pcwalton

32 ms·

451.
by pcwalton 4y ago
As pointed out by Matias Goldberg [1], neither model is particularly optimized, however. They produce too many sliver triangles. The max-area approach [2] was brought up as an alternative that can significantly improve shading performance.
452.
by pcwalton 4y ago
BCD was popular for a long time for values that had to be constantly printed in decimal, since doing repeated division by 10 could be slow on systems that had no wide multiplication and no barrel shifter (remember that integer division by a
453.
by pcwalton 4y ago
> I’ve done this. I had lots of fun—everyone, from the CEO down, was crunching—and became financially independent after it. OK, but this doesn't seem relevant to the situations Twitter's (former, in many cases) employees are in
454.
by pcwalton 4y ago
GDScript is a bad idea as a teaching language because it's so niche, practically a DSL. You really want people's first language to have broad applicability. Yes, Python has deployment problems, but so does Godot.
455.
by pcwalton 4y ago
> I don't like Python. But it is difficult for me to look past the empirical evidence that it is (a) remarkably easy for new programmers to pick up, and (b) an extraordinarily popoular, load-bearing tool for actual practitioners in
456.
by pcwalton 4y ago
I really want -> in Rust too (or perhaps .*). I don't see why such a thing would be controversial.
457.
by pcwalton 4y ago
Google absolutely uses lots of C++ in high-volume web backends today. Go didn't replace it.
458.
by pcwalton 4y ago
> A big problem with Rust, long-term, is that the kind of programs that really need it are somewhat out of today's mainstream. It's not that useful for webcrap. It's not that useful for phone apps. The AI people use Jupyte
459.
by pcwalton 4y ago
> Compare that to now, where Covid can destroy your life and you can get it just going to the grocery store...it's much worse. The chance of dying from COVID is 4/100,000 if you're 80 years old and fully vaccinated [1].
460.
by pcwalton 4y ago
Every time someone says the 90s were better than today I remind them that, back then, HIV was a death sentence. You got it, you were dead--that simple. The fact that we made it into a manageable chronic condition is an astonishing triumph o
461.
by pcwalton 4y ago
Then flag or ignore them? For me, the interesting story here isn't so much who got what record in a 42 year old arcade game but rather how frighteningly easy it remains to intimidate people over bogus nonsense by abusing the legal syst
462.
by pcwalton 4y ago
I suspect that the most popular software using DAA that folks here have used would be emulators of old systems that had BCD mode, such as the SNES. For example, ZSNES uses it: https://github.com/xyproto/zsnes/blob&
463.
by pcwalton 4y ago
> Most C++ apps are mostly memory safe and that's just fine. It's ridiculous to suggest that this secondary artifact concern is raised to primacy for all projects. It's demonstrably not "just fine", as the steady
464.
by pcwalton 4y ago
Reference counting is a form of garbage collection, and therefore Swift is a GC'd language.
465.
by pcwalton 4y ago
> Most projects do not require what Rust provides at the cost it provides it at. If this is true, it's because most projects should use a GC'd, memory safe language, not because most projects should use memory unsafe languages.
466.
by pcwalton 4y ago
> A project’s culture is about 90% the culture of its initial group. Eh, not in the case of Rust. Rust doesn't even use the MPL. The biggest Mozilla-ism I can think of that remains in the Rust project is, like, the use of "r+&
467.
by pcwalton 4y ago
> I don't know a ton about low level performance, but I'd find it surprising if it's not true. How can interpreted bytecode outperform direct machine instructions? Because the JVM and .NET compile; they don't interpre
468.
by pcwalton 4y ago
Async/await in Rust is mostly just sugar on top of futures, which you can implement manually. There's plenty of old code in Rust that predates async/await that does just that.
469.
by pcwalton 4y ago
> Another subtle difference is that most functions take a context in Go. AFAIK, most functions in Rust are synchronous functions. I'm fairly certain that most functions in Go don't take a context. There are tons of helper funct
470.
by pcwalton 4y ago
Structural vs. nominal typing has nothing to do with monomorphization. Depending on what you consider "structural typing" and how much you care about separate compilation, it may prevent you from doing monomorphization in some cas
471.
by pcwalton 4y ago
> I'd prefer not to pay 5000 or 6000 total - $3000 to a hospital and $3000 to an ER doctor agency for a small 2nd degree burn or small cut requiring stitches, or any other issue I may have been taken there in the past for. In the US
472.
by pcwalton 4y ago
People come and go from every project. I haven't noticed any particular drop in momentum for the Rust compiler. (Also, the Rust compiler dev community is hardly "secretive and opaque"--just look at the amount of drama that re
473.
by pcwalton 4y ago
It's also bad for security. IIRC code execution is easier on MIPS because 0x0 is a NOP.
474.
by pcwalton 4y ago
Rust does the same thing as C++ here. Note that rustc is pretty aggressive about using the LLVM lifetime intrinsics to allow for stack coloring (i.e. reuse of stack slots as necessary).
475.
by pcwalton 4y ago
Most of the obscure architectures people talk about are rather slow—after all, if performance were desired, then there would be a modern compiler available for them—and so code running on them isn't performance critical.
476.
by pcwalton 4y ago
Adding more IRs isn't necessarily a problem. Clang goes through AST -> LLVM IR -> SelectionDAG -> MachineInstr -> assembly already.
477.
by pcwalton 4y ago
The title undersells it: this means that Rust can now compile to C89, by first compiling to WebAssembly and then compiling that WebAssembly to C. This addresses one of the major concerns about Rust (and LLVM, for that matter) that I see bro
478.
by pcwalton 4y ago
I saw someone point out that these identifier restrictions mean that a valid C compiler optimization would be to change all instances of "x = to[a-z]+(y);" to "if (!is[a-z]+(y)) x = to[a-z]+(y);" Not that any compiler wo
479.
by pcwalton 4y ago
> For one, it's the GPU that actually does the initial boot process, and much of that is hard to find good info on. ( https://raspberrypi.stackexchange.com/questions/14862/why-do ...) x86 is similar, since I
480.
by pcwalton 4y ago
A switch can jump inside an if body that contains case labels, causing the condition not to be evaluated. C switch/case semantics are... not the most obvious.

More ›