Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

The world could run on older hardware if software optimization was a priority

twitter.com

Post reply on HN
Print view View original

791–800 of 842 posts

Re: The world could run on older hardware if software optimization was a priority

#791

Post by hellotomyrars »

CelestialMystic wrote: post #782
jen20 wrote: post #781

Earlier quoted context omitted.

Indeed, your own anecdata are as good as mine, and taken just as seriously.

Not to me it isn't. I think you are trying to justify the fact that you paid far too much for a vacuum cleaner, like most people do when they buy overpriced item and point out the obvious problems with their products. I own a Land Rover. It is old, expensive and unreliable. You know how I justify my spending on it? I like driving it.

So the person who says their Dyson works great is a liar but also their opinion is invalid because it is expensive.

Your Land Rover is good because it’s expensive but you like it.

Reading several of your comments on this thread are a real whirlwind. If you just flat out reject anyone’s experience that doesn’t reflect your own or that of your mother then I don’t know why you’re even responding to anyone.

I bought just about the cheapest toilet possible and it works identically to the one it replaced that was probably 15 years old. Maybe EU regulations are truly onerous and mad but the standards that have now been thrown in the garbage in the US have not been a problem for me literally ever. Anyone who needs to flush the toilet 10 times is doing something wrong.

I dunno what kind of cats your mom has but I’ve got 2 cats and 4 dogs and I haven’t had a problem with either a modest Shark or a (refurbished) Dyson.

Re: The world could run on older hardware if software optimization was a priority

#792

Post by hellotomyrars »

spinarrets wrote: post #582
nine_k wrote: post #516

Earlier quoted context omitted.

My wife ran a restaurant that was relatively successful due to the quality of its food and service. She was able to establish it as an upper-tier experience, by both some word of mouth, but also by catering to right events, taking part in shows, and otherwise influencing the influencers of the town, without any massive ad campaigns. As a result, there were many praises in the restaurant's visitor book, left by people…

If it paid for people's lives and sustained itself, that sounds like a huge success to me. There's a part of me that thinks, maybe we'd all be better off if we set the bar for success of a business at "sustains the lives of the people who work there and itself is sustainable."

If only. So much of the constant churn in big corporations is killing units that are profitable but just aren’t profitable enough.

Re: The world could run on older hardware if software optimization was a priority

#793

Post by CelestialMystic »

hellotomyrars wrote: post #791
CelestialMystic wrote: post #782

Earlier quoted context omitted.

Not to me it isn't. I think you are trying to justify the fact that you paid far too much for a vacuum cleaner, like most people do when they buy overpriced item and point out the obvious problems with their products. I own a Land Rover. It is old, expensive and unreliable. You know how I justify my spending on it? I like driving it.

So the person who says their Dyson works great is a liar but also their opinion is invalid because it is expensive. Your Land Rover is good because it’s expensive but you like it. Reading several of your comments on this thread are a real whirlwind. If you just flat out reject anyone’s experience that doesn’t reflect your own or that of your mother then I don’t know why you’re even responding to anyone. I bought just…

> So the person who says their Dyson works great is a liar but also their opinion is invalid because it is expensive. Your Land Rover is good because it’s expensive but you like it.

I am not trying to justify my purchase by pretending it is not bourgeois choice, that was the point I was making.

Dyson's have historically been more expensive than other brands (at least in the UK) and they aren't actually worth the extra money. I just looked on amazon for prices "air purifier fans" and it is £500, I have something similar for my living room and I bought was £50.

> Reading several of your comments on this thread are a real whirlwind. If you just flat out reject anyone’s experience that doesn’t reflect your own or that of your mother then I don’t know why you’re even responding to anyone.

My experiences was flat out rejected to begin with. I told there isn't a problem, even though I know there is because I have some of the older products and I know they work better.

Other people have told me personally that they have made similar observations. So I know it isn't just I.

Re: The world could run on older hardware if software optimization was a priority

#794

Post by TFYS »

chii wrote: post #787
TFYS wrote: post #751

Earlier quoted context omitted.

Cheating happens in competition based systems. No one cheats in games where the point is to co-operate to achieve some common goal. We should aim to have a system based on recognizing those common goals and enabling large scale co-operation to achieve them.

> co-operate to achieve some common goal. all systems are competitive, if the system involves humans - after all, even in a constrained environment like academia, where research is cooperative, the competition for recognition is still strong. This includes the order of the authorship presented in the paper. What you're asking for, regarding cooperation to achieve common goals, is altruism. This does not exist in huma…

Academia is competitive because it's designed to be competitive. If things like funding, recognition and opportunities go to "winners", people will try to win. It's possible to design systems that do not force people to compete. For example you could take away the names from papers and assign funding randomly/semi-randomly and the competition would end. Then add some form of retroactive funding (or other kinds of rewards) that's awarded to research that has produced useful results, and you'll get your incentive to do good research without the need for competition.

It's harder to design systems that avoid competitive behavior, but I don't think it's impossible. And of course competition is not all bad, it's a good tool when used carefully. But it's way too much when most of our systems are based on it.

Re: The world could run on older hardware if software optimization was a priority

#795

Post by treszkai »

caseyy wrote: post #91

There is an argument to be made that the market buys bug-filled, inefficient software about as well as it buys pristine software. And one of them is the cheapest software you could make. It's similar to the "Market for Lemons" story. In short, the market sells as if all goods were high-quality but underhandedly reduces the quality to reduce marginal costs. The buyer cannot differentiate between high and low-quality g…

> Smart Laundry with LG's AI Washing Machines: Efficient Spin Cycles & Beyond

Finally, the perfect example of AI-washing.

Re: The world could run on older hardware if software optimization was a priority

#796

Post by xondono »

cogman10 wrote: post #495
xondono wrote: post #355

Earlier quoted context omitted.

> Interpreted languages are now within a single digit multiple of natively compiled languages. You have to be either clueless or delusional if you really believe that.

Let me specify that what I'm calling interpreted (and I'm sure carmack agrees) is languages with a VM and JIT. The JVM and Javascript both fall into this category. The proof is in the pudding. [1] The JS version that ran in 8.54 seconds [2] did not use any sort of fancy escape hatches to get there. It's effectively the naive solution. But if you look at the winning C version, you'll note that it went all out pulling…

And here we get into the absurdity of microbenchmarks like these.

Yes, you can get the JVM to crunch some numbers relatively fast, particularly if the operations are repetitive enough that you can pull JIT tricks.

Now try to run something that looks a bit more like an actual application and less like a cherry picked example, and as soon as you start moving memory around the gap jumps to orders of magnitude.

Re: The world could run on older hardware if software optimization was a priority

#797

Post by rjbwork »

naasking wrote: post #538
rjbwork wrote: post #469

Earlier quoted context omitted.

Well, you can do that with LINQ + EF and embedded databases like SQL Lite or similar.

LINQ is on the right track but doesn't quite go far enough with query composition. For instance, you can't "unquote" a query within another query (although I believe there is a library that tries to add this). EF code-first is also on the right track, but the fluent and attribute mapping are awkward, foreign key associations often have to be unpacked directly as value type keys, there's no smooth transition between i…

Can you elaborate on what you mean by "unquote"? I've not heard this term and I can't find any relevant info from search.

Re: The world could run on older hardware if software optimization was a priority

#798

Post by KronisLV »

spinarrets wrote: post #582
nine_k wrote: post #516

Earlier quoted context omitted.

My wife ran a restaurant that was relatively successful due to the quality of its food and service. She was able to establish it as an upper-tier experience, by both some word of mouth, but also by catering to right events, taking part in shows, and otherwise influencing the influencers of the town, without any massive ad campaigns. As a result, there were many praises in the restaurant's visitor book, left by people…

If it paid for people's lives and sustained itself, that sounds like a huge success to me. There's a part of me that thinks, maybe we'd all be better off if we set the bar for success of a business at "sustains the lives of the people who work there and itself is sustainable."

> There's a part of me that thinks, maybe we'd all be better off if we set the bar for success of a business at "sustains the lives of the people who work there and itself is sustainable."

This would be beautiful in a world where retirement was better and it didn’t feel like inflation or financial crashes are looming around the next corner most of the time.

For many folks, trying to get savings and putting money into investments is less about wanting a lavish lifestyle later and more about just wanting financial security in case something bad happens.

Re: The world could run on older hardware if software optimization was a priority

#799

Post by cogman10 »

xondono wrote: post #796
cogman10 wrote: post #495

Earlier quoted context omitted.

Let me specify that what I'm calling interpreted (and I'm sure carmack agrees) is languages with a VM and JIT. The JVM and Javascript both fall into this category. The proof is in the pudding. [1] The JS version that ran in 8.54 seconds [2] did not use any sort of fancy escape hatches to get there. It's effectively the naive solution. But if you look at the winning C version, you'll note that it went all out pulling…

And here we get into the absurdity of microbenchmarks like these. Yes, you can get the JVM to crunch some numbers relatively fast, particularly if the operations are repetitive enough that you can pull JIT tricks. Now try to run something that looks a bit more like an actual application and less like a cherry picked example, and as soon as you start moving memory around the gap jumps to orders of magnitude.

> and as soon as you start moving memory around the gap jumps to orders of magnitude.

Depends on what you mean by "moving memory". In terms of heap allocation performance, the JVM and I suspect the JS engine will end up trouncing C/C++. Why? Because a GC allocator will outperform manual memory management in terms of absolute allocation rate. Because memory for a C++ allocator is pinned, unless you do a bunch of work utilizing things like bump allocators you'll simply flounder in terms of what the JVM does by default.

All(?) of the JVM GCs are moving collectors. That means new allocation ends being a simple pointer bump with a bounds check. Really hard to beat in terms of perf.

But it doesn't stop there. If we are talking real applications, then one real thing the JVM does better than C++ is concurrent programming. A GC works far better for managing concurrent data. So much so that when you look at high performance multithreaded C++ code you'll often find a garbage collector implementation. Something that Java gets by default. If you aren't google writing chrome, you are either aggressively copying memory or you are using something like atomic reference counters. Both of which will absolutely nuke performance.

But that's not all. One other thing the JVM does better than C++ can generally hope to do is devirtualization. In real applications, you are likely either in template hell for C++ to get performance or you are dealing with some form of interfaces and inheritance which ultimately creates code that the compiler will struggle to optimize without going to an extreme like PGO. The JVM gets PGO for free because it doesn't have to blindly optimize methods.

These simple microbenchmarks are a gift to C/C++, not the JVM.

Re: The world could run on older hardware if software optimization was a priority

#800

Post by igouy »

cogman10 wrote: post #773
igouy wrote: post #554

Earlier quoted context omitted.

fwiw There are a few naive un-optimised single-thread #8 n-body programs transliterated line-by-line literal style into different programming languages from the same original. [1] > a single digit multiple By which you mean So not those Java -Xint, PHP, Ruby, Python 3 programs? > interpreted Roberto Ierusalimschy said "the distinguishing feature of interpreted languages is not that they are not compiled, but that any…

> By which you mean Yup > So not those Java -Xint, PHP, Ruby, Python 3 programs? Pretty much. The Java -Xint stuff turns off the JIT and purely uses the interpreter. That's not how anyone (that I'm aware of) uses java. I had to look up exactly what that flag did. Ruby I think might be underperforming because `yjit` isn't enabled by default. I suspect the same of Python (it just got a jit with python 3.13. Looks like…

`yjit` was enabled explicitly.

https://benchmarksgame-team.pages.debian.net/benchmarksgame/...

> Let me specify that what I'm calling interpreted…

Is everything except native aot compilers.

Post reply on HN