hardware2win

Im always interested in comparison between top adult sites and netflix

What if it aint "dont like", but "not worth"? Like, if there is no value in wasting time on completion

>It's truly a shame how so many on HN are anti-innovation,

Inno what?

It is 2024, not 2006, what innovations are you talking about?

It isnt faang anymore - now we have mag7

The Advantage of early adopters

In the peak of hype they have years or decades of exp.

You think x86 would be changed in such a way that it'd break and?

Because what else?

If so, then i think that this is crazy because software is harder to change than hardware

I do wonder how crazy the deep dive they will have in future

Irc, cron, ldap, spying on other employees stuff

Yea, admins.

>Python is still the easiest way to write software

Try dotnet then

I really like OKRs when they are executed well

Indeed.

Standardized exams are the most fair and transparent.

Materials and tutors are available easily for free or cheaply on the internet nowadays.

Software engineeers dont work directly with hardware, it aint their expertise.

You definitely can be good SE without understanding semiconductors

Too much ORM hate in this thread

They are really useful for like 90% of the work

The rest can be made with raw sql

Combine strengths of those two powerful tools, dont be religious

Oh god, no, never.

Painful to maintain, test, debug, etc.

>That's because they've been allowed to call themselves engineers for so long that they're starting to believe they are.

What a take.

Is doctor telling bullshit about computers less doctor-ish?

This isnt some 50ppl startup

There are long ass supply chains

They will be manufacturing them for their customers.

Also arm isa is just isa

You seem to focus on it too much when it isnt THAT relevant.

Isa doesnt imply perf/energy characteristics

They will be manufacturing those ARM cpus

Have you read chip war?

It will challenge your concerns

So, teams??

Basically industry standard

Existing options like? Using 3rd party libs for basic data structures?

>C++ can omit all this, and can do simple bit shifts instead of multiplications.

Irrelevant. We need good lib in C

I do wonder why theres so little investing related content on HN

Arent "hackers" interested in this?

I think it is really unfair

Loot boxes are maybe 1% of the game and theres shitton of challenges, creativity and effort needed for the other 99%

Dont get baited folks

Those are companies hiring 100Ks of ppl, of course some of them are frustrated but it is not as relevant as news try to make it.

I bet you could read 10s of posts like that monthly about MSFT 15 years ago

What stock market taught me is that snarky comments under articles or reddit arent good future perf. predictor

Youre talking about java end to end, we are talking about just syntax.

>Sure, it's possible to write a Java allocator/GC that will yield similar performance to the C++ code,

Didnt you just agree with me that it is dependent on the impl/end product?

Because what would be the reasons in isa world to make it not desirable

>The syntax of the language has constrained the implementation so that Java will practically always be slower.

The most interesting question is: by how much?

1% 3%? 30?

>There are good reasons why 32-bit Arm failed to compete with Intel for performance, and why x86 has failed to displace Arm in low-power markets. The things that you want to optimize for at different sizes are different.

What if x86 CPUs werent designed for low power? e.g due to focus on competitivness in perf focused markets

Saying that one isa is faster or more energy efficient is like saying that c++ syntax is faster than java syntax.

While there are lang features that enable stuff, then almost everything is up to the implementation - compiler, libraries, runtime and the programs code.

One letters arent faster than the other. ISA doesnt imply perf. characteristics of the end product.

Read this:

https://chipsandcheese.com/2021/07/13/arm-or-x86-isa-doesnt-...

Even if you start talking about decoders, then:

>Another oft-repeated truism is that x86 has a significant ‘decode tax’ handicap. ARM uses fixed length instructions, while x86’s instructions vary in length. Because you have to determine the length of one instruction before knowing where the next begins, decoding x86 instructions in parallel is more difficult. This is a disadvantage for x86, yet it doesn’t really matter for high performance CPUs because in Jim Keller’s words:

Failure?

They work very fine, just not replace as many workers as expected

>pretty much everyone seems to hate the WEF

Who?

Random people on the facebook or reddit?

If so, then why their opinion should be taken into consideration?