It doesn’t work like that. If an image processing algorithm takes 2 instructions per pixel, adding a check to every access could 3-4x the cost.
This is why if you dictate bounds checking then the language becomes uncompetitive for certain tasks.
The vast majority of cases it doesn’t matter at all - much less than 5%. I think safe/unsafe or general/performance scopes are a good way to handle this.
So very rarely should it be anything like 3-4x the cost, though some complex indexing could cause it to happen, I suppose. I agree scopes are a decent way to handle it!
Your understanding of how bounds checking works in modern languages and compilers is not up to date. You're not going to find a situation where bounds checking causes an algorithm to take 3-4X longer.
A lot of people are surprised when the bounds checking in Rust is basically negligible, maybe 5% at most. In many cases if you use iterators you might not see a hit at all.
Then again, if you have an image processing algorithm that is literally reading every single pixel one-by-one to perform a 2-instruction operation and calculating bounds check on every access in the year 2025, you're doing a lot of things very wrong.
> This is why if you dictate bounds checking then the language becomes uncompetitive for certain tasks.
Do you have any examples at all? Or is this just speculation?
One I am familiar with is Swift - which does exactly this because it’s a library feature of Array.
Which languages will always be able to determine through function calls, indirect addressing, etc whether it needs to bounds check or not?
And how will I know if it succeeded or whether something silently failed?
> if you have an image processing algorithm that is literally reading every single pixel one-by-one to perform a 2-instruction operation and calculating bounds check on every access in the year 2025, you're doing a lot of things very wrong
I agree. And note this is an example of a scenario you can encounter in other forms.
> Do you have any examples at all? Or is this just speculation?
Yes. Java and python are not competitive for graphics and audio processing.
Java and Python are not on the same body of water, let alone the same boat.
You can see some comparisons here:
https://benchmarksgame-team.pages.debian.net/benchmarksgame/...
Bounds checking barely registers as a factor.
And furthermore I don’t suspect you’re proposing we should be using C + bounds checking (that’s already a Gcc flag?). But rather bounds checking is one of many safety features.
The whole pitch of Java is exactly what OP said - let’s just pay 10-30% cost to get these nice portability and memory benefits over C++, and it didn’t work that way especially as memory speeds have diverged from CPU.
Because Python is an interpreted language and Java has garbage collection, not because they have bounds checking.
Using a language like Rust the bounds checking overhead compared to C code is usually minimal, like 0-5%. The compiler can even optimize away many bounds checks (safely) if you structure your code properly, such as with iterators.
Here's an example article : https://shnatsel.medium.com/how-to-avoid-bounds-checks-in-ru...
Make sure you read the note about how the author was mistaken about the 15% speedup from bounds checking. The actual bounds check change was much less, like a couple percent. You can even structure your code so that the bounds check disappears altogether when the compiler can confirm safety at compile time.
This is a theoretical argument. It depends on the compiler being able to see that’s what you’re doing and prove that there is no other mutation.
> abominations of C and C++
Sounds like you don’t understand the design choices that made this languages successful.
Even so you haven't provided any compelling evidence that C or C++ made it's decisions to be more appealing or more popular.