovering it is part of the fun (e.g The Witness, Antichamber) ; but you never have to worry about the real-world and it's complexity.

There are some pretty good public game codebases (Penumbra/HPL, Quake I/II/III, Doom I/III), there are some okay ones (RedAlert), there are also pretty awful ones (Duke3D, Descent).

Note that this seems to be completely independent from whether the game is good or not!

You could have both a deviation (i.e tangential acceleration) and a constant speed.