←back to thread

283 points walterbell | 1 comments | | HN request time: 0.315s | source
Show context
darkamaul ◴[] No.45769289[source]
Better (or simply more) ARM processors, no matter who makes them, are a win. They tend to be far more power-efficient, and with performance-per-watt improving each generation, pushing for wider ARM adoption is a practical step toward lowering overall energy consumption.
replies(5): >>45769421 #>>45769508 #>>45769815 #>>45769973 #>>45772372 #
ahoka ◴[] No.45769508[source]
Are ARM processors inherently power efficient? I doubt.

Performance per watt is increasing due to the lithography.

Also, Devon’s paradox.

replies(5): >>45769580 #>>45770046 #>>45770800 #>>45773990 #>>45779586 #
jorvi ◴[] No.45770046[source]
They aren't inherently power efficient because of technical reasons, but because of design culture reasons.

Traditionally x86 has been built powerful and power hungry and then designers scaled the chips down whereas it's the opposite for ARM.

For whatever reason, this also makes it possible to get much bigger YoY performance gains in ARM. The Apple M4 is a mature design[0] and yet a year later the M5 is CPU +15% GPU +30% memory bandwidth +28%.

The Snapdragon Elite X series is showing a similar trajectory.

So Jim Keller ended up being wrong that ISA doesn't matter. Its just that it's the people in the ISA that matter, not the silicon.

[0] its design traces all the way back to the A12 from 2018, and in some fundamental ways even to the A10 from 2016.

replies(5): >>45770152 #>>45770164 #>>45771322 #>>45771561 #>>45780873 #
IshKebab ◴[] No.45770152[source]
Do you have any actual evidence for that? Intel does care about power efficiency - they've been making mobile CPUs for decades. And I don't think they are lacking intelligent chip designers.

I would need some strong evidence to make me think it isn't the ISA that makes the difference.

replies(3): >>45770173 #>>45770460 #>>45774829 #
high_na_euv ◴[] No.45770173[source]
Isn't Lunar Lake first mobile chip with focus on energy eff? And it is reasonably efficient

We will see how big improvement is it's successor panther lake in January on 18A node

>I would need some strong evidence to make me think it isn't the ISA that makes the difference.

It is like saying that Java syntax is faster than C# syntax.

Everything is about the implementation: compiler, jit, runtime, stdlib, etc

If you spent decades of effort on peformance and ghz then don't be shocked that someone who spent decades on energy eff is better in that category

replies(3): >>45770819 #>>45770823 #>>45772691 #
mrsmrtss ◴[] No.45770823[source]
Actually, if you had made an opposite example, it might have gone against your point. ;) C# gives you a lot more control over memory and other low-level aspects, after all.
replies(2): >>45770980 #>>45771930 #
high_na_euv ◴[] No.45770980[source]
Yet how much perf in recent dot nets comes from that, and how much comes from "Span<T>"ning whole BCL?
replies(1): >>45771187 #
mrsmrtss ◴[] No.45771187[source]
There’s much more to it than just Span<T>. Take a look at the performance improvements in .NET 10: https://devblogs.microsoft.com/dotnet/performance-improvemen.... When it comes to syntax, even something like structs (value types) can be a decisive factor in certain scenarios. C# is fast and with some effort, it can be very fast! Check out the benchmarks here: https://benchmarksgame-team.pages.debian.net/benchmarksgame/...
replies(1): >>45771419 #
1. high_na_euv ◴[] No.45771419[source]
I know that C# is fast, this is my favourite lang, but it is hard to say honestly which one is faster

I love the saying "i dont trust benchmarks that i didn't fake myself"